Click here for an overview of the Compass Users Group Archives:
Date: Sun, 2 Jan 2011 13:33:13 -0800 (PST) From: Tom Subject: RE: [compass-users] Version control: multiple editors Version control: multiple editors I want to be able to split up the project so multiple people can edit the existing files. Didn't really find anything in the manual. Is it as simple as having parts of the cave in separate dat files? Then just deleting the old dat file and adding the new dat file from windows and outside of the .MAK? The lineplot is already entered, but not passage dimensions, mag dec, and other minor details. thanks, Tom Bigfoot Cave, 16 miles kmctf.org Version control: multiple editorsI want to be able to split up the project so multiple people can edit the existing files. Didn't really find anything in the manual. Is it as simple as having parts of the cave in separate dat files? Then just deleting the old dat file and adding the new dat file from windows and outside of the .MAK? The lineplot is already entered, but not passage dimensions, mag dec, and other minor details.thanks,TomBigfoot Cave, 16 mileskmctf.org
Date: Tue, 4 Jan 2011 04:42:04 -0700 From: "Larry Fish" Subject: RE: [compass-users] Version control: multiple editors Hi Tom, I want to be able to split up the project so multiple people can edit the existing files. Didn't really find anything in the manual. Is it as simple as having parts of the cave in separate dat files? Then just deleting the old dat file and adding the new dat file from windows and outside of the .MAK? The lineplot is already entered, but not passage dimensions, mag dec, and other minor details. If the survey files are already divided up so that only one person is working on one file at a time, then you can do exactly what you described. However, I wouldn't arbitrarily split the files up just so each person is working on a different file. If you are going have more than one person working on the same file at the same time, here is what I'd do: 1. Give each person a copy of all the data, even if he/she is not working on some sections. Part of the reason for giving everyone all they data is they may need it to find and fix certain error problems such as mis-tied stations. Without the whole cave data set, it may be difficult to find the problem. 2. Make sure each person keeps track of which sections, shots and surveys they have changed. 3. When they are done editing, put each edited set of files in a separate disk folder. 4. Using the information about what person has changed what surveys; merge the data back into a "Master" copy of the data. 5. To do the Merging Operation, use the "Manipulate Surveys" option in the Compass Project Manager. You will find it by selecting the "Edit - Manipulate Surveys" option from the menu bar. This tool will allow you copy, merge, delete and re-order surveys between two files. The Left Hand panel shows the surveys in the "Merge Target." The Right Hand Panel shows the surveys from the "Source File." You can copy surveys from the Source to the Target. You can delete and reorder surveys in the Target. When you save the Target, all the changes you have made will be incorporated into the Saved file. There is detailed information on the Operation of the Manipulate Surveys tool in the Help files. While you are doing this operation, make sure you make lots of backup copies at every step of the way. That way if you make a mistake, you can just revert to one of your backups. Larry
Date: Tue, 4 Jan 2011 11:04:27 -0800 (PST) From: Tom Subject: RE: [compass-users] Version control: multiple editors Larry, As you saw the Bigfoot, there are about 5(?) dat files in the MAK. So I will have them only edit that particular dat file. That means I should be able to merge their edited dat file into my master survey. I think it would be asking too much to have them keep track of each survey modified. thanks, Tom From: Larry Fish Subject: RE: [compass-users] Version control: multiple editors To: [email protected] Date: Tuesday, January 4, 2011, 3:42 AM Hi Tom, I want to be able to split up the project so multiple people can edit the existing files. Didn't really find anything in the manual. Is it as simple as having parts of the cave in separate dat files? Then just deleting the old dat file and adding the new dat file from windows and outside of the .MAK? The lineplot is already entered, but not passage dimensions, mag dec, and other minor details. If the survey files are already divided up so that only one person is working on one file at a time, then you can do exactly what you described. However, I wouldn't arbitrarily split the files up just so each person is working on a different file. If you are going have more than one person working on the same file at the same time, here is what I'd do: 1. Give each person a copy of all the data, even if he/she is not working on some sections. Part of the reason for giving everyone all they data is they may need it to find and fix certain error problems such as mis-tied stations. Without the whole cave data set, it may be difficult to find the problem. 2. Make sure each person keeps track of which sections, shots and surveys they have changed. 3. When they are done editing, put each edited set of files in a separate disk folder. 4. Using the information about what person has changed what surveys; merge the data back into a "Master" copy of the data. 5. To do the Merging Operation, use the "Manipulate Surveys" option in the Compass Project Manager. You will find it by selecting the "Edit - Manipulate Surveys" option from the menu bar. This tool will allow you copy, merge, delete and re-order surveys between two files. The Left Hand panel shows the surveys in the "Merge Target." The Right Hand Panel shows the surveys from the "Source File." You can copy surveys from the Source to the Target. You can delete and reorder surveys in the Target. When you save the Target, all the changes you have made will be incorporated into the Saved file. There is detailed information on the Operation of the Manipulate Surveys tool in the Help files. While you are doing this operation, make sure you make lots of backup copies at every step of the way. That way if you make a mistake, you can just revert to one of your backups. Larry ------------------------------------ Yahoo! Groups Links � � [email protected]
Date: Sat, 05 Feb 2011 05:05:58 -0000 From: "derekbristol" Subject: Round Tripping Issues I'm working on a cave map with only 4 loops. I thought that when I originally compiled the data I had set Compass to close those loops, but apparently did not. It placed all of the loop error on the closing shot. I've addressed this, but had already fully drawn the cave in Illustrator CS. So I figured this would be a good opportunity to learn round tripping and fix the small deviation between the drawing and new line plot with properly closed loops. I open the new .plt file in SVGexporter, but when I go to "Load Target", and select my Illustrator file (.svg format), I get an error: "Reference to undeclared namespace prefix: 'inkscape'.". I've tried a number of things, but it seems that whenever I take any SVG export from Compass, and add anything at all to it in Illustrator, I get this same error message when I try to Merge / Morph. Anyone else have this problem or know how to solve it? I'm curious about the Inkscape reference in the error message. I'm not using Inkscape and haven't converted to Inkscape .svg format. Thanks, Derek
Date: Sat, 5 Feb 2011 03:48:49 -0700
From: "Larry Fish"
Subject: RE: [compass-users] Round Tripping Issues
Hi Derek,
It sounds like there is missing a field that the SVG-Exporter is expecting
to find. Compass puts several "Inkscape" tags in the data to insure that the
data is Inkscape-compatible. In some instances, Compass has to add that data
if it doesn't find it. For example, if the data came straight from
Illustrator, it would not have Inkscape tags.
You have probably put your data through a slightly different series of
operations than anything I tried here and some or all of the tags are
missing. If you could send me a copy of your data, I could find the problem
and fix it.
I will send you a private email with my address so you can send me the data
if you are willing to.
Send me both the plot file and the SVG file. I will keep data confidential
and it won't be shown or given to anyone else.
Larry Fish
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of derekbristol
Sent: Friday, February 04, 2011 10:06 PM
Subject: [compass-users] Round Tripping Issues
I'm working on a cave map with only 4 loops. I thought that when I
originally compiled the data I had set Compass to close those loops, but
apparently did not. It placed all of the loop error on the closing shot.
I've addressed this, but had already fully drawn the cave in Illustrator CS.
So I figured this would be a good opportunity to learn round tripping and
fix the small deviation between the drawing and new line plot with properly
closed loops.
I open the new .plt file in SVGexporter, but when I go to "Load Target", and
select my Illustrator file (.svg format), I get an error: "Reference to
undeclared namespace prefix: 'inkscape'.". I've tried a number of things,
but it seems that whenever I take any SVG export from Compass, and add
anything at all to it in Illustrator, I get this same error message when I
try to Merge / Morph.
Anyone else have this problem or know how to solve it?
I'm curious about the Inkscape reference in the error message. I'm not using
Inkscape and haven't converted to Inkscape .svg format.
Thanks,
Derek
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Sat, 05 Feb 2011 15:30:51 -0000 From: "derekbristol" Subject: Re: Round Tripping Issues Larry, Data sent. I think you'll recognize the cave... it's Premonition. Thanks for the help. Hi Derek, It sounds like there is missing a field that the SVG-Exporter is expecting to find. Compass puts several "Inkscape" tags in the data to insure that the data is Inkscape-compatible. In some instances, Compass has to add that data if it doesn't find it. For example, if the data came straight from Illustrator, it would not have Inkscape tags. You have probably put your data through a slightly different series of operations than anything I tried here and some or all of the tags are missing. If you could send me a copy of your data, I could find the problem and fix it. I will send you a private email with my address so you can send me the data if you are willing to. Send me both the plot file and the SVG file. I will keep data confidential and it won't be shown or given to anyone else. Larry Fish _____ From: [email protected] [mailto:[email protected]] On Behalf Of derekbristol Sent: Friday, February 04, 2011 10:06 PM To: [email protected] Subject: [compass-users] Round Tripping Issues I'm working on a cave map with only 4 loops. I thought that when I originally compiled the data I had set Compass to close those loops, but apparently did not. It placed all of the loop error on the closing shot. I've addressed this, but had already fully drawn the cave in Illustrator CS. So I figured this would be a good opportunity to learn round tripping and fix the small deviation between the drawing and new line plot with properly closed loops. I open the new .plt file in SVGexporter, but when I go to "Load Target", and select my Illustrator file (.svg format), I get an error: "Reference to undeclared namespace prefix: 'inkscape'.". I've tried a number of things, but it seems that whenever I take any SVG export from Compass, and add anything at all to it in Illustrator, I get this same error message when I try to Merge / Morph. Anyone else have this problem or know how to solve it? I'm curious about the Inkscape reference in the error message. I'm not using Inkscape and haven't converted to Inkscape .svg format. Thanks, Derek
Date: Thu, 17 Feb 2011 13:19:39 -0700
From: "Larry Fish"
Subject: Issue With UTM Convergence
Hi Everyone,
I thought I'd point out a potential problem that was brought to my
attention by Paul De Bie.
The UTM Convergence angle was wrong for any cave or location east of
zero degrees longitude. That would be a small part of England, Europe
excluding Spain, the Eastern 2/3s of Africa, all of Asia and Australia.
Any locations west of zero degrees longitude are unaffected. So North and
South America are excluded. So are most of England, Spain and Western
Africa.
There is a new version of Compass on the internet that fixes this
problem. If you have been having Compass adjust your maps to match the UTM
grid and the caves are in the specified areas, your caves maybe rotated by a
few degrees. You may want to recompile your data with the new version.
Larry
Date: Tue, 22 Feb 2011 20:32:12 -0700 From: "Mark, Sue, Alex, and John" Subject: kml exports? I couldn't find this in the manuals (maybe looking the wrong place?) but how do you export kml files to google earth? Thanks in advance, Mark
Date: Tue, 22 Feb 2011 21:44:30 -0700
From: "Larry Fish"
Subject: RE: [compass-users] kml exports?
Hi Mark,
Load your project into the Project Manager. Press the "Process and View
Cave" button. This will take you to the Compass Viewer. Under the "File"
menu option, select "Export 3D Formats." In the 3D Exporter, select the last
tab "KML". Change the settings to whatever you want and then press the
Export KML File button. If you have google earth installed on your computer,
you can just double click on the file, and Google Earth will load it and
zoom in on the cave location.
The cave must be geo-referenced for this to work. This means that at least
station in the cave must be referenced to UTM or Long/Lat. Let me know if
you need instructions for doing this.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Mark, Sue, Alex, and John
Sent: Tuesday, February 22, 2011 8:32 PM
Subject: [compass-users] kml exports?
I couldn't find this in the manuals (maybe looking the wrong place?) but how
do you export kml files to google earth?
Thanks in advance,
Mark
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
st1\:*{behavior:url(#default#ieooui) }
Date: Wed, 23 Feb 2011 09:32:46 -0500
From: Tony Canike
Subject: Re: [compass-users] kml exports?
The KML exports are pretty easy and the results are quite impressive,
given how little effort is required. The first time I used Larry's KML
export and plotted on Goggle Earth, I kicked myself for not doing it
earlier.
I'm just waiting for Google Geology-3d(tm) to be released, so we can
then zoom underground and see the relationship between the cave
passages, earth's surface, and the contact surfaces between the various
geological formations (sandstone, limestone, shale, etc.)
Along with Google HydroEarth(tm), which will incorporate a display the
results of dye traces between your various caves and springs in 3d.
On 2/22/2011 11:44 PM, Larry Fish ([email protected]) wrote
Hi Mark,
Load your project into the Project Manager. Press the "Process and
View Cave" button. This will take you to the Compass Viewer. Under the
"File" menu option, select "Export 3D Formats." In the 3D Exporter,
select the last tab "KML". Change the settings to whatever you want
and then press the Export KML File button. If you have google earth
installed on your computer, you can just double click on the file, and
Google Earth will load it and zoom in on the cave location.
The cave must be geo-referenced for this to work. This means that at
least station in the cave must be referenced to UTM or Long/Lat. Let
me know if you need instructions for doing this.
Larry
------------------------------------------------------------------------
*From:*[email protected]
[mailto:[email protected]] *On Behalf Of *Mark, Sue, Alex,
and John
*Sent:* Tuesday, February 22, 2011 8:32 PM
*To:* [email protected]
*Subject:* [compass-users] kml exports?
