Archive of Questions and Answers From Compass Users
Background. This page contains questions and answers that appeared on the
Compass Users Group from 2000 to 2005. For some reason, Yahoo
Groups "lost" the Compass Group and so I have had to reconstitute it. Once Yahoo
closes a group, all the messages are lost. However, with help of some very kind
Compass users, I was able to recover all the old postings and they are presented
here. Many the thanks to the Roger Schuster and Paul R. Jorgenson for
sending me their archived material.
Content. There are about 180 messages covering an
extremely wide range of topics. I have pulled out all the extraneous
header information, advertising and signature blocks so it is
less cluttered. I have also formatted it so it is easy to
pick out each letter and easier to browse through. One nice
thing about the archive, is that since all the messages are
on one page, it is very easy to search the messages using "Find" option in your
browser. You can also
download the whole file so you can keep it handy.
Joining The Group. Click here for
information about how to join the Compass User Group.
To: compass-users list
From: Roger Schuster
Date: Mon, 17 Jul 2000 20:09:32 +0200 (CEST)
Subject: [compass-users] Welcome to all!
Hello cavers!
A warm welcome to all of you! I am the maintainer of this
mailing list and I invite you to start up the discussions on
this forum and share your thoughts and experiences about
"Compass" with other cavers.
At the moment 12 people are subscribed to "compass-users" and
as far as I can see most of them are from the U.S., but also
from Austria, Germany, New Zealand and Sweden. I am sure this
is the beginning of a nice international caver's forum.
By the way, if you not already have done this, the Charta of
this group may be a good bed-time reading for some of you:
http://www.karst.net/Compass/charta.htm .
O.k., the first "real" posting to the list will follow soon...
Good caving!
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users list
From: Roger Schuster
Date: Mon, 17 Jul 2000 22:04:16 +0200 (CEST)
Subject: [compass-users] CaveBase problem
X-Evolution: 00000004-0010
Hello cavers,
here is the first question on the "Compass" mailing list.
I have CaveBase 3.000203 (the recent release) and tried to
create a new cave database and exporting the data to the cave
viewer. Unfortunately the cave viewer gives an error message
"Viewer: Invalid floating point operation" if I enabled the
"Plot" setting in the "Complex Plotting" preferences dialog
and try to leave with the "show changes" button.
The cave database contains only one dataset for test
purposes:
Name VP Temp Feucht Datum
Value 9 9,8 79 19.06.1999
Type String Float Float Date
where VP ist the station name expressed as string. Here are also the survey
data:
Finsteres Loch
SURVEY NAME: 06.1.
SURVEY DATE: 7 10 1988 COMMENT:Kat.Nr. 7225/06
SURVEY TEAM:
Döbber, Schuster
DECLINATION: 0.00 FORMAT: DMMDLRUDLDAN CORRECTIONS: 0.00 0.00 0.00
FROM TO LENGTH BEARING INC LEFT UP DOWN RIGHT FLAGS COMMENTS
0 1 27.49 57.00 -9.00 -9.90 4.59 4.92 9.84
1 2 60.86 355.00 2.00 3.28 6.56 4.92 29.53
2 3 38.71 60.00 4.00 22.97 6.56 6.56 6.56
3 4 39.89 20.00 -1.00 3.28 9.84 8.20 19.69
4 5 43.14 74.00 -1.00 7.55 5.91 4.27 0.98
5 6 43.47 17.00 12.00 1.31 1.64 3.28 8.20
6 7 25.85 347.00 6.00 0.00 1.97 1.64 3.94
7 8 41.34 31.00 34.00 6.56 8.86 6.56 9.84
8 9 59.88 40.00 -10.00 11.81 39.37 2.62 1.97
9 10 28.54 56.00 -14.00 2.62 2.30 2.95 3.94
10 11 35.66 60.00 -4.00 3.28 10.66 3.28 -9.90
11 12 18.04 85.00 -15.00 0.00 4.27 0.92 8.20
I made a simple query to get the names of the stations with
temperatures greater than 5 deg.
The whole thing runs on a German Windows 95 version and the
Borland Database Engine is set to dbase as standard file
format and uses the comma as decimal separator and the dot as
thousand separator (the common notation in many European
countries). I think, this is the well-known comma vs. dot
problem that already caused some trouble in the past.
Another question is how to query fields of the "date" type.
CaveBase (or the BDE respectively) uses the regional settings
from the Windows Control Panel automatically and stores the
date in the dd.mm.yyyy format on my box. Which expression I
must enter to get all data that are measured at a certain
date? datum = 19.06.1999 and datum = 19991906 both don't
work. The latter is the internal format of most SQL
databases.
Thanks for any advice!
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Christian Docksey
Date: Mon, 17 Jul 2000 17:45:24 -0400
Subject: Re: [compass-users] Welcome to all!
X-Evolution: 00000006-0010
Hello
Well, I'm in Venezuela at the CEE which is the Simón Bolívar University
caving group. We are dedicated to vertical caving mainly and have the
highest technical level in the country in this aspect. I hope to share
experiences with you all in addition to the surveying aspect (ie:
compass).
I have not been able to download the latest version of compass (as
trialware). I have tried both USA and European sites and conection seems
fine but when I hit the save button, it just hangs there, as if no
comunication. I have ISDN at 256K, so speed is no problem. Anybody else
have a problem here like me??>
Many regards to you all....
Christian
Roger Schuster wrote:
> Hello cavers!
>
> A warm welcome to all of you! I am the maintainer of this mailing list
> and I
> invite you to start up the discussions on this forum and share your
> thoughts
> and experiences about "Compass" with other cavers.
>
> At the moment 12 people are subscribed to "compass-users" and as far
> as I
> can see most of them are from the U.S., but also from Austria,
> Germany, New
> Zealand and Sweden. I am sure this is the beginning of a nice
> international
> caver's forum.
>
> By the way, if you not already have done this, the Charta of this
> group may
> be a good bed-time reading for some of you:
> http://www.karst.net/Compass/charta.htm .
>
> O.k., the first "real" posting to the list will follow soon...
>
> Good caving!
>
> Roger
>
> --
> Roger Schuster eMail roger at r-schuster.de
> Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Roger Mortimer
Date: Mon, 17 Jul 2000 23:35:39 -0700
Subject: [compass-users] LRUDs
X-Evolution: 00000008-0010
Hi everyone,
Here's a Compass question that might engender some general
discussion. When I sketch, I am generally behind the people actually
surveying. Because of this I want passage dimensions (LRUDs) based on
90 degrees relative to the shot from the "From Station." This gives me
a better target as I approach the station. Some (many?) prefer the
LRUDs to be relative to the upcoming shot. Some will bisect the angle
between the two.
Now Compass allows for marks to be plotted based on From or
From/To. But as I look at the plot it seems like the "Mark From"
actually is relative to the upcoming ("To") station. To be able to plot
LRUDs I have to use the Mark From/To then ignore half of the marks. The
From/To option seems most adept at estimating bisected angles since
Compass doesn't offer it. Perhaps some future edition might allow for
"Mark From," "Mark To," and "Mark Bisect" options? Or to get way too
complicated perhaps each survey in the file needs to be marked as to the
option used underground since not everyone interprets these the same
ways and in a large project multiple interpretations might happen.
Any thoughts?
And Larry, I love using your program!
Roger Mortimer
To: compass-users at egroups.com
From: "evan anderson"
Date: Tue, 18 Jul 2000 07:07:24 GMT
Subject: Re: [compass-users] LRUDs
X-Evolution: 0000000a-0010
Hello Roger,
From/To option seems most adept at estimating bisected angles since
>Compass doesn't offer it. Perhaps some future edition might allow for
>"Mark From," "Mark To," and "Mark Bisect" options? Or to get way too
>complicated perhaps each survey in the file needs to be marked as to the
>option used underground since not everyone interprets these the same
>ways and in a large project multiple interpretations might happen.
The Compiler has the option of using the "To" or "From" station as a
saveable setting. But like you said, multiple interpretations occur.
I also think being able to have some type of setting in the Editor and not
the Compiler would fix the problem. Maybe an easy way to do this is by just
shifting the entries up or down one cell, since all data is entered on a
single line.
For now we just don't have the ability to render our larger project caves
perfectly in 3-D, but I think that what we do have is sweet enough.
Just think of how far Larry has taken this program, and where it will be
this time next year.
later
evan anderson
To: compass-users at egroups.com
From: Larry Fish
Date: Tue, 18 Jul 2000 11:50:21 -0600 (MDT)
Subject: [compass-users] Download problems
X-Evolution: 0000000c-0010
In regard to Christian's download problems: I get an occasional complaint
about downloading problems, but since I got the new web site, I have had
very few complaints. The fact that the problem occurs on both the COMPASS
site and Mirror site, makes me think the problem is probably somewhere
between Christian's computer and the web sites.
I have noticed that there are times of the day when general internet
traffic is so heavy that it takes forever to download files. This is
particularly true of transmissions between countries. When I first posted
COMPASS on the net, the files were on an ftp site in Brittain. I found
that the way overseas traffic patterns worked, the most reliable time to
access the files was at 4:00 AM local time.
The other possibility is that the problem is caused by something local to
Christian's computer. Perhaps he could try another computer connected to a
different ISP or network. If the problem went away, that would isolate the
problem to a particular computer, ISP or network.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Tue, 18 Jul 2000 20:01:50 +0200 (CEST)
Subject: Re: [compass-users] Welcome to all!
X-Evolution: 0000000e-0010
On Mon, 17 Jul 2000, Christian Docksey wrote:
Hello Christian,
> I have not been able to download the latest version of compass (as
> trialware). I have tried both USA and European sites and conection seems
> fine but when I hit the save button, it just hangs there, as if no
> comunication. I have ISDN at 256K, so speed is no problem. Anybody else
> have a problem here like me??>
I tested the European site with Netscape 4.7 (Linux) few minutes ago and
it worked well. Probably it will help if you use another browser? Or get the
files directly through FTP - if you need info about how to do this, let me
know.
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Wed, 19 Jul 2000 02:12:04 -0600 (MDT)
Subject: Re: [compass-users] LRUDs
X-Evolution: 00000010-0010
In response to Roger Mortimer's question, I have added a feature to the
Viewer that marks the passage walls associated with the "TO" station. The
new version will get posted with the next release.
I don't know how many people are using the "bisected angle" method of
measuring LRUDs, but I'm not sure it is a good approach.
To review, with the bisected angle method, the surveyor takes the angle
between the last and next shot and splits it in half. He/she then measures
the LRUDs along that plane. This means that LRUDs for the last and next
shot are merged because they are measured at the exact same angle instead
of at right angles to each shot. The advantage of this is that the
polygons that describe are neatly connected together.
The big problem is that when the passage turns sharply the polygon narrows
as it goes into the corner. This occurs because of the way you have to
draw the lines when you connect the LRUDs from one station to the next. It
is like the way a hose kinks when you bend it too far. It is even worse if
you have two sharp turns in a row. Then the lines go from one narrowed
corner to another and the whole passage gets unrealistically narrow. The
only way around this problem would be to use some technique to round the
corners like "splines." COMPASS already does that, so bisecting the angle
does not improve the quality of the passage model.
The issues around passage wall modeling are complicated. The current
standards do not allow a survey program to accurately model a cave
passage. The most obvious example of this is the question: what do the
LRUDs mean on a vertical shot?
What COMPASS tries to do is come up with a reasonable compromise to give
more or less reasonable look passages, even though we all know they are
not exactly right.
I agree with Evan's suggestion of setting the LRUD mode in the Editor so
that each survey can have a different setting. It is on my list of things
to do and I will probably add the feature in the near future. However, it
is a more complicated feature to add because it means that I have to
modify the file format. That means that every program the works with the
COMPASS file must be modified to handle the new format. More than one days
work.
Larry
To: compass-users at egroups.com
From: Larry Fish
Date: Wed, 19 Jul 2000 02:21:43 -0600 (MDT)
Subject: Re: [compass-users] CaveBase problem
X-Evolution: 00000012-0010
Roger,
I was sure that your problem was another "decimal point" problem. But, I
tried everything I could to reproduce the problem and I could not. I set
the Regional Settings so the Decimal Point was a "comma" (,), and the
thousands separator was a "period" (.). I also set the date format to
match yours and I even set the units to meters. Inspite of all this, I
could not reproduce the error. When I query, the program nicely labels one
of the stations with a symbol.
I am currently running Win98 and slightly newer version of the database
engine. Beyond that, I can't think of a reason why you would be getting
this error and me not. Did the sample database I sent work? Maybe there
is a clue there.
On the subject of querying date fields, I have figured some of it out.
The date field in the Borland Database Engine is encoded as the number of
days since January 1, 00001. This means that the number 729924 matches
19.06.1999.
I have added an option to CaveBase that will do the translation from
normal dates to the encoded dates for you. I'm not entirely sure this is a
good idea, for the time being I will try it. I think that the date format
actually depends on the database engine that you are using. CaveBase can
connect through ODBC to an Access database and then the new date field
would not work. The change will get posted with the next release.
Decimal Point Discussion.
