Something VERY wrong with Compass plot

Techniques, equipment and issues. Also visit the NSS Survey & Cartography Section.

Moderator: Moderators

Something VERY wrong with Compass plot

Postby rlboyce » Mar 13, 2011 7:25 pm

Let me just start off by admitting that I'm a very new Compass and PocketTopo user.

Okay, I put a PocketTopo plot on top of a Compass plot, and this is what I got:
Image

I did not alter either plot in any way other than cropping from screen shots and proportionately resizing. Obviously the angles are different, but apparently the distances aren't proportionate either (if 1.5-1.6 and first6-first7 are equal, 1.6-1.7 and first7-first8 should also be equal):
Image

Here are the screen shots of the data if you are wondering if I entered data incorrectly:
Image

Now for the unbelievable part. I took a transparent cave survey protractor and measured the angle between stations 1.4, 1.5 and 1.6 (1.5 being the vertex, 1.4 and 1.6 being the ends) after calculating this angle to be 107.5. The PocketTopo plot gave me roughly 107 degrees, but the Compass plot gave me roughly 95 degrees!

I've double checked the Compass plot, and it's definitely on the plan view and not some weird orientation that might make the angles look different. This is a single survey with no loops tied in, so software loop error corrections shouldn't exist. I guess it's possible that I have some setting incorrect in Compass, but it doesn't make much sense to me why Compass would have a setting that made the angles messed up. The strange part is some of the angles appear to be what they are supposed to be.

Do any of you have any ideas??? Surely this software that MANY cavers use isn't inaccurate at plotting survey lines? One of my friends printed out a Compass plot to sketch around in a cave, and now I seriously question how accurate his map was.

Any help or input you can provide would be very much appreciated.
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby Extremeophile » Mar 14, 2011 1:11 am

rlboyce wrote:I did not alter either plot in any way other than cropping from screen shots and proportionately resizing.

I think your problem lies in this statement. Using Compass I entered the data and compiled it, and the line plot appears like your dark red line plot. I printed this and the angle you identified measures to 107.5 degrees. If you export the data as a metafile or svg file the correct proportions / angles should be preserved. I believe that by using the clipboard and cropping (which now creates a disproportionate image), and then resizing, you're skewing the Compass image.

Derek
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 14, 2011 7:16 am

Thanks for the help Derek! I must admit I am not quite as convinced as you that cropping and resizing has any effect on the angles or proportionality (remember that I did the same thing with the Pocket Topo plot and correct angles were maintained), but I am more hopeful that printing or exporting to an image will produce accurate angles.

To test your theory, I will measure the angle as it appears on the screen (without taking a screen shot, cropping and resizing). I will also print the plot out and do the same thing. I hope your theory is correct about screen shots, because this would be an easy fix for me. Thanks again for the help!
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby John Lovaas » Mar 14, 2011 9:37 am

rl-

I'm with Derek on the image-scaling/rotating issue. One variable to eliminate is the rotation you need to do to get these images to fit. I have PocketTopo and Compass, and north is 'up'. Both screenshots could be saved with Nm up- I'm curious why you had to do any rotation at all.

You could also export identical DXF files out of each of the survey programs, then compare them in just about any image editing program.
imbecile sheepherder.
User avatar
John Lovaas
NSS Hall Of Fame Poster
 
Posts: 590
Joined: Sep 6, 2005 9:10 am
Location: Woodstock, Illinois
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 14, 2011 7:44 pm

As I said, I hoped you guys were right, but it turns out that even after exporting as a bitmap I am still getting the wrong angles!

For those of you who doubt me, here is the 100% unaltered bitmap:
Image

Check the angle between first5, first6 and first7 for yourself... I'm still getting 95° instead of the 107.5° that I'm supposed to.

I'm at a total loss of where I should go from here. I really don't know how to solve my problem. I believe you when you say yours plots just fine, but why is it that mine does not?? I know it's not my computer because the Pocket Topo's angles are perfectly fine, so it must be some Compass setting. But once again, why would Compass have a setting that gave me the wrong angles??

I'll check the dxf files tomorrow at work with autocad, but at least shouldn't I be getting the correct angles on the bitmap???

Thanks for your help, and hopefully I can fix this problem so I am confident to use this program for mapping.
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby Extremeophile » Mar 14, 2011 8:51 pm

Here's what I get when I export as a bitmap image:

Image

My original thought on the crop and rescale was that if you change the aspect ratio of the image and then rescale it might skew the image. After thinking about it a little longer I agree with you that it should not. Seems like something else is going on.

I'm using version 5.10.10.7.164 of Compass, which I think is the most recent.

Did you try printing directly from Cave Viewer?

You might also try email Larry Fish directly. He's very helpful and responsive.

Derek
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 14, 2011 10:12 pm

I measured the angle on your plot image and got the correct angle, so now I know that it's not that I have poor protractor skills either. : )

I had not printed from Cave Viewer yet. I did so, and the plot has been printed with the correct angles! How strange! So why is it that in cave viewer and when exporting to an image file the angle is 95°, but when I print it out it's 107.5°??