I couldn't find this in the manuals (maybe looking the wrong place?)
but how do you export kml files to google earth?
Thanks in advance,
Mark
The KML exports are pretty easy and the results are quite
impressive, given how little effort is required. The first time I
used Larry's KML export and plotted on Goggle Earth, I kicked myself
for not doing it earlier.
I'm just waiting for Google Geology-3d(tm) to be released, so we can
then zoom underground and see the relationship between the cave
passages, earth's surface, and the contact surfaces between the
various geological formations (sandstone, limestone, shale, etc.)
Along with Google HydroEarth(tm), which will incorporate a display
the results of dye traces between your various caves and springs in
3d.
On 2/22/2011 11:44 PM, Larry Fish ([email protected]) wrote
Hi Mark,
Load your project
into the Project
Manager. Press the “Process and View
Cave”
button. This will take you to the Compass Viewer.
Under the “File”
menu option, select “Export 3D Formats.” In the 3D
Exporter, select
the last tab “KML”. Change the settings to whatever
you want and
then press the Export KML File button. If you have
google earth installed on
your computer, you can just double click on the
file, and Google Earth will
load it and zoom in on the cave location.
The cave must be
geo-referenced for this
to work. This means that at least station in the
cave must be referenced to UTM
or Long/Lat. Let me know if you need instructions
for doing this.
Larry
From:
[email protected]
[mailto:[email protected]] On Behalf Of Mark,
Sue, Alex, and John
Sent:
Tuesday, February 22, 2011
8:32 PM
To:
[email protected]
Subject:
[compass-users] kml
exports?
I
couldn't find this in the manuals (maybe
looking the wrong place?) but how do
you export kml files to google earth?
Thanks in advance,
Mark
Date: Wed, 23 Feb 2011 09:56:31 -0500 From: Luc Le Blanc Subject: Re: [compass-users] kml exports? 2011/2/23 Tony Canike I'm just waiting for Google Geology-3d(tm) to be released, so we can then zoom underground and see the relationship between the cave passages, earth's surface, and the contact surfaces between the various geological formations (sandstone, limestone, shale, etc.) Are you dreaming up new software? I ran into this problem of an underground cave (!) when implementing the KML export in Auriga: the cave has to be on the surface since there is no transparency setting to be able to "see" it through the surface. Just that would help a lot. Luc Le Blanc http://www.speleo.qc.ca/Auriga
Date: Thu, 17 Mar 2011 20:26:31 -0000 From: "RyanY" Subject: Cave Viewer inconsistently renders incorrect angles Hi All! I apologize if I'm asking something that's already been asked, but I've searched both this forum and the archives and couldn't find a similar problem. My friend has over 6300 feet of survey data in Compass for a cave he's been working on. He recently bought a Pocket PC to sketch the cave with using PocketTopo. I offered to convert his Compass files into PocketTopo files. When I did, the PocketTopo plot looked a little different. I overlayed the Compass plot with the PocketTopo plot, and they definitely were different. I assumed it was an error of mine during file manipulation. I decided to determine the source of the error, so I made brand new survey files in Compass and PocketTopo *using only front sights* and *the exact same data*. Somehow, the plots were STILL different! So, I pulled out my transparent cave survey protractor and began measuring. Lo and behold, the Compass plot had some angles that were incorrect! I verified that I was in the plan view, so it's not that I was looking at the plot from a skewed view. Exporting to bitmap still gives me the wrong angles. What's interesting is that when I print the plot out, the angles SUDDENLY BECOME CORRECT! Do you guys have any ideas?? I suppose this could be some sort of incorrect setting, but surely there isn't a setting to make the angles appear different that what they should be? Could this be some sort of software bug? I have version 5.11.1.1.165 Here are some pictures of the plots: http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098 Thanks for any help you can give.
Date: Thu, 17 Mar 2011 16:05:17 -0700 (PDT) From: Tom Subject: Re: [compass-users] Cave Viewer inconsistently renders incorrect angles Can you look at the raw data(text based file) and see if angles and distances work? Is it possible one program is showing 3D in view and another program showing 2D, so that when you printscreen you get skewed data? Can you post the data file so someone else can try manipulating the data to get the same resutls as you? p.s. how do you like PocketTopo? Tom From: RyanY Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles To: [email protected] Date: Thursday, March 17, 2011, 1:26 PM Hi All! I apologize if I'm asking something that's already been asked, but I've searched both this forum and the archives and couldn't find a similar problem. My friend has over 6300 feet of survey data in Compass for a cave he's been working on. He recently bought a Pocket PC to sketch the cave with using PocketTopo. I offered to convert his Compass files into PocketTopo files. When I did, the PocketTopo plot looked a little different. I overlayed the Compass plot with the PocketTopo plot, and they definitely were different. I assumed it was an error of mine during file manipulation. I decided to determine the source of the error, so I made brand new survey files in Compass and PocketTopo *using only front sights* and *the exact same data*. Somehow, the plots were STILL different! So, I pulled out my transparent cave survey protractor and began measuring. Lo and behold, the Compass plot had some angles that were incorrect! I verified that I was in the plan view, so it's not that I was looking at the plot from a skewed view. Exporting to bitmap still gives me the wrong angles. What's interesting is that when I print the plot out, the angles SUDDENLY BECOME CORRECT! Do you guys have any ideas?? I suppose this could be some sort of incorrect setting, but surely there isn't a setting to make the angles appear different that what they should be? Could this be some sort of software bug? I have version 5.11.1.1.165 Here are some pictures of the plots: http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098 Thanks for any help you can give. ------------------------------------ Yahoo! Groups Links � � [email protected]
Date: Thu, 17 Mar 2011 23:40:01 -0000 From: "RyanY" Subject: Re: Cave Viewer inconsistently renders incorrect angles Hi Tom, I'm sure yours is a simple question, but I can't seem to quite understand what you are asking. What do you mean by "work," and which file (Compass or -Topo) are you asking about? The Compass file looks fine in Notepad (as far as I can tell... not too sure about some of the header information, but the survey data all looks OK), but the PocketTopo file is some sort of encoded jibberish. For those of you who are wondering how I made a working PocketTopo file by hand, I lied... I actually made a VisualTopo file by hand which can be imported by PocketTopo and is of similar formatting to a Compass file. I had checked to see if the compass view was slightly skewed, giving me a slight 3D view, but I found that this was not the case. The plot is incorrect as soon as the Cave Viewer opens (default view is plan view, I think). To double check, I went ahead and put it in profile view and 3D-rotated it to where I was overlooking the top again (plan). The angles were still incorrect. By the way, in one instance, I am getting as much as over 10 degrees off. I have posted both the Compass file (first.DAT) and the PocketTopo file (first.top) to the Files section of this group for anyone who may want to view them. Both have been generated by their respective programs and neither have been altered manually. To see what I see, you might want to make sure to have the same Compass version as I do, and see if it's the software version or something else. Does anyone know if there's a place where I can download an older version of Compass? If my plot suddenly looks the way it's supposed to when using the older version, then we'll know that the version I've been using has a bug. Thanks. P.S. I haven't had the opportunity to use his PDA with PocketTopo yet, but his demonstration of it left me with the impression that it's a very simple yet effective and accurate program. If I had a PDA, I think I'd definitely use it. Entering the data in via stylus (depending on PDA method of input) might get somewhat annoying though. Can you look at the raw data(text based file) and see if angles and distances work? Is it possible one program is showing 3D in view and another program showing 2D, so that when you printscreen you get skewed data? Can you post the data file so someone else can try manipulating the data to get the same resutls as you? p.s. how do you like PocketTopo? Tom --- On Thu, 3/17/11, RyanY wrote: From: RyanY Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles To: [email protected] Date: Thursday, March 17, 2011, 1:26 PM Hi All! I apologize if I'm asking something that's already been asked, but I've searched both this forum and the archives and couldn't find a similar problem. My friend has over 6300 feet of survey data in Compass for a cave he's been working on. He recently bought a Pocket PC to sketch the cave with using PocketTopo. I offered to convert his Compass files into PocketTopo files. When I did, the PocketTopo plot looked a little different. I overlayed the Compass plot with the PocketTopo plot, and they definitely were different. I assumed it was an error of mine during file manipulation. I decided to determine the source of the error, so I made brand new survey files in Compass and PocketTopo *using only front sights* and *the exact same data*. Somehow, the plots were STILL different! So, I pulled out my transparent cave survey protractor and began measuring. Lo and behold, the Compass plot had some angles that were incorrect! I verified that I was in the plan view, so it's not that I was looking at the plot from a skewed view. Exporting to bitmap still gives me the wrong angles. What's interesting is that when I print the plot out, the angles SUDDENLY BECOME CORRECT! Do you guys have any ideas?? I suppose this could be some sort of incorrect setting, but surely there isn't a setting to make the angles appear different that what they should be? Could this be some sort of software bug? I have version 5.11.1.1.165 Here are some pictures of the plots: http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098 Thanks for any help you can give. ------------------------------------ Yahoo! Groups Links � � [email protected]
Date: Fri, 18 Mar 2011 03:46:32 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect angles
Ryan,
Thanks for your email and questions. I've down loaded your files and images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide. If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on the
screen. If you change the video mode to 2048 by 1536, the actual screen DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of 16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a 4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200 all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross with
equal length arms. (Be sure to expand the Viewer to full screen and zoom out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different. I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass plot
had some angles that were incorrect! I verified that I was in the plan view,
so it's not that I was looking at the plot from a skewed view. Exporting to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
Date: Fri, 18 Mar 2011 08:11:52 -0400
From: "[email protected]"
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect angles
I ran into the aspect ratio problem when I was taking screen shots from the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
From: Larry Fish [email protected]
Date: Fri, 18 Mar 2011 03:46:32 -0600
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide. If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on the
screen. If you change the video mode to 2048 by 1536, the actual screen DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of 16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a 4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200 all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross with
equal length arms. (Be sure to expand the Viewer to full screen and zoom out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different. I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass plot
had some angles that were incorrect! I verified that I was in the plan view,
so it's not that I was looking at the plot from a skewed view. Exporting to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Fri, 18 Mar 2011 14:32:00 -0000
From: "RyanY"
Subject: Re: Cave Viewer inconsistently renders incorrect angles
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually printed to PDF (via CutePDF), not on paper. I would have thought that if it were an issue with an incorrect screen setting, the PDF plot should also show incorrect angles, but as far as I can tell the PDF plot is entirely accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen image from their computer of the same survey data the plot looks perfect to me, but when I make a bitmap screen image with my computer the plot remains incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a resolution "purist" and always make sure to have my screen set at the maximum, undistorted resolution possible. My screen is 14" (diagonal) and is set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no distortion). Finally, it appears that the monitor adjustment is such that there is no black space between or extension beyond the screen borders (the screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2), but it usually does the job just fine. My operating system is XP Home Edition, 2002, SP3. I've verified that my driver is the most up to date there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of course, the PDF plot lengths measured to be equal. I used the distance tool and the program told me they were the same lengths, so Compass thinks it's showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I am of the opinion that Compass may have rendering issues with my particular resolution.
Does anyone else have the newest Compass version and have a WXGA monitor that can confirm or deny my theory? It would also be interesting to see if running an older Compass version on my computer would produce the same distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved so I can use this fantastic program with confidence!
:
I ran into the aspect ratio problem when I was taking screen shots from the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@...
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide. If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on the
screen. If you change the video mode to 2048 by 1536, the actual screen DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of 16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a 4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200 all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross with
equal length arms. (Be sure to expand the Viewer to full screen and zoom out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different. I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass plot
had some angles that were incorrect! I verified that I was in the plan view,
so it's not that I was looking at the plot from a skewed view. Exporting to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
--------------------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Fri, 18 Mar 2011 11:19:04 -0400
From: Tony Canike
Subject: Re: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles
Ryan,
What happens when you draw a cross with equal length lines in another program (Visio, Inkscape, OOo Draw, Illustrator) ?
Are the lines displayed equally long on your monitor?
Tony.
On Mar 18, 2011, at 10:32 AM, "RyanY" wrote:
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually printed to PDF (via CutePDF), not on paper. I would have thought that if it were an issue with an incorrect screen setting, the PDF plot should also show incorrect angles, but as far as I can tell the PDF plot is entirely accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen image from their computer of the same survey data the plot looks perfect to me, but when I make a bitmap screen image with my computer the plot remains incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a resolution "purist" and always make sure to have my screen set at the maximum, undistorted resolution possible. My screen is 14" (diagonal) and is set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no distortion). Finally, it appears that the monitor adjustment is such that there is no black space between or extension beyond the screen borders (the screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2), but it usually does the job just fine. My operating system is XP Home Edition, 2002, SP3. I've verified that my driver is the most up to date there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of course, the PDF plot lengths measured to be equal. I used the distance tool and the program told me they were the same lengths, so Compass thinks it's showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I am of the opinion that Compass may have rendering issues with my particular resolution.
Does anyone else have the newest Compass version and have a WXGA monitor that can confirm or deny my theory? It would also be interesting to see if running an older Compass version on my computer would produce the same distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@..." wrote:
I ran into the aspect ratio problem when I was taking screen shots from the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@...