For those of you who are trying to follow this discussion, the United
States uses a "period" as a decimal point but some other countries such as
Germany use a "comma" as the decimal point. To accommodate these
differences, Windows allows you to configure the format for your own local
settings. As Roger knows, this has been the source of a lot of problems.
The programming lanugages I use automatically check the Windows settings
and make the appropriate adjustments. You would think that that would make
it easy, but it doesn't. If I let the programming language automatically
adjust to the local settings, cavers in Germany will write a survey file
with commas in it and a caver in the United States will be unable to view
the file. So I have chosen that COMPASS will use a period when it reads or
writes files. However, it still must be aware of the local settings when
it displays or edits data. Thus I have to be very careful to turn on and
off the local settings at the right time.
Roger will probably be amused to know that after I set all my settings to
the German standard, I forgot to set them back. I spent about an hour
tonight trying to figure out why Excel wouldn't let me enter the the
number 0.3048.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Wed, 19 Jul 2000 20:56:23 +0200 (CEST)
Subject: Re: [compass-users] CaveBase problem
X-Evolution: 00000014-0010
On Wed, 19 Jul 2000, Larry Fish wrote:
Hello Larry,
thank you for your fast and friendly response.
> Did the sample database I sent work? Maybe there is a clue there.
The first and short test with your example data was positive so I am
wondering what's the problem with my own data! I'll do some more tests and
look what's wrong.
Btw: The BDE can read and write files in dbase format but which version of
dbase and which code page it uses?
> Roger will probably be amused to know that after I set all my settings to
> the German standard, I forgot to set them back. I spent about an hour
> tonight trying to figure out why Excel wouldn't let me enter the the
> number 0.3048.
Do you also have watched the interesting effect that if you hit "z" on the
keyboard you get a "y" on screen? :-)
Best regards
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Thu, 20 Jul 2000 02:23:27 -0600 (MDT)
Subject: Re: [compass-users] CaveBase problemn
X-Evolution: 00000016-0010
Roger,
> Btw: The BDE can read and write files in dbase format but which version of
> dbase and which code page it uses?
There are some tools that come with the DBE that allow you to set various
parameter for the different database options. I haven't played with them
very much, but I think it controls things like how various data types are
interpreted. Also, Borland has a lot of information about their database
engine on their web site. They even have newer versions of the database
engine. The address is:
http://www.borland.com
Also, with the right configuration, you can connect the database engine to
just about any type of database. I have connected to MS Access databases
and even RBase databases. The trick is to use ODBC. I did it a couple
years ago as an experiement, so don't ask me how to do it.
Larry
To: compass-users list
From: Roger Schuster
Date: Sat, 22 Jul 2000 14:14:01 +0200 (CEST)
Subject: [compass-users] CaveBase
X-Evolution: 00000018-0010
Hello all,
I made some more experiments with the CaveBase and Borland Database Engine
and nothing worked!
I changed the country settings in the Control Panel to U.S. English, I
tested several code pages in the BDE, I downgraded to an older version of
the Cave Viewer, I updated the BDE first to 3.5 and later to 5.11 and
nothing helped me on with the problem.
I can enter data into the data base, query them and send it to the Cave
Viewer successfully but if I set the feature info to visible in the Complex
Plotting dialog the known error message "Invalid floating point operation"
occurres. This error appeares with any kind of data, not only with numbers
of type float!
Since those days as CaveBase was working I installed three new applications
on my PC which probably cause a conflict with Cave Base. The most suspicious
was Interbase SQL Server 5.1 and Interbase ODBC driver running in
the background so I uninstalled them. Result: Nothing! I can't get CaveBase
working.
While playing around with several software updates for BDE I experienced
that Cave Base will fail completely with BDE 3.5 and 5.11: "Exception
EDatabaseError in module CAVEBASE.EXE at 000A:11ED. An error occurred while
attempting to initialize the Borland Database Engine (error $2108)."
CaveBase will only start up with BDE 1.0 which is available from the Compass
Web site but I still can't pass the data to CaveViewer.
That's it!
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Sun, 23 Jul 2000 02:10:30 -0600 (MDT)
Subject: Re: [compass-users] CaveBase
X-Evolution: 0000001a-0010
Roger,
> I changed the country settings in the Control Panel to U.S. English, I
> tested several code pages in the BDE, I downgraded to an older version of
> the Cave Viewer, I updated the BDE first to 3.5 and later to 5.11 and
> nothing helped me on with the problem.
This is very strange. Obviously, the old "decimal point" issue is not the
source of the problem.
> Since those days as CaveBase was working I installed three new applications
> on my PC which probably cause a conflict with Cave Base.
The fact that CaveBase worked in the past, but not now is interesting. It
indicates that something has changed on your system. Perhaps these other
programs are installing some component that doesn't get removed when you
uninstall them. This is a common occurance with some programs. For
example, let's say the BDE uses a common library called X.DLL. When
install your new program, it puts a newer version of X.DLL on your
computer. This new version is not compatable with the BDE. Even when you
uninstall your new program, X.DLL is not removed and replaced with the
earlier version, because WIndows determines that X.DLL is being used by
other programs.
I have a couple ideas for solving the problem. First, could you give me a
step-by-step description of exactly what you are doing to get the error?
It could be that I am leaving out a step that is crucial for recreating
the problem. That way, I can do precisely the same steps and try again to
duplicate the problem.
If that doesn't work, I will create a special version of the Viewer that
will have a memo box that will show exactly what data is being transmitted
by the database. That should tell us why you are getting the floating
point error.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Mon, 24 Jul 2000 20:51:14 +0200 (CEST)
Subject: Re: [compass-users] CaveBase
X-Evolution: 0000001c-0010
On Sun, 23 Jul 2000, Larry Fish wrote:
Hello Larry, hello *.*,
> For example, let's say the BDE uses a common library called X.DLL. When
> install your new program, it puts a newer version of X.DLL on your
> computer. This new version is not compatable with the BDE. Even when you
> uninstall your new program, X.DLL is not removed and replaced with the
> earlier version, because WIndows determines that X.DLL is being used by
> other programs.
Can you use a debugging tool to trace the dll and function calls of Cave
Viewer, Cave Base and BDE? So we will know which shared components the
software uses.
> I have a couple ideas for solving the problem. First, could you give me a
> step-by-step description of exactly what you are doing to get the error?
O.k., here is the description:
1. Country Settings in Control Panel and win.ini: German, i.e. decimal comma
and thousands separator point.
2. BDE Configuration Utility: "Drivers" Tab Langdriver dBASE DEU cp850
"System" Tab the same as above
"Number" Tab the same as Windows country
settings.
Database format dBASE
3. Starting Cave Viewer, loading arta.plt
4. Tools --> Database
5. Create New Database: Type dBase, table name arta, same directory as
arta.plt.
6. Fields: Station String, Temp Float, Humid Integer, Diskus Boolean, Lamp
String.
7. Query Temp > 16
8. In the third tab: Station Field --> Station, Number Field --> Temp
9. Do query --> result o.k.
10. As marked locations, feature name: Temp
11. Send to Viewer --> o.k.
12. In Viewer: Preferences | Set Complex | Color Surveys
13. Survey "Temp" (type "R") set to "Plot", set color blue, set symbol Bell.
Options: Color Survey, button "on".
14. Button "Close": Error "Invalid etc.". Cave Viewer shows infinite Windows
sand glass and must be closed.
15. The same with other numeric queries.
16. Numeric but "as line segments" --> does!
17. Field of type Boolean and "marked locations" --> does!
18: Field of type String: Sometimes yes, sometimes nothing happens.
The same result I got also with your example data.
I will send you the files I used in the description above.
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Mon, 24 Jul 2000 22:52:12 -0600 (MDT)
Subject: Re: [compass-users] CaveBase
X-Evolution: 0000001e-0010
Roger,
I have found and fixed the bug!! (I think.) You will be amused. It had
nothing to do with the database engine and nothing to do with the regional
setting or the decimal points.
The problem occurred because you only had one hit from a query and were
plotting symbols of different sizes. Under these circumstances, COMPASS
has to calculate the largest and smallest values query. It uses this range
to scale the symbols. The problem occurred when you had only one value.
This meant that the highest and lowest value were the same. COMPASS
subtracts the two values and then divides. If the two values are the same,
the result is zero and then dividing by zero gives the error.
The funny thing is that the bug has existed for several years. Apparently,
nobody ever tried to plot different symbols sizes from a query that only
returned one value.
I have posted a new copy on the net. Check it out and see if it solves the
problem.
Thanks for your help and persistance! The real trick was using the exact
same data and the exact query that you used.
Larry
To: compass-users list
From: Roger Schuster
Date: Thu, 27 Jul 2000 20:20:33 +0200 (CEST)
Subject: [compass-users] new Compass files
X-Evolution: 00000020-0010
Hello all,
the new Compass files are now available on the European Compass Page, too!
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users list
From: Roger Schuster
Date: Fri, 28 Jul 2000 23:19:18 +0200 (CEST)
Subject: [compass-users] new interesting feature
X-Evolution: 00000022-0010
Hello there,
I came across a new interesting "feature" of the Compass Cave Base: If you
do a query with a field of type "date" you will see a red rectangle in the
upper right corner of Cave Viewer instead of a symbol near the station!
Larry, I send you the data I used and some (small) screen shots to explain
the settings and the result of my test. If one of the readers here is
interested, too, let me know and I put the example to the File Area of the
mailing list.
Btw: There is another problem. If you query a numeric field with a "less
than" operator the query finds also fields with no content. I thought first
that Cave Base gives fields with no content the value zero implicitly so I
made a second test. The query "less than and not equal zero" gives
the same result and this is a little problem. I will only use fields which
are not left blank.
Good caving & nice week-end
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Sat, 29 Jul 2000 22:52:23 -0600 (MDT)
Subject: Re: [compass-users] new interesting feature
X-Evolution: 00000024-0010
Roger,
> I came across a new interesting "feature" of the Compass Cave Base: If
> you do a query with a field of type "date" you will see a red rectangle
> in the upper right corner of Cave Viewer instead of a symbol near the
> station!
The red rectangle is actually a very very large symbol that the Viewer is
drawing on the screen. The problem comes from inconsistancies in the way
dates are handled in the database.
When the database sends data to the Viewer, it converts the data to a
string on numbers. I relied on the database engine to do that conversion.
It works correctly with everything except the date values. There it
converts them to a string of number like "7/19/1999". When the Viewer see
the date, it only reads the first value the "7", so the size of the
symbol grows to huge proportions. I solved the problem by changing the
code so it explicitly converts all values to numbers. I have posted the
new version on the net.
> Btw: There is another problem. If you query a numeric field with a "less
> than" operator the query finds also fields with no content. I thought
> first that Cave Base gives fields with no content the value zero
> implicitly so I made a second test. The query "less than and not
> equal zero" gives the same result and this is a little problem. I will
> only use fields which are not left blank.
I talked to one of my database programmer friends and she says that this
is normal behavior for a database. The empty fields are treated as Nulls
and Nulls have a selectable value. She says that normallly you choose a
value that works for the kind of queries you are doing. I couldn't find a
way to set the default "Null" value for the BDE, but it is probably there
some where.
My friend said that you should query for the value you want and
specifically exclude nulls. It took me most of an afternoon to figure out
how to represent a null in a query. It turns out you do this with a set of
empty quotes. Either single ('') or double ("") will work. Here is a query
that works on your data:
(FEUCHT < 78) AND (FEUCHT <> '')
I also added a feature to CaveBase that enters a Null for you.
Larry
To: users group compass
From: Larry Fish
Date: Thu, 3 Aug 2000 23:14:54 -0600 (MDT)
Subject: [compass-users] Questions (fwd)
X-Evolution: 00000026-0010
Dear Compass Group,
I got this letter and I thought the question and my reply might be useful
to others. Also, other people who are working with DXF exports might have
some thoughts.
Larry
-------------------------------------------------------------------------
Mr fish,
I have been looking over the new version of COMPASS recently downloaded.
Pat Kambesis recently announced at the NSS Conv. she'd be giving out
portions of her Lech quads. I am in the process of evaulating two
software illustrators in conjunction with COMPASS. I have long been
interested in DXF increased capabilities.
With the new release of AUtoCAD 2000i I'm finding the DXF capabilities
don't allow me to bring in anything but the raw survey and stations (just
two layers). I need more versistility. Have you any recommendations at
this time.
Also, can I now dynamic rotate the wireframe?
thanks again,
Chris
Chris Andrews
427 South Loomis St. Suite B
Fort Collins, Colorado 80521
chrisandrews at juno.com
970-495-0534
LEARN RMR REP
LEARN CARTOGRAPHIC team
To: compass-users at egroups.com
From: Larry Fish
Date: Thu, 3 Aug 2000 23:18:51 -0600 (MDT)
Subject: Re: [compass-users] Questions (fwd)
X-Evolution: 00000028-0010
Chris,
DXF file conversion has been the bane of my existance. AutoCad keeps
changing their specifications and I keep having to rewrite my conversion
routines. The DXF file format worked pretty well from AutoCad version 3
through 12, but then AutoCad made a radical departure and changed
everything. Everybody complaigned, so in version 14 they went back a bit
toward the earlier versions.