I sure hope no one out there has had the same problem I've had, not realized it, exported the plot to an image, printed the image and sketched around it!

If you guys can't come up with any reason for the differences in angles, I will probably contact Mr. Fisher tomorrow afternoon. Thanks again! I think we're on the right track now.

BTW, my version of Compass is 5.11.1.1.165
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby HKalnitz » Mar 15, 2011 6:24 am

I have historically had problems with exporting bitmaps from Compass - it has been a weakness for compass for a while. Many computers manipulate bitmap prints or screen scaling and cause issues
When exporting to illustrator 9 or 10, most people used Jim Olsens excellent cave illustrator plug-in that imported compass directly into illustrator correctly - in fact many folks I know still use illustrator 10 for that reason, refusing to upgrade.
When I upgraded to illustrator CS3 i would still run illustrator 10 to import the plot and then copy paste the layer into CS3.

However, now Larry's new SVG exporter does a much better job and I have been using that - it too has its quirks but seems to work much better then exporting screen print bitmaps

Larry has a yahoo group that he monitors all the time and replies to quickly - also you can search it - plotting issues have come up before - I think I have asked about it twice myself, as far back as 2008.
here is the location:
http://tech.groups.yahoo.com/group/comp ... d=64931789

Howard
The NSS needs Cavers more then Cavers need the NSS
HKalnitz
Frequent Poster
 
Posts: 75
Joined: Aug 18, 2006 2:04 pm
Name: Howard Kalnitz
NSS #: 20678
Primary Grotto Affiliation: Great Cincinnati Grotto
  

Re: Something VERY wrong with Compass plot

Postby Caveman18 » Mar 15, 2011 4:56 pm

Actually, I have found (with much trial and error) that alot of this is caused by screen resolution and size. In fact, even a WMF can be warped if the screen isn't set properly on your viewer. I had the same problem, only dealing with 153 miles of cave passage and couldn't figure it out. When I upgraded to Map Publisher for Adobe Illustrator, and was able to import shapefiles instead of WMF or BMP files the situation resolved itself pretty easily. I would recommend playing with your screen resolution / configuration before you take a screen shot, or try exporting in a different format such as a shapefile that preserves the angles accurately.
Caveman18
Infrequent Poster
 
Posts: 18
Joined: Dec 21, 2009 4:33 pm
Name: Dan Austin
NSS #: 46880
Primary Grotto Affiliation: Paha Sapa Grotto
  

Re: Something VERY wrong with Compass plot

Postby Extremeophile » Mar 15, 2011 5:15 pm

Maybe we'll make it 154 miles after this weekend. Gotta work harder to stay ahead of those Mexican caves. :bananawhip:
User avatar
Extremeophile
NSS Hall Of Fame Poster
 
Posts: 761
Joined: Dec 7, 2009 7:37 pm
Location: Littleton, CO
Name: Derek Bristol
NSS #: 34941
Primary Grotto Affiliation: Colorado Grotto
  

Re: Something VERY wrong with Compass plot

Postby Stan Allison » Mar 16, 2011 8:27 pm

Have you checked to make sure that you have magnetic declination set to zero degrees and compass correction set to zero degrees in Compass? This could be another source of your problem.

Stan
Stan Allison
Occasional Poster
 
Posts: 45
Joined: Jun 6, 2007 5:08 pm
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 16, 2011 9:44 pm

I have not checked these, no. I will in the near future, but at the moment I must spend most of my time helping my wife with our newborn! :banana_yay:

I am curious though, wouldn't both of these corrections just rotate everything as opposed to shifting only some of the angles? Like I say, I'll check anyway just to be sure.

I haven't been able to post on the Compass forum yet either... I haven't been added as a member that can post yet. I may have to email Larry directly when I get a chance.
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 19, 2011 10:16 am

Just to give you an update, with the help of the Yahoo Compass Group my problem has been solved. Turns out that I didn't have the latest driver for my graphics card. The Compaq website said the driver I had was the latest, so I didn't bother to verify their claim. After trying a lot of different things, I came across the graphics properties once again, and I noticed there was a way to automatically check for the latest Intel drivers. After finding a newer driver and installing it, the problem was immediately solved.

BTW, Larry is incredibly helpful, and it's too bad other software developers aren't more like him.
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  

Re: Something VERY wrong with Compass plot

Postby rlboyce » Mar 19, 2011 2:58 pm

Finally, here is a link to the Compass group discussion in case one of you has a similar problem: http://tech.groups.yahoo.com/group/compass-users/message/264. Many good suggestions were made in this discussion and one of those just might help you if you have a problem.
Happiness can be measured in mud/in².
User avatar
rlboyce
NSS Hall Of Fame Poster
 
Posts: 727
Joined: Jan 22, 2010 7:26 pm
Location: Chambersburg, Pennsylvania
Name: Ryan Boyce
  


Return to Survey and Cartography Forum

Who is online

Users browsing this forum: No registered users

cron