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide. If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on the
screen. If you change the video mode to 2048 by 1536, the actual screen DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of 16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a 4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200 all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross with
equal length arms. (Be sure to expand the Viewer to full screen and zoom out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different. I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass plot
had some angles that were incorrect! I verified that I was in the plan view,
so it's not that I was looking at the plot from a skewed view. Exporting to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Ryan,What happens when you draw a cross with equal length lines in another program (Visio, Inkscape, OOo Draw, Illustrator) ?Are the lines displayed equally long on your monitor?Tony. On Mar 18, 2011, at 10:32 AM, "RyanY" <[email protected]> wrote:
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually printed to PDF (via CutePDF), not on paper. I would have thought that if it were an issue with an incorrect screen setting, the PDF plot should also show incorrect angles, but as far as I can tell the PDF plot is entirely accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen image from their computer of the same survey data the plot looks perfect to me, but when I make a bitmap screen image with my computer the plot remains incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a resolution "purist" and always make sure to have my screen set at the maximum, undistorted resolution possible. My screen is 14" (diagonal) and is set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no distortion). Finally, it appears that the monitor adjustment is such that there is no black space between or extension beyond the screen borders (the screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2), but it usually does the job just fine. My operating system is XP Home Edition, 2002, SP3. I've verified that my driver is the most up to date there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of course, the PDF plot lengths measured to be equal. I used the distance tool and the program told me they were the same lengths, so Compass thinks it's showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I am of the opinion that Compass may have rendering issues with my particular resolution.
Does anyone else have the newest Compass version and have a WXGA monitor that can confirm or deny my theory? It would also be interesting to see if running an older Compass version on my computer would produce the same distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved so I can use this fantastic program with confidence!
:[email protected]"[email protected], "dwightliv@..." <dwightliv@...> wrote:
>
> I ran into the aspect ratio problem when I was taking screen shots from the
> Compass viewer. I intended to get into my graphics card controls to see if
> I could correct it. As it happened, I first updated my ATI graphic card
> driver, and that fixed the problem.
>
> The easiest test is to create a couple of fake shots, both the same length,
> one north-south and the other east-west. Get a screen shot into Photoshop
> and compare the pixel length.
>
> Dwight
>
>
> Original Message:
> -----------------
> From: Larry Fish lfish@...
> Date: Fri, 18 Mar 2011 03:46:32 -0600
> To:
Date: Fri, 18 Mar 2011 15:52:47 -0000
From: "RyanY"
Subject: Re: Cave Viewer inconsistently renders incorrect angles
Tony,
I had none of those programs on my computer, so I went ahead and installed Inkscape. I have absolutely no experience with it so I may have drawn lines incorrectly, but I went to "Draw bezier curves and straight lines", chose the "Create a sequence of paraxial line segments" mode, created a 90� straight line in the positive y-direction for a distance of 100 px, then made another 90� straight line in the positive x-direction for a distance of 100 px. Measuring these two line segments, I found that both lines were equally long on my monitor.
G&
Ryan,
What happens when you draw a cross with equal length lines in another program (Visio, Inkscape, OOo Draw, Illustrator) ?
Are the lines displayed equally long on your monitor?
Tony.
On Mar 18, 2011, at 10:32 AM, "RyanY" wrote:
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually printed to PDF (via CutePDF), not on paper. I would have thought that if it were an issue with an incorrect screen setting, the PDF plot should also show incorrect angles, but as far as I can tell the PDF plot is entirely accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen image from their computer of the same survey data the plot looks perfect to me, but when I make a bitmap screen image with my computer the plot remains incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a resolution "purist" and always make sure to have my screen set at the maximum, undistorted resolution possible. My screen is 14" (diagonal) and is set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no distortion). Finally, it appears that the monitor adjustment is such that there is no black space between or extension beyond the screen borders (the screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2), but it usually does the job just fine. My operating system is XP Home Edition, 2002, SP3. I've verified that my driver is the most up to date there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of course, the PDF plot lengths measured to be equal. I used the distance tool and the program told me they were the same lengths, so Compass thinks it's showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I am of the opinion that Compass may have rendering issues with my particular resolution.
Does anyone else have the newest Compass version and have a WXGA monitor that can confirm or deny my theory? It would also be interesting to see if running an older Compass version on my computer would produce the same distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots from the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide. If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on the
screen. If you change the video mode to 2048 by 1536, the actual screen DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of 16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a 4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200 all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross with
equal length arms. (Be sure to expand the Viewer to full screen and zoom out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect angles
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different. I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass plot
had some angles that were incorrect! I verified that I was in the plan view,
so it's not that I was looking at the plot from a skewed view. Exporting to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 02:21:54 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles
Ryan,
I neglected to mention that when I "printed" the cave plot,
I actually printed to PDF (via CutePDF), not on paper. I
would have thought that if it were an issue with an
incorrect screen setting, the PDF plot should also show
incorrect angles, but as far as I can tell the PDF plot is
entirely accurate.
To understand my thinking on this problem, I need to explain how Compass
works in a little more detail. When Compass displays a cave on a video
screen or prints it to a printer, it uses one of Windows' standard graphics
modes. Windows allows you to draw lines using one of several different Units
of measure:
Pixels
Inches
Millimeters
http://msdn.microsoft.com/en-us/library/dd162980%28v=vs.85%29.aspx
Compass sets the Units to "Inches". This allows Compass to work in "Paper"
coordinates instead of pixels. That means that Compass doesn't have to know
anything about the display device, its resolution or aspect ratio. Instead,
Windows and the Device driver are responsible for dealing with variations in
devices and translating Inches to whatever units the device uses. This
allows Compass to Display and Print using exactly the same code and, in
theory, the printout and display should match each other exactly.
The technique relies on the fact that the device driver for each device is
responsible performing the calculations necessary to scale the image for the
device. Since the Device Driver has access to all kinds of information about
the display device, printer or video card, it is in a better position to
accurately display the image. Obviously, your printer's device driver is
doing the calculations correctly.
This suggests that the problem is with the video driver. (I have seen one
other instance where the driver caused a problem with Compass. This was on
an HP large-format color printer, where the HP driver couldn't handle a
print-out larger than 32.76 inches because of a math overflow problem in the
driver.)
The CutePDF probably uses a different method of drawing the image that
bypasses this particular driver mode. I could bypass the driver myself, but
there would be an even bigger risk of problems because I'd have to rely on
what the driver reported about the resolution. If a driver can screw up
displaying Inches, it is just as likely to report erroneous information
about screen resolutions and aspect ratios etc.
Another thing worth mentioning is that when someone sends me
a bitmap screen image from their computer of the same survey
data the plot looks perfect to me, but when I make a bitmap
screen image with my computer the plot remains incorrect.
Again, that is another indication that the device driver is not handling the
situation correctly. Your cave displays perfectly on my computer and if I
use Compass to measure the angles, they are perfect.
Another thing you might want to try is to check the "Save Screen Image"
option to export a bitmap. You've probably already used it, but it is
important to note that there are two different modes operation. If you
haven't, it is under the "File" menu option. Here are the two modes of
operation:
1. UNSCALED. If the "Scaled Checkbox" is NOT enabled, the Viewer makes a
direct copy of what appears on the screen and saves it as a bitmap. This is
the default mode.
2. SCALED. If the "Scaled Checkbox" is enabled, the Viewer plots directly on
a bitmap. This bypasses the video screen and video drivers, so it is a good
test.
If the second option exports correctly and the first one does not, it is a
further indication that there is a driver problem.
Also, if you saw Dwight's earlier post, he had similar problem that he
solved by updating the device driver. That lends credibility to the idea
that the device driver is causing the problem.
Admittedly, my screen is of oddball proportions (WXGA), but
I am a resolution "purist" and always make sure to have my
screen set at the maximum, undistorted resolution possible.
My screen is 14" (diagonal) and is set to 1280x768 pixels. I
believe with these settings, everything is 1:1 (no
distortion). Finally, it appears that the monitor adjustment
is such that there is no black space between or extension
beyond the screen borders (the screen has been automatically
adjusted).
The 14-inch diagonal number doesn't tell me what aspect ratio your screen
is. I need both the Length and Width to know that. Here are the most common
aspect ratios:
4:3 = 1.33333
16:9 = 1.777777
16:10 (8:5) = 1.6
http://en.wikipedia.org/wiki/Aspect_ratio_(image)
http://en.wikipedia.org/wiki/16:10
You've set your video mode to:
1280x768 = 1.6666666
This corresponds to a 15:9 (5:3). 15:9 is a relatively uncommon aspect
ratio. If your video screen is not actually 15:9, then using a 1280x768
resolution will certainly distort the image. If the screen is actually 15:9,
the fact that it is fairly non-standard could be a problem.
The Wikipedia article on WXGA seems to indicate that using 1280x768 mode on
a WXGA card will result in "Non-Square" pixels. That also could be causing
your problem. From the Wikipedia article:
"1280x720 provides perfectly square pixels at an aspect ratio
of 16:9, while the additional pixels in 1280x768 and
1280x800 must be ignored to give the 16:9 ratio without
vertical stretching of the image."
If the device driver has to ignore pixels to get an unstretched image, that
makes me suspicious.
http://en.wikipedia.org/wiki/Graphic_display_resolutions
WXGA supports several other resolutions that might work better:
1280x720 - 16:9
1280x768 - 15:9 (5:3)
1280x800 - 16:10 (8:5)
1360x768 - 16:9 (approx.)
1366x768 - 16:9 (approx.)
Manufacturers typical spend less time and energy writing driver for
non-standard modes. They also use special tricks to make them work. For
example, I've seen drivers that skip points and don't display every pixel.
They will also duplicate pixels. All this is done to force a certain number
pixels to fit within a certain amount of space.
For example, if the video frame buffer has 720 pixels but it needs to
display 768 pixels, it will be 48 pixels short. Duplicating every 15 pixel
will make up the difference, and wouldn't be particularly noticeable in
ordinary applications. However, duplicating those pixels would stretch the
image by about 6%, which would cause noticeable distortions in a program
like Compass.
Over the years I've experienced lots of problems with Graphics drivers, even
on Mainstream Video cards like NVidia and ATI. None of them affected Compass
per se, but they would produce odd screen anomalies that were eventually
fixed when the drivers were updated.
The chips for your graphics adapter were originally released in 2005 and the
last driver update from Intel is 2006, so it's been at least 5 years since
Intel worked on them.
http://www.intel.com/support/graphics/intel852gm/sb/cs-009059.htm
There are quite a few reports on the internet about driver problems for your
graphics chips. For example, here is one from 2007 (well after your last
driver update,) where the complaint is about screen stretching:
http://forums.techguy.org/hardware/620734-strange-problem-intel-82852-82855-
a.html
Others:
http://www.sevenforums.com/drivers/3128-intel-82852-82855-gm-gme-graphics-co
ntroller-driver-issues.html
http://social.technet.microsoft.com/Forums/en-US/itprovistahardware/thread/6
80b3aa5-7614-45ef-b213-feebf40bc232/
Perhaps unsurprisingly, I opened your cross.plt file and
the N-S lengths were indeed shorter than the W-E lengths
(approximately a 4:5 ratio). Of course, the PDF plot lengths
measured to be equal. I used the distance tool and the
program told me they were the same lengths, so Compass
thinks it's showing me an accurate plot.
That pretty much verifies that the problem is occurring somewhere between
Windows and the graphic card. That makes it very likely that it is the
driver.
Let me know if you have any other questions or thoughts.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen
image from their computer of the same survey data the plot looks perfect to
me, but when I make a bitmap screen image with my computer the plot remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal) and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no
distortion). Finally, it appears that the monitor adjustment is such that
there is no black space between or extension beyond the screen borders (the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance tool
and the program told me they were the same lengths, so Compass thinks it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I
am of the opinion that Compass may have rendering issues with my particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved
so I can use this fantastic program with confidence!
:
I ran into the aspect ratio problem when I was taking screen shots from
the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@...
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a
file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number
of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide.
If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on
the
screen. If you change the video mode to 2048 by 1536, the actual screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio
of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of
16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes
the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross
with
equal length arms. (Be sure to expand the Viewer to full screen and zoom
out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view. Exporting
to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 02:32:08 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles
Ryan,
I wrote a small program to test the different modes that are available on a
graphics card. It may give us more information about your problem. I have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines will
be thicker in "Pixel" mode and the cross might jump by one pixel because of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen
image from their computer of the same survey data the plot looks perfect to
me, but when I make a bitmap screen image with my computer the plot remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal) and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no
distortion). Finally, it appears that the monitor adjustment is such that
there is no black space between or extension beyond the screen borders (the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance tool
and the program told me they were the same lengths, so Compass thinks it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I
am of the opinion that Compass may have rendering issues with my particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved
so I can use this fantastic program with confidence!
:
I ran into the aspect ratio problem when I was taking screen shots from
the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@...
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a
file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number
of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide.