The problem for me is that every time they do this, I have to get a copy
of the latest version of AutoCad and it is not cheap. Also, all the other
drawing programs interpret the DXF standard differently. Some only import
the early versions of DXF. Some only import the 2D model. Some don't
import certain layers. Some import everything into a single layer. It is a
mess.
> With the new release of AUtoCAD 2000i I'm finding the DXF capabilities
> don't allow me to bring in anything but the raw survey and stations (just
> two layers). I need more versistility. Have you any recommendations at
> this time.
I currently have a copy of AutoCad 14. When I import the cave data into
it, it shows each survey on a separate layer, the station labels on a
separate layer, the passage walls on a separate layer, the legend and
north arrow on a separate layer. For Groaning Cave, there are about 200
layers.
What other things would you like to have in the DXF export?
Are you using the DXF exporter that is in the Windows Viewer? It sounds
like you might be using the old exporter that comes with the DOS version.
The DXF exporter was added to the Viewer in last few months. If you don't
have the latest version, you may not have the new exporter. The latest
version of COMPASS is always available on the internet. Incase you don't
have the address, it is at:
http://fountainware.com/compass
> Also, can I now dynamic rotate the wireframe?
Do you mean in COMPASS or Autocad? In AutoCad 14, you select "View|3D
Dynamic View" from the menu bar.
If you mean COMPASS, you should be able to get realtime pans, zooms and
rotation if the cave isn't too big and your computer is fast enough. Even
with large caves like Lechuguilla, I can get realtime refresh rates of 2
or 3 frames per second.
To get smooth animation, you have to set the Viewer properly. First run
the Viewer with the cave you want to view. Next select "Option|Hidden
Refresh Screen" from the Menu Bar. Also, select "Options|3D Compass" from
the menu bar. Finally, put the program in profile mode by hitting the
"Plan/Profile" button and then press the "Enter Box/Redraw" button.
You can now smoothly Pan, Zoom and Rotate by clicking on the appropriate
buttons. The "Multiplier Setting" box on the tool bar controls the amount
of movement each click accomplishes. Smaller steps make for smoother
motions. A multiplier 5 is a good choice.
If you want to smoothly rotate with the mouse, move the mouse cursor to
the "3D Compass", click the left mouse button and drag the "arms" to a new
position. The cave will track the 3D Compass position.
If you want to use the mouse to zoom or pan, push "Drag Mode" button on
the tool bar. You can now us the left mouse button to drag (pan) the cave
left, right, up or down. You can also use the middle mouse button to zoom
in and out and do "flythroughs."
I think I will post your letter on the COMPASS users group. Maybe some of
the other users will have some suggestions for you. If you would like to
subscribe to the users groups, there are instructions on the COMPASS web
page.
If you have any other questions, feel free to write.
Larry
To: compass-users at egroups.com
From: Christian Docksey
Date: Fri, 04 Aug 2000 14:51:04 -0400
Subject: [compass-users] Cave Graphics Extended
X-Evolution: 0000002a-0010
Hello Larry.
First let me show my appreciation at your efforts at solving queries and
problems quickly and thoroughly!
I managed to download the software finally. The problem was Gozilla
which didn't allow me to download any other way but through him, and it
was Gozilla that wouldn't work with this download. I uninstalled it and
all OK.
I have yet to try out the methods you describe to have an interactive
mouse control. I hope it is what I imagine. I thought that maybe I could
take this "graphical"point a little further.
In todays world where graphics are so "taken for granted" there is the
need for 2 things.
1) To simplify the task of creating the artwork for a cave publication.
2) To be able to create a 3D model of the cave which can be used to
visualize the cave better and for analysis for future expeditions.
For creating the drawings (artwork) we use CorelDraw. This is to have
info in a vectorial format which allows for any required resolution.
Also, one has to keep in mind that there is the UIS standard to maintain
for symbols, etc. What would be useful (and please correct me if Compass
can do this already) is to be able to obtain an "extended profile view"
(Please excuse me if I don't use the exact term here but I don't know
it. In Venezuela we use the spanish terms and so I can only attempt to
explain it in English). This is NOT a projected view, but more like the
"stretched out" view of certain gallerys (which could be easily selected
in Compass ie: from stationX to stationY). In other words, the length of
the drawn galleries is equivalent to the fully walked or measured
distance (and not a component of the distance like Dist x cos(angle)). I
hope you follow what I mean. If not I'm willing to try again or send an
example.
The info needed from Compass would be the vector centre-line and also
the roof and floor outline (which we could add detail to on the
drawing). As to the format for these vectors, we go to the next point.
One of the most used and flexible formats around is Adobe Illustrator.
Nearly every software package used in graphics has an import and/or
export utility for Illustrator. With this info we could use CorelDraw,
Illustrator itself, and virtually any other package for 2D work (for
creating the artwork). Also 3D packages like 3DStudio Max, Maya,
lightwave all accept this vector format too!
Now, for 3d visualization, and considering that DXF is sometimes
problematic for you, wouldn't it be better to translate "Compass" info
to a stable, widely used 3d format. The .3ds format used by Kinetix
(part of Autodesk) for their Max 3d program is widely used for importing
3d data into 3d programs. Max is probably the widest used 3d program on
the market (not saying that it is the best) and one can still import the
same .3ds file today that was created 10 years ago when the software was
DOS based. An added benefit is that one can export to DXF. The reason
for all this is that if we could get this 3d info from Compass into a 3D
animation software, we could do wonders. First of all, the interactive
rotation,etc in a 3d program is superb. Then, we can create additional
detail as needed and then even add a camera which could fly through the
inside of the galleries! Not bad, Heh! Also, pre-rendered animations
could be made and placed on a web site.
If your reply to this mail is "Shut up and keep quiet", then I don't
blame you. I'm only trying to get that stage further from having the
results as a bunch of numbers, to getting a good quality 2D drawing of
the cave. Getting the 3D info out to a 3D animation program would make
the topping on the cake. I think that Compass would definately be "the
Caving Software to have" for any serious cavers.
Forgive such a long and perhaps rediculous e-mail.
Regards...
Christian
Larry Fish wrote:
>
> Chris,
>
> DXF file conversion has been the bane of my existance. AutoCad keeps
> changing their specifications and I keep having to rewrite my
> conversion
> routines. The DXF file format worked pretty well from AutoCad version
> 3
> through 12, but then AutoCad made a radical departure and changed
> everything. Everybody complaigned, so in version 14 they went back a
> bit
> toward the earlier versions.
>
> The problem for me is that every time they do this, I have to get a
> copy
> of the latest version of AutoCad and it is not cheap. Also, all the
> other
> drawing programs interpret the DXF standard differently. Some only
> import
> the early versions of DXF. Some only import the 2D model. Some don't
> import certain layers. Some import everything into a single layer. It
> is a
> mess.
>
> > With the new release of AUtoCAD 2000i I'm finding the DXF
> capabilities
> > don't allow me to bring in anything but the raw survey and stations
> (just
> > two layers). I need more versistility. Have you any recommendations
> at
> > this time.
>
> I currently have a copy of AutoCad 14. When I import the cave data
> into
> it, it shows each survey on a separate layer, the station labels on a
> separate layer, the passage walls on a separate layer, the legend and
> north arrow on a separate layer. For Groaning Cave, there are about
> 200
> layers.
>
> What other things would you like to have in the DXF export?
>
> Are you using the DXF exporter that is in the Windows Viewer? It
> sounds
> like you might be using the old exporter that comes with the DOS
> version.
> The DXF exporter was added to the Viewer in last few months. If you
> don't
> have the latest version, you may not have the new exporter. The latest
>
> version of COMPASS is always available on the internet. Incase you
> don't
> have the address, it is at:
>
> http://fountainware.com/compass
>
> > Also, can I now dynamic rotate the wireframe?
>
> Do you mean in COMPASS or Autocad? In AutoCad 14, you select "View|3D
> Dynamic View" from the menu bar.
>
> If you mean COMPASS, you should be able to get realtime pans, zooms
> and
> rotation if the cave isn't too big and your computer is fast enough.
> Even
> with large caves like Lechuguilla, I can get realtime refresh rates of
> 2
> or 3 frames per second.
>
> To get smooth animation, you have to set the Viewer properly. First
> run
> the Viewer with the cave you want to view. Next select "Option|Hidden
> Refresh Screen" from the Menu Bar. Also, select "Options|3D Compass"
> from
> the menu bar. Finally, put the program in profile mode by hitting the
> "Plan/Profile" button and then press the "Enter Box/Redraw" button.
>
> You can now smoothly Pan, Zoom and Rotate by clicking on the
> appropriate
> buttons. The "Multiplier Setting" box on the tool bar controls the
> amount
> of movement each click accomplishes. Smaller steps make for smoother
> motions. A multiplier 5 is a good choice.
>
> If you want to smoothly rotate with the mouse, move the mouse cursor
> to
> the "3D Compass", click the left mouse button and drag the "arms" to a
> new
> position. The cave will track the 3D Compass position.
>
> If you want to use the mouse to zoom or pan, push "Drag Mode" button
> on
> the tool bar. You can now us the left mouse button to drag (pan) the
> cave
> left, right, up or down. You can also use the middle mouse button to
> zoom
> in and out and do "flythroughs."
>
> I think I will post your letter on the COMPASS users group. Maybe some
> of
> the other users will have some suggestions for you. If you would like
> to
> subscribe to the users groups, there are instructions on the COMPASS
> web
> page.
>
> If you have any other questions, feel free to write.
>
> Larry
To: compass-users at egroups.com
From: Larry Fish
Date: Sun, 6 Aug 2000 03:07:22 -0600 (MDT)
Subject: Re: [compass-users] Cave Graphics Extended
X-Evolution: 0000002c-0010
Christian,
Thanks for your letter.
> What would be useful (and please correct me if Compass can do this
> already) is to be able to obtain an "extended profile view" (Please
> excuse me if I don't use the exact term here but I don't know it. In
> Venezuela we use the spanish terms and so I can only attempt to explain
> it in English).
I think I understand what you are asking for and "extended profile" is a
good term for it. If I understand what you want, the "extended profile"
should take the cave passage and flattening it against a single vertical
plane. Normally, when you view a passage profile, the parts of the passage
that are coming toward you or going away from you are invisible. The
"extended profile" would have the effect of sort of "unwinding" these
passages so that all of the length would be visible.
I have toyed with implementing this feature for a couple of years.
Implementing the feature on a whole cave gets pretty complicated. As you
are "flattening" out the passages, the positions of all the stations move.
Since each station can be a "connecting point" for another branch of the
survey, I would have to keep track of the new position of each station.
This would require a complete recompile of the survey data every time you
wanted to change the viewing angle of the "extended profile." I could do
this, but, what I would really like to be able to do is manipulate these
"extended profiles" interactively and on-the-fly so you could get the
exact view you wanted. The bottom line is that I haven't figured out how
to do it so it "feels" right.
> One of the most used and flexible formats around is Adobe Illustrator.
I hadn't thought about Illustrator because nobody I know is using it. But
it is an interesting idea. I'm always afraid that a new file format will
just mean more work -- and of course, the risk that I will constantly be
trying to keep up with changes.
> Now, for 3d visualization, and considering that DXF is sometimes
> problematic for you, wouldn't it be better to translate "Compass" info
> to a stable, widely used 3d format.
This is an interesting idea. Have you played with CaveX yet? It will
export to "X" files, and I have seen converters on the web that will
convert "X" files to 3DS.
> If your reply to this mail is "Shut up and keep quiet", then I don't
> blame you.
No, no, no! I always appreciate new ideas and different perspectives.
Most of the best features of COMPASS have come from long discussions
with COMPASS users. So, keep making suggestions, I need them.
A proposito, he estado estudiando Espan~ol por un rato. Yo vivo in
el estado de Colorado y este estado fue un parte de Mejico en el
siglo XVII. Por eso, hay muchos lugares con nombres Espan~oles.
Tenemos dos canales de television en Espan~ol aqui. Me acuerdo que
hay una programma de Venezuela se llamada "Bien Venidos". Conoce
este programma?
Gracias
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Sun, 6 Aug 2000 17:08:34 +0200 (CEST)
Subject: Re: [compass-users] Cave Graphics Extended
X-Evolution: 0000002e-0010
On Sun, 6 Aug 2000, Larry Fish wrote:
Hello folks,
> Implementing the feature on a whole cave gets pretty complicated. As you
> are "flattening" out the passages, the positions of all the stations move.
> Since each station can be a "connecting point" for another branch of the
> survey, I would have to keep track of the new position of each station.
I think this is a real problem. Extended profiles are commonly used on the
small caves here in southern Germany where not much passages must be
"unfolded" and where extended profiles give less distorted impressions of
the cave than the usual perspectivic views. On large or complex caves it is
difficult to decide how the passages must become unfolded because the
surveys are inter-connected at many stations. Moving one station will damage
another survey.
Another point which must be kept in mind is in which direction the survey
must be flattended. Yesterday we surveyed a small pothole for example, where
the entrance pit intersects a small passage which extends to both
directions. For an extended profile one part of the cave must be flattened
to the left, the other part to the right. How should the computer decide
what to do?