If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on
the
screen. If you change the video mode to 2048 by 1536, the actual screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio
of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of
16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes
the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross
with
equal length arms. (Be sure to expand the Viewer to full screen and zoom
out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view. Exporting
to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 15:01:35 -0000
From: "RyanY"
Subject: Re: Cave Viewer inconsistently renders incorrect angles
Larry,
Unfortunately, I have made a huge error, which has resulted in a waste of your guys' time. Fortunately, the Cave Viewer is now correctly rendering plots! I told you I had the latest driver, but this was not true. You see, I own a Compaq Presario v2010us and traditionally I've always went to the computer manufacturer website to look for the latest drivers, which is what I did this time. The website claimed it was the most up to date driver, but it wasn't. While looking for the umpteenth time at my graphics properties, I noticed Intel has an option to go to their website to check for drivers. Desperate, I gave this a try and sure enough, the Compaq website had lied to me. After downloading, installing and rebooting, the Compass plotter works fine now.
I tried all your suggestions from your previous two postings before installing the new driver including using your small program, and if you are still interested in my findings I can elaborate. It's probably not worth mentioning, but after installing the driver, the crosses are occurring at different inch origins (pixel mode - 0,0 ; other modes - 8.2,8.2) and the 0.01 inch mode is as thick as the pixel mode. But as long as the Compass Cave Viewer is working correctly this doesn't concern me.
I apologize for wasting your time, and thank you for all the help you've given. I've definitely learned some things about Compass (and the reliability of Compaq support) that I never knew before. Kudos to you Larry for your excellent program and phenomenal support. Thanks also to this wonderful community of Compass users.
G&
Ryan,
I wrote a small program to test the different modes that are available on a
graphics card. It may give us more information about your problem. I have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines will
be thicker in "Pixel" mode and the cross might jump by one pixel because of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen
image from their computer of the same survey data the plot looks perfect to
me, but when I make a bitmap screen image with my computer the plot remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal) and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no
distortion). Finally, it appears that the monitor adjustment is such that
there is no black space between or extension beyond the screen borders (the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance tool
and the program told me they were the same lengths, so Compass thinks it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I
am of the opinion that Compass may have rendering issues with my particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved
so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots from
the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a
file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number
of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide.
If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on
the
screen. If you change the video mode to 2048 by 1536, the actual screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio
of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of
16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes
the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross
with
equal length arms. (Be sure to expand the Viewer to full screen and zoom
out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view. Exporting
to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 15:27:07 -0000
From: "RyanY"
Subject: Re: Cave Viewer inconsistently renders incorrect angles
All,
I am not quite sure how downloading a later driver solved the problem, but I am glad Compass is performing perfectly now!
I am curious though, if I had used the older driver when it first came out, would Compass have rendered an incorrect plot then as well? Or, did the driver become inaccurate for Compass over time? Perhaps installing a new driver reset a setting that was set wrong?
Knowing that one's computer can influence the plot of Compass, what is the easiest and quickest way for one ensure Compass is plotting correctly? To periodically open your cross.plt file and measure the lengths?
Thanks,
Ryan
Larry,
Unfortunately, I have made a huge error, which has resulted in a waste of your guys' time. Fortunately, the Cave Viewer is now correctly rendering plots! I told you I had the latest driver, but this was not true. You see, I own a Compaq Presario v2010us and traditionally I've always went to the computer manufacturer website to look for the latest drivers, which is what I did this time. The website claimed it was the most up to date driver, but it wasn't. While looking for the umpteenth time at my graphics properties, I noticed Intel has an option to go to their website to check for drivers. Desperate, I gave this a try and sure enough, the Compaq website had lied to me. After downloading, installing and rebooting, the Compass plotter works fine now.
I tried all your suggestions from your previous two postings before installing the new driver including using your small program, and if you are still interested in my findings I can elaborate. It's probably not worth mentioning, but after installing the driver, the crosses are occurring at different inch origins (pixel mode - 0,0 ; other modes - 8.2,8.2) and the 0.01 inch mode is as thick as the pixel mode. But as long as the Compass Cave Viewer is working correctly this doesn't concern me.
I apologize for wasting your time, and thank you for all the help you've given. I've definitely learned some things about Compass (and the reliability of Compaq support) that I never knew before. Kudos to you Larry for your excellent program and phenomenal support. Thanks also to this wonderful community of Compass users.
Ryan
--- In [email protected], "Larry Fish" wrote:
Ryan,
I wrote a small program to test the different modes that are available on a
graphics card. It may give us more information about your problem. I have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines will
be thicker in "Pixel" mode and the cross might jump by one pixel because of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap screen
image from their computer of the same survey data the plot looks perfect to
me, but when I make a bitmap screen image with my computer the plot remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal) and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1 (no
distortion). Finally, it appears that the monitor adjustment is such that
there is no black space between or extension beyond the screen borders (the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics 2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance tool
and the program told me they were the same lengths, so Compass thinks it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the moment I
am of the opinion that Compass may have rendering issues with my particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be solved
so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots from
the
Compass viewer. I intended to get into my graphics card controls to see if
I could correct it. As it happened, I first updated my ATI graphic card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an "aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display a
file
on the screen, so that rules out a mathematical or graphics error inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact number
of
Pixels per inch in the left-right and up-down direction. As a result, when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution are
based on a several extraneous factors. For example, Windows assumes the
screen has a certain resolution. Typically, it is 96 Dots Per Inch (DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't really
mean that a line 96 dots long will really be 1-inch long. The actual size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches wide.
If
your video card is set to a resolution of 1024 by 768, there will be 1024
pixels across the 16 inch width of the screen. That means that the actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long on
the
screen. If you change the video mode to 2048 by 1536, the actual screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long. The
important thing to understand is that, unlike the printer, the monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the ratio
of
the width to the height is 4 to 3. Newer, high def monitors have aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios of
16
to 10.
When you setup your computer for a particular monitor, you must select a
screen resolution that matches your monitor. If you don't, the pixels will
be stretched or squashed horizontally or vertically. If this happens, the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem you
are seeing.
The first thing I would try is to check the monitor's resolution and make
sure your video card was set for a matching resolution. Some people don't
like to set their video resolution to match the monitor because it makes
the
text too small. If you are doing this, make sure the resolution you choose
match the aspect ratio of your monitor. For example, if your monitor has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and 1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of the
image on the monitor using the monitor's own controls. For example, you
might prefer to have an 1/4 inch border on the right and left side of the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows you
to lock a cursor on each survey station ("Nearest Station" button) so the
measurements are extremely accurate. When I do this, the angles exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is NOT
measuring it visually, so it ignores all kinds of resolution, scaling and
aspect ratio issues.
This is something you can do too and it would give us more information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You should
download it and double click on it. If Compass is installed properly, the
Viewer will automatically open the file. The image will display a cross
with
equal length arms. (Be sure to expand the Viewer to full screen and zoom
out
a bit so you can see the ends of the cross arms.) If you measure the cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked, but
I've searched both this forum and the archives and couldn't find a similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with using
PocketTopo. I offered to convert his Compass files into PocketTopo files.
When I did, the PocketTopo plot looked a little different. I overlayed the
Compass plot with the PocketTopo plot, and they definitely were different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view. Exporting
to
bitmap still gives me the wrong angles. What's interesting is that when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 09:53:23 -0600
From: [email protected]
Subject: Re: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles
Ryan,
I can understand your desire for your application to have the lineplot be
exact on the screen. When I am using Compass to view lineplots on the
screen, I really don't worry too much about the lineplot not being exact. I
only view lineplots on the screen to look at approximate relationships and
since my eyeballs aren't calibrated very precisely, as long as the
lineplots are close I'm OK. Where I do care about the lineplots being
correct is when I export them to ARCGIS to view the lineplot in
relationship to surface features, other caves, geology maps and etcetera,
however I export those maps as Shape files which shouldn't have the
problems you experienced. I also care about the lineplots being correct
when I export them to Adobe Illustrator to draft a map. However, since I
use the SVG exporter to export maps to Illustrator I shouldn't have the
problems you experienced. Since most of the caves I work with are maze
caves with lots of loop closures, I am always careful to check the close
loop option on the Project Manager screen when exporting a lineplot where I
care about exact relationships.
I applaud your tenacity and thoroughness in tracking down the problem you
were having with your driver. I admit that I have never checked the Compass
Shape File and SVG exports to make sure that they are exactly correct, but
merely assume that they work properly. I did learn something from your
experiences concerning video drivers, screen resolutions and ratios. I also
learned from Larry that exporting lineplots as a scaled BMP or WMF is a way
to make sure the lineplot is correct despite video driver, screen
resolution and ratio problems. Before the SVG exporter I used to export
lineplots as WMF for drafting cave maps. I had intuitively exported them as
scaled files, since that seemed like a better option and now I know why. I
wonder if it is really ever desirable to export an unscaled BMP or WMF and
whether it might be better to have the default be to have the scaled output
box checked.
Good luck with your cave survey and cartography projects,
Stan Allison
"RyanY"
To
Sent by: [email protected]
compass-users@yah cc
oogroups.com
Subject
[compass-users] Re: Cave Viewer
03/19/2011 09:27 inconsistently renders incorrect
AM angles
Please respond to
compass-users@yah
oogroups.com
All,
I am not quite sure how downloading a later driver solved the problem, but
I am glad Compass is performing perfectly now!
I am curious though, if I had used the older driver when it first came out,
would Compass have rendered an incorrect plot then as well? Or, did the
driver become inaccurate for Compass over time? Perhaps installing a new
driver reset a setting that was set wrong?
Knowing that one's computer can influence the plot of Compass, what is the
easiest and quickest way for one ensure Compass is plotting correctly? To
periodically open your cross.plt file and measure the lengths?
Thanks,
Ryan
--- In [email protected], "RyanY" wrote:
Larry,
Unfortunately, I have made a huge error, which has resulted in a waste of
your guys' time. Fortunately, the Cave Viewer is now correctly rendering
plots! I told you I had the latest driver, but this was not true. You see,
I own a Compaq Presario v2010us and traditionally I've always went to the
computer manufacturer website to look for the latest drivers, which is what
I did this time. The website claimed it was the most up to date driver, but
it wasn't. While looking for the umpteenth time at my graphics properties,
I noticed Intel has an option to go to their website to check for drivers.
Desperate, I gave this a try and sure enough, the Compaq website had lied
to me. After downloading, installing and rebooting, the Compass plotter
works fine now.
I tried all your suggestions from your previous two postings before
installing the new driver including using your small program, and if you
are still interested in my findings I can elaborate. It's probably not
worth mentioning, but after installing the driver, the crosses are
occurring at different inch origins (pixel mode - 0,0 ; other modes -
8.2,8.2) and the 0.01 inch mode is as thick as the pixel mode. But as long
as the Compass Cave Viewer is working correctly this doesn't concern me.
I apologize for wasting your time, and thank you for all the help you've
given. I've definitely learned some things about Compass (and the
reliability of Compaq support) that I never knew before. Kudos to you Larry
for your excellent program and phenomenal support. Thanks also to this
wonderful community of Compass users.
Ryan
--- In [email protected], "Larry Fish" wrote:
Ryan,
I wrote a small program to test the different modes that are available
on a
graphics card. It may give us more information about your problem. I
have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your
hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the
screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines
will
be thicker in "Pixel" mode and the cross might jump by one pixel
because of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected] [mailto:
[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders
incorrect
angles
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that
if it
were an issue with an incorrect screen setting, the PDF plot should
also
show incorrect angles, but as far as I can tell the PDF plot is
entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap
screen
image from their computer of the same survey data the plot looks
perfect to
me, but when I make a bitmap screen image with my computer the plot
remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal)
and is
set to 1280x768 pixels. I believe with these settings, everything is
1:1 (no
distortion). Finally, it appears that the monitor adjustment is such
that
there is no black space between or extension beyond the screen borders
(the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard
with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme
Graphics 2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S
lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio).
Of
course, the PDF plot lengths measured to be equal. I used the distance
tool
and the program told me they were the same lengths, so Compass thinks
it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the
moment I
am of the opinion that Compass may have rendering issues with my
particular
resolution.
Does anyone else have the newest Compass version and have a WXGA
monitor
that can confirm or deny my theory? It would also be interesting to see
if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be
solved
so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots
from
the
Compass viewer. I intended to get into my graphics card controls to
see if
I could correct it. As it happened, I first updated my ATI graphic
card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into
Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders
incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an
"aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can
understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display
a
file
on the screen, so that rules out a mathematical or graphics error
inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed
explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact
number
of
Pixels per inch in the left-right and up-down direction. As a result,
when
you print a line that is supposed to be 1-inch long, it will be
1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution
are
based on a several extraneous factors. For example, Windows assumes
the
screen has a certain resolution. Typically, it is 96 Dots Per Inch
(DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't
really
mean that a line 96 dots long will really be 1-inch long. The actual
size
depends on the Resolution of the monitor, the size of the monitor,
the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches
wide.
If
your video card is set to a resolution of 1024 by 768, there will be
1024
pixels across the 16 inch width of the screen. That means that the
actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long
on
the
screen. If you change the video mode to 2048 by 1536, the actual
screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch
long. The
important thing to understand is that, unlike the printer, the
monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the
ratio
of
the width to the height is 4 to 3. Newer, high def monitors have
aspect
ratios of 16 to 9. Some of the very newest monitors have aspect
ratios of
16
to 10.
When you setup your computer for a particular monitor, you must
select a
screen resolution that matches your monitor. If you don't, the pixels
will
be stretched or squashed horizontally or vertically. If this happens,
the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem
you
are seeing.
The first thing I would try is to check the monitor's resolution and
make
sure your video card was set for a matching resolution. Some people
don't
like to set their video resolution to match the monitor because it
makes
the
text too small. If you are doing this, make sure the resolution you
choose
match the aspect ratio of your monitor. For example, if your monitor
has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and
1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to
test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of
the
image on the monitor using the monitor's own controls. For example,
you
might prefer to have an 1/4 inch border on the right and left side of
the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the
survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure
Distance/Angles"
option to measure the angles of each shot in the cave. The tool
allows you
to lock a cursor on each survey station ("Nearest Station" button) so
the
measurements are extremely accurate. When I do this, the angles
exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it
is NOT
measuring it visually, so it ignores all kinds of resolution, scaling
and
aspect ratio issues.
This is something you can do too and it would give us more
information to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You
should
download it and double click on it. If Compass is installed properly,
the
Viewer will automatically open the file. The image will display a
cross
with
equal length arms. (Be sure to expand the Viewer to full screen and
zoom
out
a bit so you can see the ends of the cross arms.) If you measure the
cross
arms on the screen and they aren't equal, then the problem is
probably
caused because the video resolution settings don't match your
monitor.
It would also be useful to know more about your computer. What type
of
monitor and video card are you using, what resolution your are
running and
what operating system have you got?
Larry
________________________________________
From: [email protected] [mailto:
[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
Hi All! I apologize if I'm asking something that's already been
asked, but
I've searched both this forum and the archives and couldn't find a
similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave
he's
been
working on. He recently bought a Pocket PC to sketch the cave with
using
PocketTopo. I offered to convert his Compass files into PocketTopo
files.
When I did, the PocketTopo plot looked a little different. I
overlayed the
Compass plot with the PocketTopo plot, and they definitely were
different.
I
assumed it was an error of mine during file manipulation. I decided
to
determine the source of the error, so I made brand new survey files
in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my
transparent
cave survey protractor and began measuring. Lo and behold, the
Compass
plot
had some angles that were incorrect! I verified that I was in the
plan
view,
so it's not that I was looking at the plot from a skewed view.
Exporting
to
bitmap still gives me the wrong angles. What's interesting is that
when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of
software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 13:02:14 -0600 From: "Larry Fish" Subject: RE: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles Ryan, I apologize for wasting your time, and thank you for all the help you've given. Don't worry about it. It was not a waste of time. One of the nice things about the Users Group is that other people can read about your problems and learn from them. In the future your discoveries may save someone a lot of time and effort. At the same time, I now have a test tool that I can send out to people so they can test to see if their graphics drivers are handling the mapping modes correctly. Larry
Date: Sat, 19 Mar 2011 13:29:15 -0600 From: "Larry Fish" Subject: RE: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles Ryan, You might want to post something on the Cave Chat page letting people know that you solved the problem. The information would help somebody who has a similar problem. Larry
Date: Sat, 19 Mar 2011 13:32:29 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Re: Cave Viewer inconsistently renders incorrect angles
Ryan,
I am not quite sure how downloading a later driver solved
the problem, but I am glad Compass is performing perfectly
now!
The driver is responsible for converting Windows commands into actions. For
example, if I tell Windows to draw a line from the coordinates (0,0) to
(1000,0) it will draw a horizontal line 1,000 pixels long. The driver is
responsible lighting all the pixels between those two points.
If I tell Windows to go into "Inch" mode, that 1000 number is now
interpreted to be 1-inch and the Driver has to figure out how many pixels it
has to light up to equal 1-inch. On my computer, it turns out that it has to
light up 72 pixels to equal one inch.
There are lots of things that go into that calculation. And the calculation
may be different in the Horizontal direction than in the Vertical direction.
If the driver software gets any of those calculations wrong, the line will
have an error in it. Since driver software can be very complicated, it is
not unusual to have bugs in the software that the programmer didn't notice.
That is apparently what happened here.
I am curious though, if I had used the older driver when it
first came out, would Compass have rendered an incorrect
plot then as well? Or, did the driver become inaccurate for
Compass over time? Perhaps installing a new driver reset a
setting that was set wrong?
It is possible that the problem was caused by some kind of "Setting Error"
that occurred after the older driver was originally installed, but I doubt
it. It is most likely that the error was there from the beginning. You
either didn't notice it in other applications or Compass was the first
program to use the driver in this particular video mode.
Knowing that one's computer can influence the plot of
Compass, what is the easiest and quickest way for one ensure
Compass is plotting correctly? To periodically open your
cross.plt file and measure the lengths?
This kind of problem is relatively rare. In all the years of working with
Compass, this is the first time I've seen anything similar. The "Cross.plt"
file is a good way to test for these kinds of problems. I have several
similar files that I use to verify that the draw/printing routines are
working properly.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Saturday, March 19, 2011 9:27 AM
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
All,
I am not quite sure how downloading a later driver solved the problem, but I
am glad Compass is performing perfectly now!
I am curious though, if I had used the older driver when it first came out,
would Compass have rendered an incorrect plot then as well? Or, did the
driver become inaccurate for Compass over time? Perhaps installing a new
driver reset a setting that was set wrong?
Knowing that one's computer can influence the plot of Compass, what is the
easiest and quickest way for one ensure Compass is plotting correctly? To
periodically open your cross.plt file and measure the lengths?
Thanks,
Ryan
Larry,
Unfortunately, I have made a huge error, which has resulted in a waste of
your guys' time. Fortunately, the Cave Viewer is now correctly rendering
plots! I told you I had the latest driver, but this was not true. You see, I
own a Compaq Presario v2010us and traditionally I've always went to the
computer manufacturer website to look for the latest drivers, which is what
I did this time. The website claimed it was the most up to date driver, but
it wasn't. While looking for the umpteenth time at my graphics properties, I
noticed Intel has an option to go to their website to check for drivers.
Desperate, I gave this a try and sure enough, the Compaq website had lied to
me. After downloading, installing and rebooting, the Compass plotter works
fine now.
I tried all your suggestions from your previous two postings before
installing the new driver including using your small program, and if you are
still interested in my findings I can elaborate. It's probably not worth
mentioning, but after installing the driver, the crosses are occurring at
different inch origins (pixel mode - 0,0 ; other modes - 8.2,8.2) and the
0.01 inch mode is as thick as the pixel mode. But as long as the Compass
Cave Viewer is working correctly this doesn't concern me.
I apologize for wasting your time, and thank you for all the help you've
given. I've definitely learned some things about Compass (and the
reliability of Compaq support) that I never knew before. Kudos to you Larry
for your excellent program and phenomenal support. Thanks also to this
wonderful community of Compass users.
Ryan
--- In [email protected], "Larry Fish" wrote:
Ryan,
I wrote a small program to test the different modes that are available
on a
graphics card. It may give us more information about your problem. I
have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your
hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the
screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines
will
be thicker in "Pixel" mode and the cross might jump by one pixel because
of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected]
[mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders
incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if
it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap
screen
image from their computer of the same survey data the plot looks perfect
to
me, but when I make a bitmap screen image with my computer the plot
remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal)
and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1
(no
distortion). Finally, it appears that the monitor adjustment is such
that
there is no black space between or extension beyond the screen borders
(the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard
with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics
2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance
tool
and the program told me they were the same lengths, so Compass thinks
it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the
moment I
am of the opinion that Compass may have rendering issues with my
particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see
if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be
solved
so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots
from
the
Compass viewer. I intended to get into my graphics card controls to
see if
I could correct it. As it happened, I first updated my ATI graphic
card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into
Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders
incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an
"aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can
understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display
a
file
on the screen, so that rules out a mathematical or graphics error
inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed
explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact
number
of
Pixels per inch in the left-right and up-down direction. As a result,
when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution
are
based on a several extraneous factors. For example, Windows assumes
the
screen has a certain resolution. Typically, it is 96 Dots Per Inch
(DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't
really
mean that a line 96 dots long will really be 1-inch long. The actual
size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches
wide.
If
your video card is set to a resolution of 1024 by 768, there will be
1024
pixels across the 16 inch width of the screen. That means that the
actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long
on
the
screen. If you change the video mode to 2048 by 1536, the actual
screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long.
The
important thing to understand is that, unlike the printer, the
monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the
ratio
of
the width to the height is 4 to 3. Newer, high def monitors have
aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios
of
16
to 10.
When you setup your computer for a particular monitor, you must select
a
screen resolution that matches your monitor. If you don't, the pixels
will
be stretched or squashed horizontally or vertically. If this happens,
the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem
you
are seeing.
The first thing I would try is to check the monitor's resolution and
make
sure your video card was set for a matching resolution. Some people
don't
like to set their video resolution to match the monitor because it
makes
the
text too small. If you are doing this, make sure the resolution you
choose
match the aspect ratio of your monitor. For example, if your monitor
has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and
1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to
test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of
the
image on the monitor using the monitor's own controls. For example,
you
might prefer to have an 1/4 inch border on the right and left side of
the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the
survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure
Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows
you
to lock a cursor on each survey station ("Nearest Station" button) so
the
measurements are extremely accurate. When I do this, the angles
exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is
NOT
measuring it visually, so it ignores all kinds of resolution, scaling
and
aspect ratio issues.
This is something you can do too and it would give us more information
to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You
should
download it and double click on it. If Compass is installed properly,
the
Viewer will automatically open the file. The image will display a
cross
with
equal length arms. (Be sure to expand the Viewer to full screen and
zoom
out
a bit so you can see the ends of the cross arms.) If you measure the
cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running
and
what operating system have you got?
Larry
________________________________________
From: [email protected]
[mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked,
but
I've searched both this forum and the archives and couldn't find a
similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with
using
PocketTopo. I offered to convert his Compass files into PocketTopo
files.
When I did, the PocketTopo plot looked a little different. I overlayed
the
Compass plot with the PocketTopo plot, and they definitely were
different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my
transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view.
Exporting
to
bitmap still gives me the wrong angles. What's interesting is that
when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of
software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sat, 19 Mar 2011 19:40:31 -0000
From: "RyanY"
Subject: Re: Cave Viewer inconsistently renders incorrect angles
Thanks Larry, this pretty much answers every remaining question I had. Looking very forward to using this program more in the future!
I've already let Cavechat know that it was as simple as downloading the latest driver and that it wasn't any sort of software bug. Now that you mention it though, I'll go ahead and provide a link to the specific discussion so that they can more easily find the details.
Happy mapping!
Ryan,
I am not quite sure how downloading a later driver solved
the problem, but I am glad Compass is performing perfectly
now!
The driver is responsible for converting Windows commands into actions. For
example, if I tell Windows to draw a line from the coordinates (0,0) to
(1000,0) it will draw a horizontal line 1,000 pixels long. The driver is
responsible lighting all the pixels between those two points.
If I tell Windows to go into "Inch" mode, that 1000 number is now
interpreted to be 1-inch and the Driver has to figure out how many pixels it
has to light up to equal 1-inch. On my computer, it turns out that it has to
light up 72 pixels to equal one inch.
There are lots of things that go into that calculation. And the calculation
may be different in the Horizontal direction than in the Vertical direction.
If the driver software gets any of those calculations wrong, the line will
have an error in it. Since driver software can be very complicated, it is
not unusual to have bugs in the software that the programmer didn't notice.
That is apparently what happened here.
I am curious though, if I had used the older driver when it
first came out, would Compass have rendered an incorrect
plot then as well? Or, did the driver become inaccurate for
Compass over time? Perhaps installing a new driver reset a
setting that was set wrong?
It is possible that the problem was caused by some kind of "Setting Error"
that occurred after the older driver was originally installed, but I doubt
it. It is most likely that the error was there from the beginning. You
either didn't notice it in other applications or Compass was the first
program to use the driver in this particular video mode.
Knowing that one's computer can influence the plot of
Compass, what is the easiest and quickest way for one ensure
Compass is plotting correctly? To periodically open your
cross.plt file and measure the lengths?
This kind of problem is relatively rare. In all the years of working with
Compass, this is the first time I've seen anything similar. The "Cross.plt"
file is a good way to test for these kinds of problems. I have several
similar files that I use to verify that the draw/printing routines are
working properly.
Larry
________________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of RyanY
Sent: Saturday, March 19, 2011 9:27 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders incorrect
angles
�
All,
I am not quite sure how downloading a later driver solved the problem, but I
am glad Compass is performing perfectly now!
I am curious though, if I had used the older driver when it first came out,
would Compass have rendered an incorrect plot then as well? Or, did the
driver become inaccurate for Compass over time? Perhaps installing a new
driver reset a setting that was set wrong?
Knowing that one's computer can influence the plot of Compass, what is the
easiest and quickest way for one ensure Compass is plotting correctly? To
periodically open your cross.plt file and measure the lengths?
Thanks,
Ryan
--- In [email protected], "RyanY" wrote:
Larry,
Unfortunately, I have made a huge error, which has resulted in a waste of
your guys' time. Fortunately, the Cave Viewer is now correctly rendering
plots! I told you I had the latest driver, but this was not true. You see, I
own a Compaq Presario v2010us and traditionally I've always went to the
computer manufacturer website to look for the latest drivers, which is what
I did this time. The website claimed it was the most up to date driver, but
it wasn't. While looking for the umpteenth time at my graphics properties, I
noticed Intel has an option to go to their website to check for drivers.