I create such extended profiles manually. I make a copy of the original
survey data file and open it in the Cave Editor. Here I set the acimut of
all passages which should be flattened to the right to 0 degrees and that of
the passages to the left to 180 degrees. All the other shots I flagged with
the "exclude from plotting" option. If you view this in profile mode from 90
or 270 degrees this is exactly what you want: An extended profile.
I know two programs which already support e.p.. One is "Cave Render" in
which you must enter numeric values for each shot which describe how to
handle the data for extension. The other is a French product called
"Cybertopo" that has an automatic option for unfolding all surveys. I don't
know how good the results of this algorythm are.
> > One of the most used and flexible formats around is Adobe Illustrator.
Nearly as expensive as AutoCAD... :-) Why not Corel Draw which is widely
used and available for modest prices?
BTW: Have you tested the method that is described on
http://www.karst.net/Compass/hpgl.htm ? It works very good and even with old
versions of Corel Draw like 4.0 which is my favorite drawing tool.
> This is an interesting idea. Have you played with CaveX yet? It will
> export to "X" files, and I have seen converters on the web that will
> convert "X" files to 3DS.
Is there something that can convert *.x to VRML?
> > If your reply to this mail is "Shut up and keep quiet", then I don't
> > blame you.
As moderator of this list I welcome questions and suggestions like yours.
> Gracias
Freundliche Grüße
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Christian Docksey
Date: Mon, 07 Aug 2000 13:25:58 -0400
Subject: Re: [compass-users] Cave Graphics Extended
X-Evolution: 00000030-0010
Hello Larry,
I'm grateful towards your receptiveness towards some of my comments.
I totally agree with you and with Roger, that trying to "extend" the
whole of a cave is an extremely complicated task and the computer
wouldn't know what to do sometimes. No, what I mean is for "Compass" to
ask for a station to station input (say station 1A to station 1S), which
would probably be the main passageway of the cave. The stations must be
joint by a single "route". All other galleries would be ignored. Then
Compass would work out the "flattened" route and display the results. It
would be like a separate et of data so as not to cause confusion with
the rest of the data. Internally, I imagine that maybe Compass would
create new stations (say x-1A to x-1S or whatever) but display them on
screen with the same labels but in another colour or whichever way you
deem more appropiate. Why only some galleries and not all? Well, because
I'm thinking of when its time for plotting, when really it is only some
portions of the cave that you tend to plot in "extended" mode (usually
the main passage from entrance to end). It IS possible to require
another "extended" gallery when plotting, but it is 99% sure that it
will be a separate drawing on the artwork. So there is no problem that
the second be a separate set of data given by inputting the new station
to station info (say 3C to 3F).
I believe this "extended" method is really only valid for cave which are
generally horizontal in nature. This is because you are flattening the
compass readings but not the clinometer redings. A pothole, for example,
usually needs to be described by section views. And one should sketch as
many sections as needded for the shape of a cave to be understood.
Now, Roger, your example of the pothole you visited is great, because
you could tell "Compass" that you want an extended set of data for your
intersected passage. You would enter the names of the stations of the
passage, and it would give you the set of data you require. Then you
would add the drawing of that "extended" passage to your artwork in
addition to the main views of the pothole and its sections.
When I suggested Adobe Illustrator, I meant it as the file format and
not as the software itself. I don't use Illustrator as a software
either. To me Corel is great too and its what I use for this type of
thing, but then you have hundreds of preferences from different users.
Corel Imports Illustrator files like a treat, just like nearly all 2D
and 3D software, and its because of the wide acceptance of this format
that I suggested it.
Thanks Roger for the link about the PLT method of transferring files. I
had not seen it. I will try it out. Maybe there's no need for any
vector file export after all, although I don't think this method will
work for importing into 3D software. I'll check it out though!
Many Thanks to all of you. I think it's great to have the chance to
discuss a topic at all with you.
By the way, Larry. Thats pretty good spanish you have! Yes, I've seen
the odd episode of "Bienvenidos". Its a "Joke telling" program with some
girls in bikinis forming part of the cast (The best part-Right?)
Cheers to all
Chris
Larry Fish wrote:
>
> Christian,
>
> Thanks for your letter.
>
> > What would be useful (and please correct me if Compass can do this
> > already) is to be able to obtain an "extended profile view" (Please
> > excuse me if I don't use the exact term here but I don't know it. In
>
> > Venezuela we use the spanish terms and so I can only attempt to
> explain
> > it in English).
>
> I think I understand what you are asking for and "extended profile" is
> a
> good term for it. If I understand what you want, the "extended
> profile"
> should take the cave passage and flattening it against a single
> vertical
> plane. Normally, when you view a passage profile, the parts of the
> passage
> that are coming toward you or going away from you are invisible. The
> "extended profile" would have the effect of sort of "unwinding" these
> passages so that all of the length would be visible.
>
> I have toyed with implementing this feature for a couple of years.
> Implementing the feature on a whole cave gets pretty complicated. As
> you
> are "flattening" out the passages, the positions of all the stations
> move.
> Since each station can be a "connecting point" for another branch of
> the
> survey, I would have to keep track of the new position of each
> station.
> This would require a complete recompile of the survey data every time
> you
> wanted to change the viewing angle of the "extended profile." I could
> do
> this, but, what I would really like to be able to do is manipulate
> these
> "extended profiles" interactively and on-the-fly so you could get the
> exact view you wanted. The bottom line is that I haven't figured out
> how
> to do it so it "feels" right.
>
> > One of the most used and flexible formats around is Adobe
> Illustrator.
>
> I hadn't thought about Illustrator because nobody I know is using it.
> But
> it is an interesting idea. I'm always afraid that a new file format
> will
> just mean more work -- and of course, the risk that I will constantly
> be
> trying to keep up with changes.
>
> > Now, for 3d visualization, and considering that DXF is sometimes
> > problematic for you, wouldn't it be better to translate "Compass"
> info
> > to a stable, widely used 3d format.
>
> This is an interesting idea. Have you played with CaveX yet? It will
> export to "X" files, and I have seen converters on the web that will
> convert "X" files to 3DS.
>
> > If your reply to this mail is "Shut up and keep quiet", then I don't
>
> > blame you.
>
> No, no, no! I always appreciate new ideas and different perspectives.
> Most of the best features of COMPASS have come from long discussions
> with COMPASS users. So, keep making suggestions, I need them.
>
> A proposito, he estado estudiando Espan~ol por un rato. Yo vivo in
> el estado de Colorado y este estado fue un parte de Mejico en el
> siglo XVII. Por eso, hay muchos lugares con nombres Espan~oles.
> Tenemos dos canales de television en Espan~ol aqui. Me acuerdo que
> hay una programma de Venezuela se llamada "Bien Venidos". Conoce
> este programma?
>
> Gracias
>
> Larry
> -----------------------------------------------------------------------
>
>
To: compass-users at egroups.com
From: mr.doj at mailbox.swipnet.se
Date: Mon, 14 Aug 2000 13:59:50 -0000
Subject: [compass-users] Locking of mapping accuracy
X-Evolution: 00000032-0010
Hello!
Is there a way of locking survey stations when closing loops?
or
Is there some way to give surveys less/more corrections when closing
loops ie by inserting UIS/BCRA survey accuracy?
We would like to survey the main passage of a complex cave with a
higher grade and side passages with less timeconsuming surveying, but
not bend or alter the main passage when closing loops.
T Doj Sweden
To: compass-users at egroups.com
From: Larry Fish
Date: Tue, 15 Aug 2000 03:31:56 -0600 (MDT)
Subject: Re: [compass-users] Locking of mapping accuracy
X-Evolution: 00000034-0010
Hello,
Thanks for your question.
> Is there a way of locking survey stations when closing loops? or Is
> there some way to give surveys less/more corrections when closing
> loops ie by inserting UIS/BCRA survey accuracy?
You can exclude any shot from the closing process using the "Shot
Flags" in the Editor. However, you cannot set a variable level of
accuracy for different parts of the cave.
The processing of setting an accuracy for each shot is in a survey is
called "weighting." And, there is a very good reason why COMPASS does not
allow you to assign "weights." This is because COMPASS automatically
assigns "weights" to every shot in the cave based on the shot's length
and how well each loop closes.
The problem with allowing the user to assign weights, is that these
weights are only estimates. So even if you survey very very carefully,
you can still make mistakes that will destroy the accuracy of the survey.
Likewise, even a very sloppy survey can turn out to be accurate. What
COMPASS does is use the actual data to get a precise measurement of the
quality of each loop. Loops with large closure errors obviously cannot be
trusted and so they are given low weights. Likewise, loops with small
closure errors can be trusted and are given high weights. Thus COMPASS
creates a combination of weights that gives the best accuracy for all the
stations in the cave.
The actual process that COMPASS uses to analyze the quality of each loop
is a little more complicated. It begins by making a prediction about how
much error you would expect to find in a loop if all the errors were
small random errors. This is done by going through each shot in the loop
and seeing how each shot would effect the total error in the loop. This
gives an error predictions for the loop. Loops are then weighted based on
how well they match the prediction.
Assigning weights is a very important part of land surveying
and land surveyors have studied the problem for more than a century.
John Halleck has written extensively on the subject as it applies to
cave surveying. You might find his ideas interesting. His web page is
at:
http://www.cc.utah.edu/~nahaj/cave/survey/
Larry
To: compass-users at egroups.com
From: Devin Kouts
Date: Tue, 15 Aug 2000 10:13:55 -0700 (PDT)
Subject: Re: [compass-users] Locking of mapping accuracy
X-Evolution: 00000036-0010
Larry,
I understand your explanation of how compass develops
weights for certain loops in the survey. I would like
to offer some pointers that might help "increase" the
weight given to a specific survey over another. Please
comment on my conjectures (they've not been tried in
the field, just thoughts I've discussed with other
cave surveyors).
In our friend's case he wants the main trunk survey to
bear a heavier weight than any potential loops created
by side passage surveys. In that instance I would
suppose a couple changes in one's survey technique
could be useful.
First, I would limit the length of individual shots
along the trunk passage to 20 feet or 7 meters maximum
(an arbitrary number, I'm working on the assumption
that shorter distances don't propogate as much error
in the azimuth / inclination as do longer distances).
Second, I would always record forward and backward
azimuth and inclination between every consecutive
station along the trunk passage survey.
Third, I would decrease the allowable error between
forward and backward azimuth / inclinations. E.g. the
folks I survey with will allow a maximum of two
degrees difference between the foresight and the
corrected backsight azimuth and inclination. To
increase the accuracy (and the weight) of the trunk
survey I would decrease my allowable error to 1 degree
difference.
Fourth, I would avoid problematic shots, like
high-angle shots, in favor of multiple short,
low-angle, precise shots to cover the same distance.
Admittedly these techniques increase the level of
effort required to complete the survey but that's
simply the nature of information. If you desire
greater accuracy (i.e. weight) then you need to
increase the frequency and precision of the data you
collect (in this case survey data).
After the main trunk has been surveyed with these
increased standards of accuracy I presume the loops of
the main trunk (and they are a chain of consecutive
loops, foresight and backsight make a loop between
each station) will receive a greater weight from the
algorithm you described in Compass. Afterward,
surveying side passages with standard practices (i.e.
2 degree difference in fore/back sights and no
distance limits impossed on shot length) will create
relatively accurate survey but not to such a degree
that it's "weight" will degrade the accuracy of the
main trunk survey.
This approach should help our friend reach his goal of
increased deference to the main trunk survey over
errors introduced by lower quality side passage
surveys.
Comments?
Devin
--- Larry Fish wrote:
>
>
> Hello,
>
> Thanks for your question.
>
> > Is there a way of locking survey stations when
> closing loops? or Is
> > there some way to give surveys less/more
> corrections when closing
> > loops ie by inserting UIS/BCRA survey accuracy?
>
> You can exclude any shot from the closing process
> using the "Shot
> Flags" in the Editor. However, you cannot set a
> variable level of
> accuracy for different parts of the cave.
>
> The processing of setting an accuracy for each shot
> is in a survey is
> called "weighting." And, there is a very good reason
> why COMPASS does not
> allow you to assign "weights." This is because
> COMPASS automatically
> assigns "weights" to every shot in the cave based on
> the shot's length
> and how well each loop closes.
>
> The problem with allowing the user to assign
> weights, is that these
> weights are only estimates. So even if you survey
> very very carefully,
> you can still make mistakes that will destroy the
> accuracy of the survey.
> Likewise, even a very sloppy survey can turn out to
> be accurate. What
> COMPASS does is use the actual data to get a precise
> measurement of the
> quality of each loop. Loops with large closure
> errors obviously cannot be
> trusted and so they are given low weights. Likewise,
> loops with small
> closure errors can be trusted and are given high
> weights. Thus COMPASS
> creates a combination of weights that gives the best
> accuracy for all the
> stations in the cave.
>
> The actual process that COMPASS uses to analyze the
> quality of each loop
> is a little more complicated. It begins by making a
> prediction about how
> much error you would expect to find in a loop if all
> the errors were
> small random errors. This is done by going through
> each shot in the loop
> and seeing how each shot would effect the total
> error in the loop. This
> gives an error predictions for the loop. Loops are
> then weighted based on
> how well they match the prediction.
>
> Assigning weights is a very important part of land
> surveying
> and land surveyors have studied the problem for more
> than a century.