Desperate, I gave this a try and sure enough, the Compaq website had lied to
me. After downloading, installing and rebooting, the Compass plotter works
fine now.
I tried all your suggestions from your previous two postings before
installing the new driver including using your small program, and if you are
still interested in my findings I can elaborate. It's probably not worth
mentioning, but after installing the driver, the crosses are occurring at
different inch origins (pixel mode - 0,0 ; other modes - 8.2,8.2) and the
0.01 inch mode is as thick as the pixel mode. But as long as the Compass
Cave Viewer is working correctly this doesn't concern me.
I apologize for wasting your time, and thank you for all the help you've
given. I've definitely learned some things about Compass (and the
reliability of Compaq support) that I never knew before. Kudos to you Larry
for your excellent program and phenomenal support. Thanks also to this
wonderful community of Compass users.
Ryan
--- In [email protected], "Larry Fish" wrote:
Ryan,
I wrote a small program to test the different modes that are available
on a
graphics card. It may give us more information about your problem. I
have
uploaded the file to the Compass Group site and it has the name:
MapModeTest.zip
After you've downloaded the file, unzip it and put it somewhere on your
hard
drive. Double click on the file. The program shows a list of Windows
"mapping modes." Pressing the "Cross" button will draw a cross on the
screen
using the mapping mode you've selected. The cross should be in the same
place and the same size no matter what mode it is drawn in. The lines
will
be thicker in "Pixel" mode and the cross might jump by one pixel because
of
rounding errors. Otherwise they should be identical.
Let me know what you find out.
Larry
________________________________________
From: [email protected]
[mailto:[email protected]]
On Behalf Of RyanY
Sent: Friday, March 18, 2011 8:32 AM
To: [email protected]
Subject: [compass-users] Re: Cave Viewer inconsistently renders
incorrect
angles
�
Larry et al,
Thank you for the quick response!
I neglected to mention that when I "printed" the cave plot, I actually
printed to PDF (via CutePDF), not on paper. I would have thought that if
it
were an issue with an incorrect screen setting, the PDF plot should also
show incorrect angles, but as far as I can tell the PDF plot is entirely
accurate.
Another thing worth mentioning is that when someone sends me a bitmap
screen
image from their computer of the same survey data the plot looks perfect
to
me, but when I make a bitmap screen image with my computer the plot
remains
incorrect.
Admittedly, my screen is of oddball proportions (WXGA), but I am a
resolution "purist" and always make sure to have my screen set at the
maximum, undistorted resolution possible. My screen is 14" (diagonal)
and is
set to 1280x768 pixels. I believe with these settings, everything is 1:1
(no
distortion). Finally, it appears that the monitor adjustment is such
that
there is no black space between or extension beyond the screen borders
(the
screen has been automatically adjusted).
My graphics card is actually the integrated rubbish that comes standard
with
most low end computers (Intel 82852/82855 GM/GME, Intel Extreme Graphics
2),
but it usually does the job just fine. My operating system is XP Home
Edition, 2002, SP3. I've verified that my driver is the most up to date
there is.
Perhaps unsurprisingly, I opened your cross.plt file and the N-S lengths
were indeed shorter than the W-E lengths (approximately a 4:5 ratio). Of
course, the PDF plot lengths measured to be equal. I used the distance
tool
and the program told me they were the same lengths, so Compass thinks
it's
showing me an accurate plot.
I am not entirely sure what the cause of my problem is, but at the
moment I
am of the opinion that Compass may have rendering issues with my
particular
resolution.
Does anyone else have the newest Compass version and have a WXGA monitor
that can confirm or deny my theory? It would also be interesting to see
if
running an older Compass version on my computer would produce the same
distorted plot.
Thanks again for all your help guys, and I hope this problem can be
solved
so I can use this fantastic program with confidence!
--- In [email protected], "dwightliv@" wrote:
I ran into the aspect ratio problem when I was taking screen shots
from
the
Compass viewer. I intended to get into my graphics card controls to
see if
I could correct it. As it happened, I first updated my ATI graphic
card
driver, and that fixed the problem.
The easiest test is to create a couple of fake shots, both the same
length,
one north-south and the other east-west. Get a screen shot into
Photoshop
and compare the pixel length.
Dwight
Original Message:
-----------------
From: Larry Fish lfish@
Date: Fri, 18 Mar 2011 03:46:32 -0600
To: [email protected]
Subject: RE: [compass-users] Cave Viewer inconsistently renders
incorrect
angles
Ryan,
Thanks for your email and questions. I've down loaded your files and
images
and spent some time looking at them. I think this is probably an
"aspect
ratio" issue, but I don't have quite enough information to be sure.
I'm going to give you a fairly large "brain-dump" so you can
understand
the
issues and focus on certain details.
This statement gives me a clue:
What's interesting is that when I print the plot out, the
angles SUDDENLY BECOME CORRECT!
Compass uses exactly the same subroutines to print a file and display
a
file
on the screen, so that rules out a mathematical or graphics error
inside
Compass. The most likely cause is the display and how it is setup,
especially things that the "Aspect Ratio." Here is a detailed
explanation
of
the issues:
1. PRINTERS. When you print a cave plot, the printer uses an exact
number
of
Pixels per inch in the left-right and up-down direction. As a result,
when
you print a line that is supposed to be 1-inch long, it will be 1-inch
long
on the paper. That's why your print out has the correct angles.
2. SCREEN SCALING. The screen is different. Its scale and resolution
are
based on a several extraneous factors. For example, Windows assumes
the
screen has a certain resolution. Typically, it is 96 Dots Per Inch
(DPI),
but it can be other values. If Windows is set to 96 DPI, it doesn't
really
mean that a line 96 dots long will really be 1-inch long. The actual
size
depends on the Resolution of the monitor, the size of the monitor, the
video
card settings and how the monitor is adjusted.
For example, let's say you have a monitor whose screen is 16 inches
wide.
If
your video card is set to a resolution of 1024 by 768, there will be
1024
pixels across the 16 inch width of the screen. That means that the
actual
scale is 64 dpi so a 96 pixel lines would actually be 1.5 inches long
on
the
screen. If you change the video mode to 2048 by 1536, the actual
screen
DPI
will now be 128 dpi and the 96 pixel line will only be 0.75-inch long.
The
important thing to understand is that, unlike the printer, the
monitors
scale and resolution depends of many factors and the computer can't
control
all of them.
3. ASPECT RATIO. The problem gets more complicated because different
monitors have different "aspect ratios." For example, on an old style
non-HiDef monitor the "Aspect Ratio" is 4 to 3. That means that the
ratio
of
the width to the height is 4 to 3. Newer, high def monitors have
aspect
ratios of 16 to 9. Some of the very newest monitors have aspect ratios
of
16
to 10.
When you setup your computer for a particular monitor, you must select
a
screen resolution that matches your monitor. If you don't, the pixels
will
be stretched or squashed horizontally or vertically. If this happens,
the
pixels won't be "Square" and the image will be distorted. The angles
between
lines will be distorted. This is the most likely cause of the problem
you
are seeing.
The first thing I would try is to check the monitor's resolution and
make
sure your video card was set for a matching resolution. Some people
don't
like to set their video resolution to match the monitor because it
makes
the
text too small. If you are doing this, make sure the resolution you
choose
match the aspect ratio of your monitor. For example, if your monitor
has a
4
by 3 aspect ratio, screen resolutions of 800x600, 1024x768 and
1600x1200
all
have a 4 to 3 aspect ratio. I've have posted file that you can use to
test
to see if the aspect ratio is wrong. More on this below.
4. MONITOR ADJUSTMENTS. On top of that, you can adjust the width of
the
image on the monitor using the monitor's own controls. For example,
you
might prefer to have an 1/4 inch border on the right and left side of
the
image to make sure you can the you are not cutting off part of the
display.
That small gap would change the Aspect Ratio enough to through the
survey
angles.
5. MEASURED ANGLES. I loaded the file you posted into Compass and
displayed
the file in the Viewer. I then used the "Tools - Measure
Distance/Angles"
option to measure the angles of each shot in the cave. The tool allows
you
to lock a cursor on each survey station ("Nearest Station" button) so
the
measurements are extremely accurate. When I do this, the angles
exactly
match the compass angles in the survey data.
This tools measures angles mathematically by calculating difference
between
the locations of each point on the screen. The key thing is that it is
NOT
measuring it visually, so it ignores all kinds of resolution, scaling
and
aspect ratio issues.
This is something you can do too and it would give us more information
to
isolate the cause. Let me know if you need help using the tool.
I have uploaded a file to the Compass Group called "Cross.plt." You
should
download it and double click on it. If Compass is installed properly,
the
Viewer will automatically open the file. The image will display a
cross
with
equal length arms. (Be sure to expand the Viewer to full screen and
zoom
out
a bit so you can see the ends of the cross arms.) If you measure the
cross
arms on the screen and they aren't equal, then the problem is probably
caused because the video resolution settings don't match your monitor.
It would also be useful to know more about your computer. What type of
monitor and video card are you using, what resolution your are running
and
what operating system have you got?
Larry
________________________________________
From: [email protected]
[mailto:[email protected]]
On Behalf Of RyanY
Sent: Thursday, March 17, 2011 2:27 PM
To: [email protected]
Subject: [compass-users] Cave Viewer inconsistently renders incorrect
angles
�
Hi All! I apologize if I'm asking something that's already been asked,
but
I've searched both this forum and the archives and couldn't find a
similar
problem.
My friend has over 6300 feet of survey data in Compass for a cave he's
been
working on. He recently bought a Pocket PC to sketch the cave with
using
PocketTopo. I offered to convert his Compass files into PocketTopo
files.
When I did, the PocketTopo plot looked a little different. I overlayed
the
Compass plot with the PocketTopo plot, and they definitely were
different.
I
assumed it was an error of mine during file manipulation. I decided to
determine the source of the error, so I made brand new survey files in
Compass and PocketTopo *using only front sights* and *the exact same
data*.
Somehow, the plots were STILL different! So, I pulled out my
transparent
cave survey protractor and began measuring. Lo and behold, the Compass
plot
had some angles that were incorrect! I verified that I was in the plan
view,
so it's not that I was looking at the plot from a skewed view.
Exporting
to
bitmap still gives me the wrong angles. What's interesting is that
when I
print the plot out, the angles SUDDENLY BECOME CORRECT!
Do you guys have any ideas?? I suppose this could be some sort of
incorrect
setting, but surely there isn't a setting to make the angles appear
different that what they should be? Could this be some sort of
software
bug?
I have version 5.11.1.1.165
Here are some pictures of the plots:
http://forums.caves.org/viewtopic.php?f=26&t=11762&p=101230#p101098
Thanks for any help you can give.
----------------------------------------------------------
mail2web - Check your email from the web at
http://link.mail2web.com/mail2web
Date: Sun, 20 Mar 2011 01:47:33 -0600
From: "Larry Fish"
Subject: New Version of the SVG Exporter
Hi Everyone,
I wanted to announce that I've just finished up a new version of the SVG
Exporter. Most of the changes were based on compatibility issues that I
discovered working with Illustrator CS2. I also added lots of options that
allow you to repair and deal with file corruption and compatibility issues.
Here is a description of the problems and changes I made:
1. ILLUSTRATOR SPLITS AND MOVES LAYERS. Illustrator sometimes takes
Compass/Walls tags and splits them up or embeds that at deeper level. This
would cause the SVG Exporter to miss part or all of the tag and not process
them. The result was that certain parts of the map would be adjusted
properly and certain parts would not.
To solve this problem, the SVG Exporter is now smart enough to re-assemble
tags. It also searches to the bottom of all layers, not just the top layer,
for Compass/Walls tags.
2. ILLUSTRATOR REMOVES INKSCAPE TAGS. Adobe Illustrator sometimes removes
key "Inkscape" tags from the SVG file. This wouldn't be so bad except that
it leaves some other Inkscape tags in place. When that happens, the Inkscape
tags that are left in the file cannot be tied to a reference and the file
cannot be parsed correctly. That means that even Illustrator will get an
"Invalid File" error and tell you to "validate" the file before trying to
load it again. This is particularly disconcerting because, Illustrator will
load the file perfectly one time and then not the next. Even the Compass SVG
Exporter would get an error when it tried to load the SVG file for merging,
(although Inkscape would load it just fine.)
To solve this problem, the SVG Exporter now checks to make sure all the tags
are in place before it attempts to work with the file. To give the user
control over the tags, there is now a "Compatibility" option in the exporter
and merge/morph tools that allows you to select which tags are added or
removed from the file. There is a also a Repair Option that allows you fix
problems with the tags without doing any morphing or merging. These options
will not only add tags that you want in the file, it will also remove
unwanted tags if they are causing compatibility problems.
If you are working exclusively with Illustrator, you should set the SVG
Exporter to remove all Inkscape tags. It you want to use both Inkscape and
Illustrator on the file, you need to be aware that you will have to run the
file through the Repair Tool in the Exporter periodically to reinsert all
the Inkscape tags.
4. ILLUSTRATOR RENAMES LAYERS. Adobe Illustrator also renames some of the
key layers on a whim. For example, it might rename "w2d_vectors" to
"w2d_vectors_4_." When this happens, Compass cannot find the key layers. To
solve the problem, the program looks for key layers whose names have been
changed and corrects them.