> John Halleck has written extensively on the subject
> as it applies to
> cave surveying. You might find his ideas
> interesting. His web page is
> at:
>
> http://www.cc.utah.edu/~nahaj/cave/survey/
>
> Larry
>
>
=====
Devin S. Kouts
devinkouts at yahoo.com
http://sites.netscape.net/devinkouts/
To: compass-users at egroups.com
From: Larry Fish
Date: Tue, 15 Aug 2000 12:50:23 -0600 (MDT)
Subject: Re: [compass-users] Locking of mapping accuracy
X-Evolution: 00000038-0010
Devin,
Your ideas are great and should result in higher accuracy of surveys. A
couple of comments:
If you cannot get the fore and backsight compass to agree with only a
small error, it is probably due magnet anomolies in the cave. This is due
to the fact that the fore and back compass are taken in different
locations and the magnetic anomoly will be different at those locations.
See John Halleck article on this idea on his web page. There is a link to
his web page on the COMPASS web page.
If you are absolutely sure that your survey is 100% accurate, you can
exclude all the stations from closing. Carlsbad Cavern has a theodolite
survey down the backbone of the cave and they exclude all those shot from
closing. If you don't know for sure that the survey is 100% accurate, it
is probably better to let COMPASS do the weighting.
Larry
To:
From: "Pel Torbjvrn Doj"
Date: Thu, 17 Aug 2000 11:35:27 +0200
Subject: SV: [compass-users] Locking of mapping accuracy
X-Evolution: 0000003a-0010
Thank you for youranswers. I can see that compass is valuing
my mapping and thus actually making its own survey accuracy
grade. And as Larry sais, even if you aim to to a good
survey, the "grade 5" can under circumstanses be less
acuurate than intended, and compass gives me a percentage
failure of the loop.
The techniques of surveying accurate is a too large subject
to comment, but Devins commenta about compass backsights:
I presume the loops of the main trunk (and they are a chain
of consecutive loops, foresight and backsight make a loop
between each station) will receive a greater weight from the
algorithm you described in Compass.
.......................
is interesting. This could be an answer to give the more
accurate surveyline (we will use backsights, as we do on the
surface survey) its credit. I hope compass works that way.
As for accuracy, we have around 1-2% on the larger loops
(300m+) except some unfortunate loops.
Torbjoern Doj
To: compass-users at egroups.com
From: Larry Fish
Date: Thu, 17 Aug 2000 05:11:27 -0600 (MDT)
Subject: Re: SV: [compass-users] Locking of mapping accuracy
X-Evolution: 0000003c-0010
Torbjoern,
> ............I presume the loops of
> the main trunk (and they are a chain of consecutive
> loops, foresight and backsight make a loop between
> each station) will receive a greater weight from the
> algorithm you described in Compass.
Actually, no. COMPASS does not make loops out of the foresight/backsight
pair. The accuracy comes from the fact that you have checked the fore and
back measurement against each other. If you make sure that your fore and
back readings agree, then you automatically get higher accuracy.
However, the only way you can know in a cave survey whether a survey is
accurate is by having it in a loop. If the loop closes well, then the
combined errors of all the shots in the loop must be low and so the shots
in the loop should be weighted high. That is the only thing you can know
about the accuracy of shots in a cave.
Now if you have a special survey and check and double check it, then you
might be able to say that it is more accurate than the rest. If you are
really sure, you could then override COMPASS and flag all the shots not to
be closed. However, even under those circumstances, you could make a
mistake and your assumption be wrong.
> As for accuracy, we have around 1-2% on the larger loops (300m+) except
> some unfortunate loops.
The best measure of loop quality is the Standard Deviation measurement
that COMPASS gives. Percentage is based the loops length, but the Standard
Deviation takes into account all of the individual shots and how they
might contribute to the error.
Larry
To:
From: "Pel Torbjvrn Doj"
Date: Thu, 17 Aug 2000 17:20:20 +0200
Subject: SV: SV: [compass-users] Locking of mapping accuracy
X-Evolution: 0000003e-0010
>However, the only way you can know in a cave survey whether a survey is
>accurate is by having it in a loop. If the loop closes well, then the
>combined errors of all the shots in the loop must be low and so the shots
>in the loop should be weighted high. That is the only thing you can know
>about the accuracy of shots in a cave.
Well.So if we survey the mainpassage with precisioncompass and the side-loops with handheld, compass will correct the whole loop, giving all the individual shots a correction. If we have multiple loops covering the shots in the main passage repeadetly, compass will correct the shots in the main line several times?! And if I exclude the mainline, the loop wont close at all, since the loop also includes the mainline?
>Now if you have a special survey and check and double check it, then you
>might be able to say that it is more accurate than the rest. If you are
>really sure, you could then override COMPASS and flag all the shots not to
>be closed. However, even under those circumstances, you could make a
>mistake and your assumption be wrong.
A special survey could be the surface shots or if I can get the entrances fixed with a very high accuracy. It would be nice if the entrances does not move around when I close the loops.
>> As for accuracy, we have around 1-2% on the larger loops (300m+) except
>> some unfortunate loops.
>
>The best measure of loop quality is the Standard Deviation measurement
>that COMPASS gives. Percentage is based the loops length, but the Standard
>Deviation takes into account all of the individual shots and how they
>might contribute to the error.
OK thanks!!
Torbjoern
To: compass-users at egroups.com
From: Larry Fish
Date: Thu, 17 Aug 2000 11:47:04 -0600 (MDT)
Subject: Re: SV: SV: [compass-users] Locking of mapping accuracy
X-Evolution: 00000040-0010
> Well.So if we survey the mainpassage with precision compass and the
> side-loops with handheld, compass will correct the whole loop, giving
> all the individual shots a correction.
Yes, but if the shots are part of a good loop, very little correction will
be done. Because, by definition, a good loop has a small amount of error.
Likewise, if it is part of a bad loop, a larger correction will be done.
For example, let's say you have a loop in which four shots are part of
precision survey and four shots are part of a non-precision survey. If
that loop closes badly, you cannot say whether the error was in the
precision survey or in the non-precision survey. In other words, using the
only emperical measure of the quality of the shots in the cave, you cannot
know which shots were bad, just that somewhere in the loop, there was one
or more problems.
Now you have to understand that closing loops with a survey program cannot
fix survey errors. All you can do is distrubute the error over several
shots so cosmetically, it distorts the map in a minimal way.
COMPASS does have tools which help to try to zero in on the individual
shot that has the error, but it can only narrow the search. You have to
look at your survey notes or recheck your measurements to be sure you have
found the source of the error.
> If we have multiple loops covering the shots in the main passage
> repeadetly, compass will correct the shots in the main line several
> times?!
No, COMPASS will only correct each shot one time. It begins by closing the
best loops first and then it locks all the shots in that loop so that it
cannot be adjusted again. This means that shots in the best quality loops
don't change much, while the shots in the worst loops do. This has the
effect of isolating the errors to a particular part of the cave, instead
of spreading them around like some survey programs do.
> And if I exclude the mainline, the loop wont close at all, since
> the loop also includes the mainline?
No, only the excluded shots won't be adjusted. Any adjustment will take
place on the un-excluded shots.
> A special survey could be the surface shots or if I can get the
> entrances fixed with a very high accuracy. It would be nice if the
> entrances does not move around when I close the loops.
The entrance and any fixed stations are never moved by the closure
process. All the other shots move around them. By the way, in the last
few weeks I fixed a small bug in this area. So if you don't have the
very latest version, you probably should get it.
Larry
To: compass-users list
From: Roger Schuster
Date: Sat, 19 Aug 2000 16:31:59 +0200 (CEST)
Subject: [compass-users] a CaveBase idea
X-Evolution: 00000042-0010
Hi all,
I am just back from a caving trip where we recorded some temperature and
climate data. I will enter them in Cave Base and this leads me to an idea
for a probably useful enhancement of Compass.
CaveBase can export numeric values to CaveViewer and the latter displays the
data as symbols of different size. I think it would be also interesting to
export the number itself, i.e. if you entered a temperature of 5.7 degrees
in CaveBase the number should appear in Cave Viewer near the corresponding
station.
Larry, what's your opinion?
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Sat, 19 Aug 2000 21:05:35 -0600 (MDT)
Subject: Re: [compass-users] a CaveBase idea
X-Evolution: 00000044-0010
Roger,
> I think it would be also interesting to
> export the number itself, i.e. if you entered a temperature of 5.7 degrees
> in CaveBase the number should appear in Cave Viewer near the corresponding
> station.
> Larry, what's your opinion?
I liked the idea so much that I went ahead and implemented it. It is up on
the net. Download it and see what you think.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Sun, 20 Aug 2000 21:13:52 +0200 (CEST)
Subject: Re: [compass-users] a CaveBase idea
X-Evolution: 00000046-0010
On Sat, 19 Aug 2000, Larry Fish wrote:
Hello again,
> I liked the idea so much that I went ahead and implemented it. It is up on
> the net. Download it and see what you think.
Guys, that's really hot stuff! :-)
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: "Thomas Oakes"
Date: Thu, 24 Aug 2000 20:43:10 -0000
Subject: [compass-users] CaveX
X-Evolution: 00000048-0010
Hi Larry,
I am brand new to Compass and have been trying to run CaveX but I
keep getting an error message. I am running under win95 ver. 4.0 on a
266 MHz Pentium Pro. I installed DirectX 6.0, rebooted and then
installed CaveX. When I try to run CaveX I get the following series
of error messages:
Unknown error: $80040111
External exception C000001D
Could Not Initialize DirectX, Be Sure that DirectX is Installed
The CaveX window does open but when I click OK on the last error
message the application closes. Is this something you can help with?
Thanks,
Tom Oakes
To: compass-users list
From: Roger Schuster
Date: Mon, 28 Aug 2000 14:11:37 +0200 (CEST)
Subject: [compass-users] x files and MSIE
X-Evolution: 0000004a-0010
Hello Larry,
for a public demonstration of Compass next weekend I prepared my old
computer: fresh Windows 95 installation, DirectX 6.1 core, MSIE 5.0, current
Compass releases and anything goes well with one exception. I can't view the
x files (generated with CaveX) in Internet Explorer.
On my new machine the first time I accessed your live3d cave page MSIE asked
for downloading a specific ActiveX control and then it worked but on the old
computer nothing happened. ActiveX and scripting are enabled but the browser
doesn't download the 3d viewer.
Probably the problem is because my old video card doesn't have hardware 3d
acceleration. The DirectX diagnostics tools says that Direct3D runs only in
software mode. Your CaveX viewer on the other hand doesn't cause trouble.
Any hints or experiences?
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Mon, 28 Aug 2000 12:38:06 -0600 (MDT)
Subject: Re: [compass-users] x files and MSIE
X-Evolution: 0000004c-0010
Roger,
I have a couple of thoughts on the subject. The fact that Internet
Explorer doesn't try to download the 3D viewer probably means that
Explorer thinks that it has already been loaded. My guess is that when you
installed Win95, it left the viewer half installed.
The 3D viewer is a Microsoft product that is not very well documented. I
discovered it by accident when I found that Explorer could display X
files. It took quite a bit of research before I was able to figure out
that the 3D viewing capabilities in Explorer came from an ActiveX
component called XWEB.OCX.
In the process of figuring out how XWEB.OCX worked, I had to install and
uninstall it several times. I did this by renaming the file. When I did
this, Windows decided that XWEB was gone and downloaded a new version when
I connected to my 3D Web Page.
The tricky part is that XWEB.OCX is difficult to find, even if it is on
your computer. This is because Windows doesn't want you messing with
installed components. This means the "Find Files and Folders" from the
Start Menu will not "see" XWEB.OCX. At least that is the case in Windows
98. Windows 95 may allow you to see the file. What I had to do to find it
was go to the DOS prompt and probe around in various hidden directories.
Currently on my machine, XWEB is in:
c:\windows\downloaded program files\xweb.ocx
I think what you need to do is find XWEB and delete or rename it. Then try
reconnecting to the web page. Hopefully, it will download and reinstall
the component.
The tricky part is going to be finding XWEB. I just tried one of the DOS
tools I have called "WHEREIS" and it could not find XWEB. Yet, when if I
go to the DOS prompt and look in the directory, it is there. If you have
trouble finding the file, you might try searching the registry for the
word XWEB. I find several entries for XWEB, and one of them gives me a
directory. In my case, the registry key is:
\HEY_CLASSES_ROOT\CLSID\{970C7E08-05A7-11D0-89AA-00A0C9054129}\InprocServer32
This method of removing and reinstalling XWEB is probably not the correct
way to do it, but it does seem to work. I probably should look around on
the net and find the proper way.
I don't think the problem has to do with lack of graphics acceleration.
DirectX is supposed to work fine without it. It would just be slow.
I'll be interested to know how your demo goes.
Larry
> for a public demonstration of Compass next weekend I prepared my old
> computer: fresh Windows 95 installation, DirectX 6.1 core, MSIE 5.0, current
> Compass releases and anything goes well with one exception. I can't view the
> x files (generated with CaveX) in Internet Explorer.