5. PIXEL RESOLUTION. The SVG Exporter uses "Points" as the unit of measure
when specifying Paper Sizes and Margins in the SVG file. Since Points
translate directly into Inches or Millimeters, SVG files exported from
Compass will always show the correct paper size and margin.
On the other hand, a drawing generated by other programs or directly from
Inkscape or Illustrator will specify the paper size in pixels. Since there
is no universal standard for converting pixels to inches or millimeters, the
pixel values are hard to interpret.
For example, Adobe Illustrator defaults to using a scale 72 dots per inch
when converting pixels to a real-world unit such as inches. (This makes one
pixel equal to one "Point.") Inkscape, on the other hand, defaults to 90
dots per inch. If the SVG Exporter reads a document using the wrong number
of DPIs, the paper and margin sizes will be wrong. For this reason, there is
now a compatibility option that allows you to choose between Inkscape and
Illustrator modes.
If you want use both Inkscape and Illustrator on the same map, you should
either set Inkscape to 72 dpi or Illustrator to 90 dpi.
SETTING INKSCAPE. To set Inkscape to a different resolution like 72 dpi,
choose the "File - Inkscape Options" from the menu bar. Next, select the
"Import/Export" option from the panel on the left. Finally, set "Default
export resolution."
SETTING ILLUSTRATOR. To set Illustrator to a different resolution such as 90
dpi, select the "Effect - Document Raster Effect Settings," from the menu
bar. In "Resolution Box," select the "Other" option and then enter the
resolution in the edit box to the right.
6. ORGANIZATION. When Compass processes an SVG file, it can adjust one item
at a time or it can adjust a group of items all at once. Adjusting items as
a group is useful when you have several items that have to be adjusted
together. For example, you might have a Number with a Circle around it to
represent Ceiling Heights. Because they have slightly different starting
positions, if they were adjusted separately, the number would no longer be
centered in the circle.
You control which items are adjusted together by putting them in "unnamed
groups." Compass looks for Unnamed groups adjusts all the items in the group
at once instead of one-at-a-time.
This can also lead to problems if you are not careful how you group items.
For example, let's say you have two groups that are organized like this:
Line
Text
The "Line" inside "Group 2" could get adjusted twice and move twice as far
as you would expect. This is because the Exporter would adjust both Group-1
and Group-2 since they both Unnamed groups. Because the "Line" item is
inside both groups, the adjustment would be twice as much as necessary.
To solve this problem, Compass now only adjusts the top most un-named group.
This solves most of the problems with nested Groups, but it may not fix
every possible combination of groupings. For this reason, you probably want
to avoid having unnamed groups nested inside other unnamed groups.
7. ILLUSTRATOR EMBEDS BINARY DATA. When you save an SVG file from
Illustrator, Illustrator can save a "binary" copy of the Image inside the
SVG file. The Binary image is saved in a SVG group called "i:pgf" and is
actually just a copy of the standard "AI" file that Illustrator normally
uses. Saving this Binary data allows Illustrator to the load image faster
and save certain setting that SVG doesn't handle.
The Binary data causes a problem because Illustrator will ignore the rest of
the SVG file. If you have made any changes to the SVG part of the file,
Illustrator will ignore those changes and instead read the old, unchanged
binary data. In other words, Compass could make changes to the data and
Illustrator would ignore the changes.
To solve this problem, the SVG Exporter now has the option of removing the
Binary Data as a part of the Export or Merge/Morph process. It is also an
option when you Repair a file.
To avoid the problem in the first place, you should always disable the
"Preserve Illustrator Editing Capabilities" when you export an SVG file from
Illustrator. It is also a good idea to turn off all the following options
when you export a file:
Preserve Illustrator Editing Capabilities
Optimize for Adobe SVG Viewer
Include Adobe Graphic Server Data
Include Slicing Data
Use element for Text on Path
10. ILLUSTRATOR/INKSCAPE TAGS. Illustrator requires a special tag
(i:layer="yes") in each SVG group before it will be treated as a "layer." In
the past, you had to manually convert the groups to layers. The SVG Exporter
now adds the special tag if you have the "Illustrator Compatibility" option
enabled. Likewise, Inkscape requires its own special tag for a group to be
seen as a Layer. The Compatibility Option now allow you add or remove these
tags.
11. CLIPPING WINDOW. The SVG Exporter normally exports a clipping window
around your map. This tells Illustrator or Inkscape not to draw any portion
of the map that might fall outside the margins. Unfortunately, Adobe
Illustrator versions 10, CS1 and CS2 do not support this kind of a clipping
window. If AI10, CS1 or CS2 tries to read a file with the clipping window,
it would two errors. First, Illustrator would give the Error message
"Clipping will be lost on roundtrip to Tiny." Second, the drawing would not
appear on the paper window but would appear up and to the right of the
paper.
The SVG Exporter has always had the ability to not export the Clipping
Windows if the "10/CS1/CS2 Compatible" option was enabled. However, it could
not remove the Clipping Window if you accidentally put one in. Your only
choice was to manually delete it. The new version SVG Exporter will remove
the Clipping Window during the Repair or Merge/Morph operation if the option
is turned off.
The new version is on the Compass web site.
Larry Fish
Date: Sat, 06 Aug 2011 23:45:36 -0000 From: [email protected] Subject: use of 'pass' in LRUD Hello Larry I have been using compass for many years and recently another caver has inputed data into the program and not used 'pass' for passages at L and R but written in a distance as if the wall continued along the passage being surveyed. Question: will this make a difference to the program and how so? Thanks Bob
Date: Sat, 6 Aug 2011 18:26:56 -0600
From: "Larry Fish"
Subject: RE: [compass-users] use of 'pass' in LRUD
Hi Bob,
It shouldn't make much difference. When you enter the "pass" option,
depending on circumstances and settings, Compass may substitute the last
valid passage-wall distance and use it to draw the passage walls. What the
other caver is doing is very similar to what Compass does, except the caver
is better able to figure out what the correct distance should be. What
he/she is doing might actually improve the accuracy of the Compass display.
On the other hand, we don't enter enough Passage Wall information to
accurately display the true shape and size of the passage. To do it right,
we'd need more LRUDs along the length of a shot. It would also be better to
have more than just the four measurements - for example, eight passage
dimensions would be much better. Since we don't have more data, the 3D model
Compass constructs is just a rough approximation. As a result, entering or
not entering the "Pass" option won't make much difference.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of [email protected]
Sent: Saturday, August 06, 2011 5:46 PM
Subject: [compass-users] use of 'pass' in LRUD
Hello Larry
I have been using compass for many years and recently another caver has
inputed data into the program and not used 'pass' for passages at L and R
but written in a distance as if the wall continued along the passage being
surveyed.
Question: will this make a difference to the program and how so?
Thanks
Bob
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Sun, 07 Aug 2011 04:02:53 -0000 From: [email protected] Subject: Re: use of 'pass' in LRUD hello LArry thanks for the quick reply Bob Hi Bob, It shouldn't make much difference. When you enter the "pass" option, depending on circumstances and settings, Compass may substitute the last valid passage-wall distance and use it to draw the passage walls. What the other caver is doing is very similar to what Compass does, except the caver is better able to figure out what the correct distance should be. What he/she is doing might actually improve the accuracy of the Compass display. On the other hand, we don't enter enough Passage Wall information to accurately display the true shape and size of the passage. To do it right, we'd need more LRUDs along the length of a shot. It would also be better to have more than just the four measurements - for example, eight passage dimensions would be much better. Since we don't have more data, the 3D model Compass constructs is just a rough approximation. As a result, entering or not entering the "Pass" option won't make much difference. Larry _____ From: [email protected] [mailto:[email protected]] On Behalf Of rkershaw@... Sent: Saturday, August 06, 2011 5:46 PM To: [email protected] Subject: [compass-users] use of 'pass' in LRUD Hello Larry I have been using compass for many years and recently another caver has inputed data into the program and not used 'pass' for passages at L and R but written in a distance as if the wall continued along the passage being surveyed. Question: will this make a difference to the program and how so? Thanks Bob
Date: Tue, 13 Sep 2011 16:12:43 -0000 From: "[email protected]" Subject: Section displays Can anyone please inform if Compass has the facility to display extended sections ? I have seen and used the various projected sections but for a very curvy cave would prefer to show as extended instead. Thanks, Mark Tringham
Date: Tue, 13 Sep 2011 12:43:26 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Section displays
Mark,
Thanks for your question. I assume that when you say "extended sections,"
you mean unfolding a profile so you see every shot as it's full extended
length. Compass calls this an "Unfolded Profile." It requires that you
recompile the cave to do this. In the Project Manager, choose the
"Options-Settings" item from the menu bar, then select the "Profile" tab.
Enable the "Unfolded Profile" option and then set the various options to
your liking. (If you have any questions about how the options work, press
the "Help" button and look at the section about "Unfolded Profiles."
Finally, recompile the cave. In the Viewer, select profile mode. The Viewer
will now show the cave in as an extended or unfolded profile. The option is
only available for profiles.
Let me know if you have any questions.
Larry Fish
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of [email protected]
Sent: Tuesday, September 13, 2011 10:13 AM
Subject: [compass-users] Section displays
Can anyone please inform if Compass has the facility to display extended
sections ? I have seen and used the various projected sections but for a
very curvy cave would prefer to show as extended instead.
Thanks, Mark Tringham
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Sat, 17 Sep 2011 16:31:31 -0000 From: "Tony Canike" Subject: Layers in SVG round tripping I've been using Compass and Illustrator CS4 for a while, but have never used the SVG roundtripping feature. Just got another round of 2011 survey data from the team, and they added few more loops in 2011. The line plot creaked and groaned, so I thought it would be a good time to learn the SVG roundtripping process before I start drawing the 2011 updates to the map. I downloaded Larry's latest version of SVG Exporter, read his tutorial on adopting roundtripping, and think I understand what I need to do. I am at the stage where I need to organize my artwork into the SVG roundtrip-specific layers (e.g "w2d_*") and I have a couple questions about organizing the artwork into roundtrip-specific layers. 1. Can I create sub-sub-layers of the w2d sym and shp layers to keep my art organized how I like it? For example, can I define new layers "walls" and "ledges" under the "w2d_Walls_shp" layer? 2. Does anyone have a cheat sheet to help figure out what artwork to put into which w2d layer. For example, I think the leader lines for cross sections should be moved but not warped or rotated, so I think they need to go into a shp layer. But what works best for everyone? If I could save a day of trial and error it would really help! Thanks, Tony. PS 7073 I'm working on your map, honest :) !
Date: Sat, 17 Sep 2011 15:59:12 -0600
From: "Larry Fish"
Subject: RE: [compass-users] Layers in SVG round tripping
Hi Tony,
You can create sublayers under the standard layers and the SVG Exporter will
adjust all the sublayers the same as the standard layer.
The closest I can come to a "Cheat Sheet," is the "Layer Details" section of
the help file. You've probably already seen it, but in case you haven't, it
is under the "Concepts, Techniques and Details" section of the help file. At
the bottom is a list of all the standard layers and what they are used for.
There are some suggestions about what things could go in each layer.
Let me know if you have any more questions.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Tony Canike
Sent: Saturday, September 17, 2011 10:32 AM
Subject: [compass-users] Layers in SVG round tripping
I've been using Compass and Illustrator CS4 for a while, but have never used
the SVG roundtripping feature. Just got another round of 2011 survey data
from the team, and they added few more loops in 2011. The line plot creaked
and groaned, so I thought it would be a good time to learn the SVG
roundtripping process before I start drawing the 2011 updates to the map.
I downloaded Larry's latest version of SVG Exporter, read his tutorial on
adopting roundtripping, and think I understand what I need to do.
I am at the stage where I need to organize my artwork into the SVG
roundtrip-specific layers (e.g "w2d_*") and I have a couple questions about
organizing the artwork into roundtrip-specific layers.
1. Can I create sub-sub-layers of the w2d sym and shp layers to keep my art
organized how I like it? For example, can I define new layers "walls" and
"ledges" under the "w2d_Walls_shp" layer?
2. Does anyone have a cheat sheet to help figure out what artwork to put
into which w2d layer. For example, I think the leader lines for cross
sections should be moved but not warped or rotated, so I think they need to
go into a shp layer. But what works best for everyone? If I could save a day
of trial and error it would really help!
Thanks,
Tony.
PS 7073 I'm working on your map, honest :) !