>
> On my new machine the first time I accessed your live3d cave page MSIE asked
> for downloading a specific ActiveX control and then it worked but on the old
> computer nothing happened. ActiveX and scripting are enabled but the browser
> doesn't download the 3d viewer.
>
> Probably the problem is because my old video card doesn't have hardware 3d
> acceleration. The DirectX diagnostics tools says that Direct3D runs only in
> software mode. Your CaveX viewer on the other hand doesn't cause trouble.
To: compass-users list
From: Roger Schuster
Date: Wed, 6 Sep 2000 21:40:32 +0200 (CEST)
Subject: [compass-users] x to vrml converter
X-Evolution: 0000004e-0010
Hi all,
does anybody know if there is a converter available which can convert
Microsoft x files (which the CaveX Viewer produces as output) to vrml?
Happy caving
Roger
PS: A report about the public presentation of Compass last weekend will
follow soon.
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: "evan anderson"
Date: Fri, 06 Oct 2000 04:52:18 GMT
Subject: [compass-users] Re: File date problems
X-Evolution: 00000054-0010
Roger wrote-
>Probably it *is* a problem with your control panel settings. My own
>experiences with COMPASS (on a box with the German edition of Win 95) >are
>that it uses the "short date" which stores only the last two digits of the
>year by default. This may in some circumstances end up with y2k >problems.
>Go to the control panel and select the regional settings and change the
>"short date" to something with a four digit year. There is a pull-down
> >menu with several choices, also some where the year has all four digits.
>After that you must reboot your PC and fix your cave data manually.
>Hopefully this error will never appear again.
I am running Windows 98 and the control panel displays the date as 2000...
not 00. There is no option to change or manipulate what the year is other
than:
1. Toggle the years up or down one year at a time.
2. Highlight the year and re-write it.
3. Hit the OK, CANCEL, or APPLY buttons at the bottom.
The Editor does not change the saved 2000 year every time. Just randomly
every once in a while.
thanks for the Textpad.com thing. I'm going to try to learn Larry's XEDIT
program first.
later
evan
To: users group compass
From: Larry Fish
Date: Fri, 6 Oct 2000 15:21:55 -0600 (MDT)
Subject: [compass-users] Date Problem
X-Evolution: 00000056-0010
Evan,
I have found the problem! Roger was right, it has to do with the
Windows Regional settings and the way the editor handles the date. I
went through your step-by-step procedure and still could not
duplicate the problem. Finally, I went to the Control Panel and
changed the "Short Date Format" so that only two digits were
displayed for the year and the problem suddenly appeared.
For an immediate quick fix, carry out the following steps:
1. Goto "My Computer," then "Control Panel", finally select "Regional
Settings."
2. Select the "Date" tab and examine the "Short Date Style." The
"Style" string uses letters to specify the number of digits that will
appear in the date. The letter "m" is used to specify the month, "d"
is used to specify the day and "y" is used to specify the the year.
For example, if the style string were "mm/dd/yy" it would specify
that all dates are displayed with a two digit year.
3. Change the "Style" string so it has four "y"s in it. Press the
apply button. Besure to close any COMPASS programs and re-run them so
the new setting can take effect. You may need to reboot the computer
for the changes to take effect.
These instructions work for Win95 or Win98. Window 3.1 has something
similiar, but I don't have it available so I can't give you the
step-by-step procedure.
I have fixed COMPASS so it ignores the year part of the "Style" string.
I have posted the new version on the net, but it is not an official
release.
Check it out and see if it solves the problem for you.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Mon, 09 Oct 2000 09:22:04 +0200
Subject: [compass-users] [Fwd: Spud stuff: other survey software; and a few other bits.]
X-Evolution: 00000058-0010
Hi Compass Users,
this was posted on the Survex mailing list and I think it may be of some
interest to see what other survey software developers and users are
thinking.
Roger
Mark Shinwell wrote:
>
> Hi
>
> I have been browsing around various cave surveying web sites and collated
> some things which might be of use for Spud design. This is the first
> installment of a summary of these things.
>
> Obviously some of these are specific features which do not have direct
> relevance on the fundamental architecture of Spud, but it is useful to
> have some inkling of what other developers are up to.
>
> Two prevelent features seem to be
> (a) integrated editors for the data;
> (b) 3D "tunnel" features for visualising passage shapes.
>
>
> A page for those "fed up with Survex"...
>
> There are issues raised relating to:
> (a) the lack of an editor tailored to entering svx files;
> (b) no depth colouring or surface grids;
> (c) no support for stereo visualisation;
> (d) no 3D tunnel support;
> (e) lack of features like the ones Chasm has for highlighting
> bits of caves, etc.;
> (f) lack of blunder detection;
> (g) lack of print preview;
> (h) difficulty in organising svx files.
>
> (c) is easy for us to do. (b), (e) and (g) we certainly already have
> the technology to do and I believe (f) is covered to some extent.
>
> (h) is an awkward one -- how can one make it easy to do this? Can we
> maybe produce a short document on it, using examples drawn from large
> datasets such as the CUCC Austria one? (See later about "philosophy").
>
> (a) is in hand for Spud.
>
> (d) is more awkward. Persuade Julian to port Tunnel to C++? Dunno :)
>
> There is an interesting comment about a text editor "Aurora" which can
> select columns of text, not just normal blocks. Might be worth
> considering... (then there are all the other questions to do with the
> editor which need sorting out -- eg are there two editors (plain text
> and a more sophisticated one)?).
>
> Compass
> "...originally written in 1979 and ran on a PDP-10..."
>
> I suggest someone endowed with a machine running Windows has a look at
> this... here are a few notes on it anyway.
>
> The main thrust of the program is to allow easy input and easy
> visualisation.
>
> Some of the main features, condensed from
> http://fountainware.com/compass/wdetails.htm:
>
> - Ghastly web site.
> - Project manager. Survey files can participate in more than one
> project (wow :) Whether we need to have something like this I
> don't know; the one thought that occurs to me which might be
> relevant is that we need to figure out an interface to the revision
> control system.
> You can easily arrange and disorganise your survey data by dragging
> and dropping. There is a serious point here, related to the one
> above, about the philosophy of the software. With Survex at the
> moment it's up to the user how to structure their dataset;
> presumably we want to preserve this for Spud? There are of course
> the hierarchies of country/system etc. which Olly was proposing to
> think about.
> - It has a built-in editor, which looks nasty and does things like
> forcing a certain number of decimal places. It has 12-character
> station name restrictions.
> - Blunder detection is incorporated.
> - You can assemble 3D passage shapes as with Tunnel. The whole
> support for this looks quite sophisticated.
> - Eye-candy: fancy viewer with all the twiddles, export of
> fly-through videos and some sort of way of exporting to the
> Web (presumably VRML/Java).
> - Rose graphs and so on are all implemented.
> - Has integration with GIS stuff like ArcView, and can read standard
> DEM (Digital Elevation Map) files for terrain data. We ought to do
> the latter (I will investigate).
> - Context-sensitive help. We need to think about how we will provide
> help.
> - It costs money.
>
> There is also some kind of interface to external database software to
> allow the recording of specific features within the cave. I wonder if
> this could be used for CUCC's Austria dataset to record, for example,
> the position of cave entrances as marked via GPS. Something like this
> could probably be very easily implemented as a plugin which acts as
> an arbiter between the rendering engine and an external database such
> as MySQL, for e
>
> Olly might want to look at
> http://fountainware.com/compass/compart2.htm
> but I suspect he's already seen it :)
>
> It strikes me that Compass has featuritis, but I can't really make
> an informed decision not having tried it out. Can anyone advise on the
> merits or otherwise of this software? The thing is that it has
> hundreds of features in addition to the few I've noted above, and many
> of them would be easily implemented via our plugin architecture.
>
> Does anyone know much about the data model in Compass?
>
> is a bad joke, but cites the
> main problem of Survex as being "poor qualities concerning the
> visualization of survey data".
>
> "home of the best software
> for the study and mapping of caves".
>
> Looks pretty good. One point I notice is that it can understand
> coordinate systems, GPS, etc. (e.g. it allows export of waypoint data
> for entrances). This could be a really nice feature to have, but again,
> it doesn't really affect our underlying architecture (save for the fact
> that we need to make the editors understand commands, etc, implemented
> by the different plug-ins -- but when one considers that the data
> processing engine itself is just a plug-in, this becomes a non-issue).
>
> Has various fancy diagrams like 3D rose diagrams; see the web site.
>
> The tunnel features seem to be given much attention. I seriously think
> we should make sure Spud can do something like this.
>
> It strikes me that WinKarst has the sort of overall features we might
> be aiming for and does seem to be targetting specific useful features
> rather than having lots of little twiddles. Does anyone have any
> experience with it?
>
> The four major advantages which Spud will have over any of these other
> programs is that it will be multi-platform, free, open source and very
> easy to extend; from the look of it a Mac port will be plausible
> (assuming the GTK+ Mac port keeps going OK).
>
> How are we going to approach Spud design? We need to think about overall
> architecture, data model etc. Input from surveyors around the place would
> be very useful... Then we can move onto the nitty gritty of specific
> features within each plugin (which can probably be designed fairly
> independently in some cases and then subjected to peer review). What do
> people think? Olly's mail has set the ball rolling and I'll reply to
> that as soon as I have time.
>
> To be honest I reckon it would be great to organise a meeting in person
> between people who have good ideas for the project, as then a lot more
> can get done at once...
>
> Anyway, that's all for now.
>
> Mark
>
> --
> Mark Shinwell -- President, Cambridge University Caving Club
> Queens' College, Cambridge, UK
> Mail: mrs30 at cam.ac.uk Web: http://mrs30.quns.cam.ac.uk
>
> --
> Survex http://lists.survex.com/mailman/listinfo/survex
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Tue, 10 Oct 2000 03:18:30 -0600 (MDT)
Subject: [compass-users] Survex ideas.
X-Evolution: 0000005a-0010
The information from the Survex mailing was interesting and I wanted
to make a few comments. First, I have a great deal of respect for
Olly the author of Survex. I think that he has done a very nice job
with Survex. In fact, Survex appears to be the only cave survey
program that correctly implements the Least Squares technique for
loop closure. He has been very friendly and helpful. Olly, Wookey,
John Halleck and I have had extensive and very interesting
conversations about loop closures and blunders.
One of the big differences between Survex and COMPASS is the way data
is organized. This seems to be related to the Brittish cave surveying
style. I may not have this exactly right, but their survey stations
don't have unique names. For example, many surveys in the Survex data
I have seen have station names like "1, 2, 3, 4, etc." They organize
their data by sections and subsections, and these names are used to
distinguish the station names between surveys.
Survex has an elegant and sophisticated system for handling this.
COMPASS can do similiar things with its "Links," but it is not really
intended to do the kind of things that Survex needs to do. I don't
know if other parts of the world use the same surveying technique as
the Brittish, but so far, I haven't had any requests to improve the
linking features.
> - Ghastly web site.
Maybe a little chaotic. My frustration with most cave survey web sites
is that you cannot tell very much about the program. My goal has been
to put enough information on the web site so people could really tell
how the program works and what features it has.
> - It has a built-in editor, which looks nasty and does things like
> forcing a certain number of decimal places. It has 12-character
> station name restrictions.
There are two philosophical approaches to cave survey editors. The first
is the "Free Form" editor. With this type of editor, you just enter the
data pretty much any way you want, adding commands at the begining of the
text that define the where and how the data will appear. The best example
I have seen of this type of editor is the one in David MacKenzie's Walls
program.
The second type of editor is the "Spreadsheet" style editor. COMPASS uses
this type of editor. Although you can configure the editor to match the
measurement order and units used, you are forced to enter the information
in specific cells, in a specific way.
I chose the second type of editor because there is more opportunity
for error checking at the point that the data is entered. For
example, with Free Form editors, the data is only checked when the
data is compiled. This means that you can accidently do things like
enter letters in a numeric field, enter contradictory depth gauge
information, etc. With a Free Form editor, these types of errors will
not get caught until the data is compiled.
The 12-character limit can be increased at any point. It is a compromise.
The more characters you have in the station name, the more memory you use
and the slower the processing.
> - Eye-candy: fancy viewer with all the twiddles, export of
> fly-through videos and some sort of way of exporting to the
> Web (presumably VRML/Java).
> It strikes me that Compass has featuritis, but I can't really make
> an informed decision not having tried it out.
This is interesting because most of the features in COMPASS come from
the users. I have yet to hear someone say they want less features.
Usually, they want more and more.
> The four major advantages which Spud will have over any of these other
> programs is that it will be multi-platform, free, open source and very
> easy to extend.
The current version of COMPASS is written in Delphi a Boland product.
Borland is working on a version of Delphi for Linx called Kylix.
Supposedly, you can take a program written for Windows and convert it
directly to Linx. If there really is an interest from cavers for a Linx
version, it may be fairly easy to do.
Larry
To: compass-users at egroups.com
From: Roger Schuster
Date: Tue, 10 Oct 2000 20:27:32 +0200 (CEST)
Subject: Re: [compass-users] Survex ideas.
X-Evolution: 0000005c-0010
On Tue, 10 Oct 2000, Larry Fish wrote:
Hello folks,
> The current version of COMPASS is written in Delphi a Boland product.
> Borland is working on a version of Delphi for Linx called Kylix.
> Supposedly, you can take a program written for Windows and convert it
> directly to Linx. If there really is an interest from cavers for a Linx
> version, it may be fairly easy to do.
Kylix is a "rapid development IDE" for C++ and Delphi on Linux machines. I
don't know how difficult it is to port a Delphi application like Compass to
Linux but I think it is possible. Compass can be used on Linux today with the
help of some DOS and Windows emulators but these are somewhat slow and there
are several bugs. See http://www.karst.net/Compass/complin.htm for details.
Because of the problems with Compass in an emulated environment and the lack
of really good native cave surveying software for Linux I think we should
take it seriously into consideration to port Compass to Linux. I expect some
problems in those parts which make use of OLE and other Windows specific
technologies (e.g. the interface between Cave Base and Cave Viewer) but on
Linux there are good alternatives. AFAIK Kylix uses Qt as GUI toolkit which
means that Kylix applications will fit seamless into KDE and the latter
offers powerful methods for data transfer.
I don't know how much Kylix will cost and what's about the license stuff (Qt
is since version 2.2 under the GNU GPL but only for open source projects and
Compass isn't OSS) but in my opinion Compass for Linux is exactly that thing
which the (under)world is waiting for.
If the porting of Compass becomes reality I am strongly interested in
supporting this effort.
Best regards
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Tue, 10 Oct 2000 14:30:57 -0600 (MDT)
Subject: Re: [compass-users] Survex ideas.
X-Evolution: 0000005e-0010
According to the Delphi web page, applications developed with Delphi
version 5, will port directly to Kylix. COMPASS is now written in Delphi-1
and with portions like CaveX and the DEM Reader written in Delphi-3. The
first step would be to port everything to Delphi-5. This shouldn't be too
much of a problem, although I have written my own memory managers for the
Win3.1 environment and they would have be rewritten. Actually, I have been
planning to upgrade everything to Delphi-5 for a long time because it
would dramatically speed up some applications like the Viewer.
This disadvantange of porting everything to Delphi-5 is it would no longer
work with Win3.1. I don't know how many cavers are still using Win3.1, but
I know of one Colorado caver who is and I suspect there are others. (I
have also heard that people in Eastern European countries are currently
using older computer and old versions of the operating systems.) Still, I
think that most of the lagging cavers could be nudged into the
Win95/98/ME/2000 world.
After that, it would simply be a matter of doing whatever conversion are
necessary to make everything work with Kylix. Roger is correct that
programs like CaveBase would not work because it needs OLE, a Windows
specific feature, to work. Also, CaveX, which uses the Windows only
DirectX, would not work.
Beyond these issues, I have some practical questions. Currently, I barely
have time to keep up with the Windows version of COMPASS. I don't think I
would have time to deal with porting and maintaining a whole other
version. I might be able to do it if both version could be maintained from
the same source code, however, as Roger points out, at the very least,
certain modules would have to be different to deal with the Windows
specific features. In addition, to get maxium preformance, you always want
to take advantage of certain operating system specific features. I would
guess that before too long, the Kylix version would gradually evolve so
that it was very different from the Windows version. Finally, there are
all the bugs, configuration problems, and version specific issues that go
with a different operating system. For example, there are several
different flavors of Linux, each one created by different companies. I
suspect that each version would have its own issues and it would create a
lot of work for me.
Also, I have questions about the viability of Linux in this environment.
Inspite of the fact that Linux is becoming very popular, Windows machines
still dominate. I use Linux and Unix everyday and I'm not sure I see any
advantage that would compel the average computer user to switch. Most
people say that Linux is more stable than Windows, but that has not been
my experience. A number of my friends who are programmer types have loaded
Linux on there system, played with it for a while and then gone back to
Windows or DOS. I think that part of Linux's popularity is a result of
everyone hating Bill Gates and Microsoft and although it is fun to hate
Bill, most cavers are not involved in that fight. My guess is that a few
cavers who are programmers or tinkerers will be using Linux, but the vast
majority will be in the Windows environment.
Larry
To: compass-users at egroups.com
From: "rob harper"
Date: Thu, 12 Oct 2000 23:54:12 GMT
Subject: [compass-users] general questions
X-Evolution: 00000060-0010
Just a couple of quick questions.
1. How do you fix the entrance stations so that more than one cave can be
entered on the same survey and, possibly more useful, their relationship to
each other can be assessed.
2. Regarding Larry's comment that there are still people using Win 3.1.
There are also some people still just using the DOS Version. This is mainly
because it requires much less sophisticated, and hence cheaper, hardware.
Following on from this - has anyone any experience of trying to run Compass
from one of the little palmtop computers?
Cheers,
Rob Harper
To: compass-users at egroups.com
From: Larry Fish
Date: Thu, 12 Oct 2000 22:46:50 -0600 (MDT)
Subject: Re: [compass-users] general questions
X-Evolution: 00000062-0010
Rob,
Thanks for the letter.
> 1. How do you fix the entrance stations so that more than one cave can
> be entered on the same survey and, possibly more useful, their
> relationship to each other can be assessed.
COMPASS works with both single data files and Projects. Fixed stations
have to entered as a part of a Project. If you are just working with
a small cave that you have in a single data file, you have to create a
project that goes with it. Once you have created the project, you can
enter one or more fixed stations. In the case of single caves, you would
probably be fixing the location of the entrance. In the case, multiple
caves or multiple entrances, you may be fixing many more stations.
Remember, survey files in COMPASS can be connected to more than one
Project. For example, in the Williams Canyon area of Colorado, there are
more than 30 caves. As a result, I have several different projects. Some
cover the whole canyon, others cover subsections of the canyon and others
show surface trails and terrain. Many of the caves participate in more
than one Project.
I have attached a complete description of setting up fixed stations at the
end of this letter.
Larry
==========================================================================
Linking Cave Surveys To Fixed Stations Or Real-World Coordinates.
It is sometimes useful to set one or more stations in a cave to a fixed
location. When you do this, all other stations in the cave are positioned
relative to the fixed stations. This is useful for tying together multiple
caves. For example, you can tie the entrances of several caves together
and create plot of the whole system.
It also allows you locate the position of rooms and passages on the
surface using a GPS receiver. Finally, it is useful for making the surveys
more accurate by tying them to benchmarks and precision (theodolite)
surface surveys.
Fixed stations can be applied to any station in the cave. Usually, they
are applied to entrances, but if have a "cave radio" coordinates or fixed
coordinates, you can apply them to station in the interior of the cave.
GEOGRAPHIC REFERENCE SYSTEMS
The fixed location can be based on a geographic reference system such as
UTM and longitude and latitude. COMPASS uses UTM as it standard geographic
measurement unit, but you can also enter locations in Longitude and
Latitude. You can get this information from maps, benchmarks and from GPS
receivers. Once you have connected a cave to a geographic reference point,
all stations and locations within the cave will be relative to this
location. For example, if you tie the entrance of a cave to its UTM
coordinates, location of all passages will be in UTM. This means that you
can find the elevation above sealevel for any passage in the cave. You can
also find location of passage on the surface, by comparing the coordinates
of the passage to the coordinates from a GPS receiver.
LOCAL REFERENCE SYSTEMS
You can also use other coordinate systems. If you are not using Longitude,
Latitude or UTM, the fixed location can be based on the relative distance
between fixed location on the earth and a point in the cave. This type of
fixed location can be anything like a telephone pole, fence post, building
corner, a land surveyors mark, a section corner, a bench mark etc.
FIXED STATIONS AS LOOPS
If you have a cave with more than one fixed location and you "Close" the
cave, the Loop Closer will treat the intermediate stations as a loop. This
means that the Loop Closer will look for errors between the two fixed
station and "adjust" intermediate station to reduce the effect of the
error. Fixed Stations themselves are never adjusted or moved by as the
cave data is processed.
FIXED STATIONS AND PROJECTS
In COMPASS, Fixed Stations can only be entered at the "Project" level. The
advantage of this is that since COMPASS can have more than one Project for
the same data, you can have more than one set of geographic reference
points for a particular cave. For example, you could have one Project
where the cave is tied to a local reference and another where it is tied
to UTM.
STEP-BY-STEP INSTRUCTIONS
1. PROJECT MANAGER. Since Fixed Stations are created at the project level,
you must go to the Project Manager to enter a Fixed Station.
2. CREATING A PROJECT. The Project Manager can work with either individual
".DAT" survey files or ".MAK" project files. Since the fixed stations are
entered at the project level, you must have a Project File before you can
enter a Fixed Station. If you don't have a Project File, go the "Project
Creation Wizard" on the "File" menu of the Project Manager. Just follow
the instructions. If you already have a Project, you can use it.
3. OPENING THE PROJECT. Once you have Project File, open it with the
Project Manager. Remember, this must be a file with the "MAK" extension
like LECH.MAK. You cannot use "DAT" file to create a link. Once you have
opened the Project File, click on the "Expand Tree" button on the tool bar
at the top of the screen. This will expand the "Survey Tree" to show all
the "Nodes."
4. SELECTING SURVEY FILE. You now need to select the survey file that has
the station or stations you want to use as a Fixed Station. Just click on
and highlight the survey file in the "Tree." Now click on the "Edit Node"
button at the top of the screen. This will display the Node Editing dialog
box. Select the "Links Tab" at the bottom of the page. This will take you
to the Links Editor.
5. ENTERING LINKS. The Links Editor displays a grid that allows you to
enter the station name and the North, East and Vertical coordinates of the
station. Enter the coordinates of the station in the cells to the right of
the station name. The units can be feet or meters as selected by the check
boxes below. As an example:
A1 10.1 20.2 30.3
This illustrates a fixed station. The station A1 is fixed to the location
10.1 east, 20.2 north and 30.3 vertical from the fixed reference point.
The East, North and Vertical coordinates can be in either feet or meters.
The coordinates can be relative a local coordinate system or a world-wide
geographic reference system. COMPASS uses UTM as its internal reference
system, however it can convert to and from longitude and latitude. To use
geographic referenced coordinates, enable the "Use UTM" checkbox. If you
are using UTM, be sure to set a "Zone" number. Only one zone is allowed
for each cave file, but you can have different files in different zones.
You can enter up to 99 Links for each survey file.
If you want to enter the data in Longitude and Latitude, first highlight
the row in the grid for the station you want to set. Then, click on the
"Georeference" button. The program will then display a dialog box that
will allow you to convert back and forth between Longitude and Latitude
and UTM. At the bottom of the box will be the UTM coordinates in meters.
(Note: UTM is always displayed in meters on maps.) If you press the "UTM
to Long/Lat" button, the UTM coordinates will be converted to longitude
and latitude. Like wise, if press the "Long/Lat to UTM" button, the
longitude/latitude values will be converted to UTM. This way, you can
enter longitude and latitude values. You can also use the dialog to verify
you UTM coordinates by converting them to longitude and latitude and then
checking the map.
6. OTHER USES FOR LINKS. Links can also be used to "link" together
different caves or different subsections of caves. You can create this
type of link by entering the station name without the coordinates. In this
situation, the links tell COMPASS which stations are the connecting
stations between survey files. COMPASS then discards all other stations
except the links.
This kind of link has two purposes. First, when you link together two
separate caves, it is possible to have duplicate station names. Since
COMPASS discards all station except links, duplicate stations do not
interfere with each other. Second, because the majority of stations are
discarded between files, this kind of link saves memory. This was
important in the DOS environment, where memory was limited. Under Windows,
memory is virtually unlimited, so there is little need to conserve memory.
7. VIEWING GEO-REFERENCED COORDINATES. Once you have created a project
with geo-referenced coordinates, all surveys station in the cave are
relative to the fixed stations. You can view these locations in the
Viewer. If press the "Label Elevations" button on the Viewer Tool Bar, the
elevation will be relative to fixed stations. If you have set the
elevation of the fixed stations to the altitude above sea level, all the
station elevations will be displayed as their elevation above sea level.
You can also find any East, North or Vertical coordinate in or around the
cave using the Measuring Tool. Just select "Tools|Measure
Distances/Angles" from the menu bar.
To: compass-users at egroups.com
From: Christian Docksey
Date: Fri, 13 Oct 2000 11:29:17 -0400
Subject: Re: [compass-users] general questions
X-Evolution: 00000064-0010
Hello Larry,
First let me say that it is really nice to be able to read and write
with the programmer himself. To get answers, problems solved, doubts
explained, etc, is really fantastic.
For my few queries:
1) Your previous reply: "This means that you
can find the elevation above sealevel for any passage in the cave. You
can
also find location of passage on the surface, by comparing the
coordinates
of the passage to the coordinates from a GPS receiver."
Can you make compass give the UTM (and/or lat/long)coordinates of any
particular station in a cave once it has been linked to a geographical
reference system? If one could get this readout, the advantages are
enormous! We many times discover a cave, and want to see if it connects
to another cave nearby. Also, when a chimney is found in a cave, one
could look for a possible connection to the surface by looking on the
surface with GPS coordinates given by "Compass". Anyway, you can imagine
the benefits of this. If "Compass" can't now, would it be possible in
next version??