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Wed, 2 Nov 2011 02:11:10 +0000 (GMT) From: Matt Cracknell Subject: SVG Exporter error Hi all, I am encountering an error when attempting to load an existing SVG map (created with SVG Exporter and edited with Inkscape) into the SVG Exporter. The error states "Device is not ready". I have had a quick look in the Compass group archives but have not seen any discussions for a similar error.� I have tried a few things like reinstalling the SVG Exporter, loading other PLT files and SVGs, mucking around with the compatability settings and even trying to repair the SVG. But alas the same error comes up. Any help would be greatly appreciated. Regards, Matt Cracknell (Ph) 0409 438 924 Alternative email [email protected] Hi all,I am encountering an error when attempting to load an existing SVG map (created with SVG Exporter and edited with Inkscape) into the SVG Exporter. The error states "Device is not ready". I have had a quick look in the Compass group archives but have not seen any discussions for a similar error. I have tried a few things like reinstalling the SVG Exporter, loading other PLT files and SVGs, mucking around with the compatability settings and even trying to repair the SVG. But alas the same error comes up.Any help would be greatly appreciated.Regards,Matt Cracknell(Ph) 0409 438 924Alternative [email protected]
Date: Wed, 2 Nov 2011 04:25:26 -0600
From: "Larry Fish"
Subject: RE: [compass-users] SVG Exporter error
Hi Matt,
I have not seen this error before. I will need to look at the data to
diagnose the problem. I will send you a private message with an email
address where you can send the data.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Matt Cracknell
Sent: Tuesday, November 01, 2011 8:11 PM
Subject: [compass-users] SVG Exporter error
Hi all,
I am encountering an error when attempting to load an existing SVG map
(created with SVG Exporter and edited with Inkscape) into the SVG Exporter.
The error states "Device is not ready".
I have had a quick look in the Compass group archives but have not seen any
discussions for a similar error.
I have tried a few things like reinstalling the SVG Exporter, loading other
PLT files and SVGs, mucking around with the compatability settings and even
trying to repair the SVG. But alas the same error comes up.
Any help would be greatly appreciated.
Regards,
Matt Cracknell
(Ph) 0409 438 924
Alternative email
[email protected]
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Wed, 2 Nov 2011 11:09:36 +0000 (GMT) From: Matt Cracknell Subject: Re: [compass-users] SVG Exporter error Larry, Thanks for that. A� Matt Cracknell (Ph) 0409 438 924 Alternative email [email protected] ________________________________ From: Larry Fish Sent: Wednesday, 2 November 2011, 21:25 Subject: RE: [compass-users] SVG Exporter error A� Hi Matt, A� I have not seen this error before. I will need to look at the data to diagnose the problem. I will send you a private message with an email address where you can send the data. A� Larry A� A� ________________________________ From:[email protected] [mailto:[email protected]] On Behalf Of Matt Cracknell Sent: Tuesday, November 01, 2011 8:11 PM Subject: [compass-users] SVG Exporter error A� A� Hi all, A� I am encountering an error when attempting to load an existing SVG map (created with SVG Exporter and edited with Inkscape) into the SVG Exporter. The error states "Device is not ready". A� I have had a quick look in the Compass group archives but have not seen any discussions for a similar error.A� A� I have tried a few things like reinstalling the SVG Exporter, loading other PLT files and SVGs, mucking around with the compatability settings and even trying to repair the SVG. But alas the same error comes up. A� Any help would be greatly appreciated. A� Regards, A� Matt Cracknell (Ph) 0409 438 924 Alternative email [email protected] Larry,Thanks for that. Matt Cracknell(Ph) 0409 438 924Alternative [email protected]: Larry Fish <[email protected]>To: [email protected]: Wednesday, 2 November 2011, 21:25Subject: RE: [compass-users] SVG Exporter error Hi Matt, I have not seen this error before. I will need to look at the data to diagnose the problem. I will send you a private message with an email address where you can send the data. Larry From: [email protected] [mailto:[email protected]] On Behalf Of Matt Cracknell Sent: Tuesday, November 01, 2011 8:11 PM To: [email protected] Subject: [compass-users] SVG Exporter error Hi all, I am encountering an error when attempting to load an existing SVG map (created with SVG Exporter and edited with Inkscape) into the SVG Exporter. The error states "Device is not ready". I have had a quick look in the Compass group archives but have not seen any discussions for a similar error. I have tried a few things like reinstalling the SVG Exporter, loading other PLT files and SVGs, mucking around with the compatability settings and even trying to repair the SVG. But alas the same error comes up. Any help would be greatly appreciated. Regards, Matt Cracknell (Ph) 0409 438 924 Alternative email [email protected]
Date: Fri, 4 Nov 2011 03:56:40 -0600
From: "Larry Fish"
Subject: RE: [compass-users] SVG Exporter error
Matt,
Did you get the private email I sent? If not, I will retransmit.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Matt Cracknell
Sent: Wednesday, November 02, 2011 5:10 AM
Subject: Re: [compass-users] SVG Exporter error
Larry,
Thanks for that.
Matt Cracknell
(Ph) 0409 438 924
Alternative email
[email protected]
_____
From: Larry Fish
Sent: Wednesday, 2 November 2011, 21:25
Subject: RE: [compass-users] SVG Exporter error
Hi Matt,
I have not seen this error before. I will need to look at the data to
diagnose the problem. I will send you a private message with an email
address where you can send the data.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Matt Cracknell
Sent: Tuesday, November 01, 2011 8:11 PM
Subject: [compass-users] SVG Exporter error
Hi all,
I am encountering an error when attempting to load an existing SVG map
(created with SVG Exporter and edited with Inkscape) into the SVG Exporter.
The error states "Device is not ready".
I have had a quick look in the Compass group archives but have not seen any
discussions for a similar error.
I have tried a few things like reinstalling the SVG Exporter, loading other
PLT files and SVGs, mucking around with the compatability settings and even
trying to repair the SVG. But alas the same error comes up.
Any help would be greatly appreciated.
Regards,
Matt Cracknell
(Ph) 0409 438 924
Alternative email
[email protected]
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
Date: Sun, 6 Nov 2011 22:25:05 +0000 (GMT)
From: Matt Cracknell
Subject: Re: SVG Exporter error - FIXED!
Larry,
After a bit of searching I fixed the SVG Exporter error ("Device is not ready") that occured when attempting to load SVG files for round tripping. It appears that there was a bug in a Windows 7-64 bit MS XML handling code. This is fixed by getting the latest Microsoft MSXML 4.0 Service Pack 3 (Microsoft XML Core Services).
This can be downloaded at
http://www.microsoft.com/download/en/details.aspx?id=15697
Cheers,
Matt Cracknell
(Ph) 0409 438 924
Alternative email
[email protected]
________________________________
From: Larry Fish
Sent: Sunday, 6 November 2011, 19:30
Subject: RE: SVG Exporter error
Matt,
A�
I checked the files you sent and I can't
get them to fail. I tested them on both Windows XP and Window 7. I don't have
64-bit Win7 installed, so I couldn't test that version, but I kind of doubt
that is the issue. Here are some thoughts about the problem:
A�
1. The error message "Device is not
ready" is an operating system error that usually indicates a problem with
a disk drive. Searching the internet, I see lots of people getting that error
message with USB hard drives and Windows 7. That makes me wonder if the files
might be on some kind of USB drive or thumb drive and that is causing the
problem. You might try copying the files to different drive or folder and
re-testing the operation.
A�
2. I'm not sure I'm doing exactly the same
steps that you are doing. To test the problem, I loaded the
"Newdegate_Sept_2011.PLT" in the SVG Exporter then load the
"Newdegate_Sept_2011_example.svg." Perhaps you doing this in a
different order or you are working with different files. It would be very
helpful for you to explain exactly what steps you are performing and exactly
where the error occurs. Sometimes very small differences in procedure can make
a big difference. For example, one problem I tracked down a few years only
happened on a floppy disk, not on the hard drive. I couldn't duplicate the
problem no matter what I did, until the user happened mentioned that he had the
data on a floppy. Since I rarely used floppies, it never occurred to me to test
the problem on a floppy disk.
A�
3. Another possibility is "Regional
Settings." Since you appear to be in the UK , I wonder if you might have
computer setup in a special way. If so, it might explain why I cannot reproduce
the problem.
A�
For example, in much of Europe ,
they use a comma "," as the "decimal point" and a period
"." as the "thousands separator." In the US it is the
opposite. I have had several bugs in Compass that were caused by those differences.
If your computer was setup in a way that I haven�?Tt anticipated, that
could explain why I can�?Tt duplicate the problem.
A�
You can get to the Regional Setting under
Win7, by going to the Start Menu and selecting the "Control Panel"
option. In the Control Panel, select "Region and Language." Finally,
on the "Formats" page, select the "Advanced" option. There
are a number of different settings here. The ones I have had the most trouble
with are the number format settings.
A�
Let me know what you think.
A�
Larry
Larry,After a bit of searching I fixed the SVG Exporter error ("Device is not ready") that occured when attempting to load SVG files for round tripping. It appears that there was a bug in a Windows 7-64 bit MS XML handling code. This is fixed by getting the latest Microsoft MSXML 4.0 Service Pack 3 (Microsoft XML Core Services).This can be downloaded athttp://www.microsoft.com/download/en/details.aspx?id=15697Cheers,Matt Cracknell(Ph) 0409 438
924Alternative [email protected]: Larry Fish <[email protected]>To: 'Matt Cracknell' <[email protected]>Sent: Sunday, 6 November 2011, 19:30Subject: RE: SVG Exporter error
Date: Sun, 6 Nov 2011 15:50:51 -0700
From: "Larry Fish"
Subject: RE: [compass-users] Re: SVG Exporter error - FIXED!
Matt,
Thank you very much. Great find! That saves me a lot of head scratching. I'm
not sure I would have found it considering I don't have Win7-64 bits.
Larry
_____
From: [email protected] [mailto:[email protected]]
On Behalf Of Matt Cracknell
Sent: Sunday, November 06, 2011 3:25 PM
Subject: [compass-users] Re: SVG Exporter error - FIXED!
Larry,
After a bit of searching I fixed the SVG Exporter error ("Device is not
ready") that occured when attempting to load SVG files for round tripping.
It appears that there was a bug in a Windows 7-64 bit MS XML handling code.
This is fixed by getting the latest Microsoft MSXML 4.0 Service Pack 3
(Microsoft XML Core Services).
This can be downloaded at
http://www.microsoft.com/download/en/details.aspx?id697
Cheers,
Matt Cracknell
(Ph) 0409 438 924
Alternative email
[email protected]
_____
From: Larry Fish
Sent: Sunday, 6 November 2011, 19:30
Subject: RE: SVG Exporter error
Matt,
I checked the files you sent and I can't get them to fail. I tested them on
both Windows XP and Window 7. I don't have 64-bit Win7 installed, so I
couldn't test that version, but I kind of doubt that is the issue. Here are
some thoughts about the problem:
1. The error message "Device is not ready" is an operating system error that
usually indicates a problem with a disk drive. Searching the internet, I see
lots of people getting that error message with USB hard drives and Windows
7. That makes me wonder if the files might be on some kind of USB drive or
thumb drive and that is causing the problem. You might try copying the files
to different drive or folder and re-testing the operation.
2. I'm not sure I'm doing exactly the same steps that you are doing. To test
the problem, I loaded the "Newdegate_Sept_2011.PLT" in the SVG Exporter then
load the "Newdegate_Sept_2011_example.svg." Perhaps you doing this in a
different order or you are working with different files. It would be very
helpful for you to explain exactly what steps you are performing and exactly
where the error occurs. Sometimes very small differences in procedure can
make a big difference. For example, one problem I tracked down a few years
only happened on a floppy disk, not on the hard drive. I couldn't duplicate
the problem no matter what I did, until the user happened mentioned that he
had the data on a floppy. Since I rarely used floppies, it never occurred to
me to test the problem on a floppy disk.
3. Another possibility is "Regional Settings." Since you appear to be in the
UK , I wonder if you might have computer setup in a special way. If so, it
might explain why I cannot reproduce the problem.
For example, in much of Europe , they use a comma "," as the "decimal point"
and a period "." as the "thousands separator." In the US it is the opposite.
I have had several bugs in Compass that were caused by those differences. If
your computer was setup in a way that I haven't anticipated, that could
explain why I can't duplicate the problem.
You can get to the Regional Setting under Win7, by going to the Start Menu
and selecting the "Control Panel" option. In the Control Panel, select
"Region and Language." Finally, on the "Formats" page, select the "Advanced"
option. There are a number of different settings here. The ones I have had
the most trouble with are the number format settings.
Let me know what you think.
Larry
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
st1\:*{behavior:url(#default#ieooui) }
Date: Mon, 07 Nov 2011 17:16:36 -0000 From: "[email protected]" Subject: Loasding compass into Widows 7 I had a problem loading compass software. No icons appeared and I cannot start the software even tho the software folders are present on 'C' drive. Any similar experiences with other users or fix advice please. Mark Tringham
Date: Tue, 8 Nov 2011 04:15:34 -0700 From: "Larry Fish" Subject: RE: [compass-users] Loasding compass into Widows 7 Hi Mark, Thanks for you question. I don't think I've heard of anyone having similar problem. Compass should put the Project Manager Icon on the Desk Top when you install Compass. You don't provide many details, so I need to ask a few questions: 1. Are you installing the Base Compass program or are you installing one of the other Compass support programs? 2. When you install Compass, do you see the Compass Installing walk you through a series of questions about the installation? 3. What is in the Compass installation folder? You should find a copy of "comp32.exe," which is the Project Manager. 4. You should be able to double click on "comp32.exe" and it should run the program. In general, you need to give me as many details about the process as you can. Larry Fish _____ From: [email protected] [mailto:[email protected]] On Behalf Of [email protected] Sent: Monday, November 07, 2011 10:17 AM Subject: [compass-users] Loasding compass into Widows 7 I had a problem loading compass software. No icons appeared and I cannot start the software even tho the software folders are present on 'C' drive. Any similar experiences with other users or fix advice please. Mark Tringham