2)Since trying to keep "Compass" for windows compatible with DOS and 3.1
limits the power of compass in many ways (and its expansion), wouldn't
you consider separating the versions; just as you have a DOS version and
a windows version, you would have a DOS, a 3.1 and a 98/2000 version
which would permit all sorts of possibilities. As a note to this, would
you find it easier to use windows elements to simplify your work and
mantain a known interface. I mean it in the sense that you wouldn't have
to create buttons, menus, etc, because you would use those of windows. I
think this is what Winkarst has done.
3)I agree that the information section in the printout should be more
extensive. I know you want to make a nice feature of this. But maybe in
the meanwhile one could have the option of having, say, 6 to 8 cells in
which one could type in any info one wishes (as text). No interpretation
necessary by "Compass". Maybe this info could be put in at the printing
command stage. Anyway,to me, more important than this is to be able to
have a more sophisticated North arrow. Perhaps larger and not in its
current position, but more within the map drawn by compass. Ideally, it
could be positioned where one wanted by using a menu maybe. Same goes
with that extra info. Too complicated??
Also very necessary, is to have a scale bar outputted, whether printed
out or exported. I know you have the scale info (say 1:320) but to have
the scale bar (as in Winkarst) printed or exported allows one to scale
the whole map in another application until it fits the desired size. The
scale bar would change size along with the cave map, allowing a precise
scaling info to be always available. Please don't think I'm comparing
Compass to Winkarst. It's just that it does have some nice features too
(as many progs have) and why not take advantage of some of these ideas?
I have one or two other suggestions and ideas, but I think you may say,
maybe this is enough for one day!
Once again let me congratulate you for your work on this software and
let us hope that it becomes the undoubted software of choice for
everyone.
warm regards...
Christian
Larry Fish wrote:
>
> Rob,
>
> Thanks for the letter.
>
> > 1. How do you fix the entrance stations so that more than one cave
> can
> > be entered on the same survey and, possibly more useful, their
> > relationship to each other can be assessed.
>
> COMPASS works with both single data files and Projects. Fixed stations
>
> have to entered as a part of a Project. If you are just working with
> a small cave that you have in a single data file, you have to create a
>
> project that goes with it. Once you have created the project, you can
> enter one or more fixed stations. In the case of single caves, you
> would
> probably be fixing the location of the entrance. In the case, multiple
>
> caves or multiple entrances, you may be fixing many more stations.
>
> Remember, survey files in COMPASS can be connected to more than one
> Project. For example, in the Williams Canyon area of Colorado, there
> are
> more than 30 caves. As a result, I have several different projects.
> Some
> cover the whole canyon, others cover subsections of the canyon and
> others
> show surface trails and terrain. Many of the caves participate in more
>
> than one Project.
>
> I have attached a complete description of setting up fixed stations at
> the
> end of this letter.
>
> Larry
>
>
> =========================================================================
>
> Linking Cave Surveys To Fixed Stations Or Real-World Coordinates.
>
> It is sometimes useful to set one or more stations in a cave to a
> fixed
> location. When you do this, all other stations in the cave are
> positioned
> relative to the fixed stations. This is useful for tying together
> multiple
> caves. For example, you can tie the entrances of several caves
> together
> and create plot of the whole system.
>
> It also allows you locate the position of rooms and passages on the
> surface using a GPS receiver. Finally, it is useful for making the
> surveys
> more accurate by tying them to benchmarks and precision (theodolite)
> surface surveys.
>
> Fixed stations can be applied to any station in the cave. Usually,
> they
> are applied to entrances, but if have a "cave radio" coordinates or
> fixed
> coordinates, you can apply them to station in the interior of the
> cave.
>
> GEOGRAPHIC REFERENCE SYSTEMS
>
> The fixed location can be based on a geographic reference system such
> as
> UTM and longitude and latitude. COMPASS uses UTM as it standard
> geographic
> measurement unit, but you can also enter locations in Longitude and
> Latitude. You can get this information from maps, benchmarks and from
> GPS
> receivers. Once you have connected a cave to a geographic reference
> point,
> all stations and locations within the cave will be relative to this
> location. For example, if you tie the entrance of a cave to its UTM
> coordinates, location of all passages will be in UTM. This means that
> you
> can find the elevation above sealevel for any passage in the cave. You
> can
> also find location of passage on the surface, by comparing the
> coordinates
> of the passage to the coordinates from a GPS receiver.
>
> LOCAL REFERENCE SYSTEMS
>
> You can also use other coordinate systems. If you are not using
> Longitude,
> Latitude or UTM, the fixed location can be based on the relative
> distance
> between fixed location on the earth and a point in the cave. This type
> of
> fixed location can be anything like a telephone pole, fence post,
> building
> corner, a land surveyors mark, a section corner, a bench mark etc.
>
> FIXED STATIONS AS LOOPS
>
> If you have a cave with more than one fixed location and you "Close"
> the
> cave, the Loop Closer will treat the intermediate stations as a loop.
> This
> means that the Loop Closer will look for errors between the two fixed
> station and "adjust" intermediate station to reduce the effect of the
> error. Fixed Stations themselves are never adjusted or moved by as the
>
> cave data is processed.
>
> FIXED STATIONS AND PROJECTS
>
> In COMPASS, Fixed Stations can only be entered at the "Project" level.
> The
> advantage of this is that since COMPASS can have more than one Project
> for
> the same data, you can have more than one set of geographic reference
> points for a particular cave. For example, you could have one Project
> where the cave is tied to a local reference and another where it is
> tied
> to UTM.
>
> STEP-BY-STEP INSTRUCTIONS
>
> 1. PROJECT MANAGER. Since Fixed Stations are created at the project
> level,
> you must go to the Project Manager to enter a Fixed Station.
>
> 2. CREATING A PROJECT. The Project Manager can work with either
> individual
> ".DAT" survey files or ".MAK" project files. Since the fixed stations
> are
> entered at the project level, you must have a Project File before you
> can
> enter a Fixed Station. If you don't have a Project File, go the
> "Project
> Creation Wizard" on the "File" menu of the Project Manager. Just
> follow
> the instructions. If you already have a Project, you can use it.
>
> 3. OPENING THE PROJECT. Once you have Project File, open it with the
> Project Manager. Remember, this must be a file with the "MAK"
> extension
> like LECH.MAK. You cannot use "DAT" file to create a link. Once you
> have
> opened the Project File, click on the "Expand Tree" button on the tool
> bar
> at the top of the screen. This will expand the "Survey Tree" to show
> all
> the "Nodes."
>
> 4. SELECTING SURVEY FILE. You now need to select the survey file that
> has
> the station or stations you want to use as a Fixed Station. Just click
> on
> and highlight the survey file in the "Tree." Now click on the "Edit
> Node"
> button at the top of the screen. This will display the Node Editing
> dialog
> box. Select the "Links Tab" at the bottom of the page. This will take
> you
> to the Links Editor.
>
> 5. ENTERING LINKS. The Links Editor displays a grid that allows you to
>
> enter the station name and the North, East and Vertical coordinates of
> the
> station. Enter the coordinates of the station in the cells to the
> right of
> the station name. The units can be feet or meters as selected by the
> check
> boxes below. As an example:
>
> A1 10.1 20.2 30.3
>
> This illustrates a fixed station. The station A1 is fixed to the
> location
> 10.1 east, 20.2 north and 30.3 vertical from the fixed reference
> point.
>
> The East, North and Vertical coordinates can be in either feet or
> meters.
> The coordinates can be relative a local coordinate system or a
> world-wide
> geographic reference system. COMPASS uses UTM as its internal
> reference
> system, however it can convert to and from longitude and latitude. To
> use
> geographic referenced coordinates, enable the "Use UTM" checkbox. If
> you
> are using UTM, be sure to set a "Zone" number. Only one zone is
> allowed
> for each cave file, but you can have different files in different
> zones.
> You can enter up to 99 Links for each survey file.
>
> If you want to enter the data in Longitude and Latitude, first
> highlight
> the row in the grid for the station you want to set. Then, click on
> the
> "Georeference" button. The program will then display a dialog box that
>
> will allow you to convert back and forth between Longitude and
> Latitude
> and UTM. At the bottom of the box will be the UTM coordinates in
> meters.
> (Note: UTM is always displayed in meters on maps.) If you press the
> "UTM
> to Long/Lat" button, the UTM coordinates will be converted to
> longitude
> and latitude. Like wise, if press the "Long/Lat to UTM" button, the
> longitude/latitude values will be converted to UTM. This way, you can
> enter longitude and latitude values. You can also use the dialog to
> verify
> you UTM coordinates by converting them to longitude and latitude and
> then
> checking the map.
>
> 6. OTHER USES FOR LINKS. Links can also be used to "link" together
> different caves or different subsections of caves. You can create this
>
> type of link by entering the station name without the coordinates. In
> this
> situation, the links tell COMPASS which stations are the connecting
> stations between survey files. COMPASS then discards all other
> stations
> except the links.
>
> This kind of link has two purposes. First, when you link together two
> separate caves, it is possible to have duplicate station names. Since
> COMPASS discards all station except links, duplicate stations do not
> interfere with each other. Second, because the majority of stations
> are
> discarded between files, this kind of link saves memory. This was
> important in the DOS environment, where memory was limited. Under
> Windows,
> memory is virtually unlimited, so there is little need to conserve
> memory.
>
> 7. VIEWING GEO-REFERENCED COORDINATES. Once you have created a project
>
> with geo-referenced coordinates, all surveys station in the cave are
> relative to the fixed stations. You can view these locations in the
> Viewer. If press the "Label Elevations" button on the Viewer Tool Bar,
> the
> elevation will be relative to fixed stations. If you have set the
> elevation of the fixed stations to the altitude above sea level, all
> the
> station elevations will be displayed as their elevation above sea
> level.
> You can also find any East, North or Vertical coordinate in or around
> the
> cave using the Measuring Tool. Just select "Tools|Measure
> Distances/Angles" from the menu bar.
>
>
>
To: compass-users at egroups.com
From: Roger Schuster
Date: Fri, 13 Oct 2000 20:32:26 +0200 (CEST)
Subject: Re: [compass-users] general questions
X-Evolution: 00000066-0010
On Thu, 12 Oct 2000, rob harper wrote:
Hello Rob,
nice to meet you here!
> Following on from this - has anyone any experience of trying to run Compass
> from one of the little palmtop computers?
No, I don't have any experiences with palmtop computers but one suggestion I
can offer.
Microsoft made a special OS for palmtops, called "Windows CE". This product
was a desaster because it was too bulky and slow for little palmtop PCs and
only useful on machines with colour screens. Anyway, Compass will probably
run on Windows CE but I'm not sure about this because the hardware of
palmtops differs from that of common desktop machines.
Some weeks ago on our national caver convention one guy presented a
surveying software with the name "Auriga" which runs on a Palm Pilot and has
a connector kit to an electronic compass and clinometer. You can also enter
the data manually. Auriga can export its data in a Toporobot compatible
format which can be converted to Compass. Auriga also has a home page but I
don't have an URL at hand. The author of Auriga is Martin Melzer or Metzler
(or so) so you may find the web page with the help of a search engine.
Lucky caving!
Roger
--
Roger Schuster eMail roger at r-schuster.de
Cavepage http://www.karst.net
To: compass-users at egroups.com
From: Larry Fish
Date: Sat, 14 Oct 2000 03:41:42 -0600 (MDT)
Subject: Re: [compass-users] general questions
X-Evolution: 00000068-0010
Christian,
> First let me say that it is really nice to be able to read and write
> with the programmer himself. To get answers, problems solved, doubts
> explained, etc, is really fantastic.
Actually, it works out very nicely for me. I usually get several
COMPASS questions a week. Before the Mailing List, only me and the
questioner got to see the question and the answer. Now everybody gets
to see them and they are archived so people who come along later can
benifit from the information.
> Can you make compass give the UTM (and/or lat/long)coordinates of any
> particular station in a cave once it has been linked to a
> geographical reference system? If one could get this readout, the
> advantages are enormous!
Unless I don't understand your suggestion, you can already do this for
UTM. If you fix one or more stations to UTM or Long/Lat, the location
of each station in the cave is available in UTM.
The UTM information is available in two places. The first place is in
the Compiler. (From the Project Manager, you press the "Compile Only
(Blunder/Stats)" button.) The information is under the "View|Cave
Statistics" menu option in the Compiler. Here is a sample of the UTM
data for Fulford Cave:
Station COORDINATES:
Index Station Parent east north vertical
===============================================================
0 A1 Ent 359000.000m. 4372000.000m. 3048.000m.
1 A2 A1 359005.238m. 4372002.612m. 3044.888m.
2 A3 A2 359007.088m. 4372009.283m. 3042.771m.
The other place you can find UTM coordinates is in the Viewer. If you
have a Geo-referenced cave, the measurements in the "Tools|Measure
Distance/Angles" will be in UTM.
COMPASS uses UTM as the standard for geo-referenced coordinates. If
you enter coordinates in Lat/Long, they are converted to UTM. The
reason for this is that Lat/Long is a spherical coordinate system and
inorder to plot the cave, everything has to be in cartesian
coordinates. I suppose COMPASS could carry its measurements in
Long/Lat, but conversion between Long/Lat is very complicated and
slow. This is because the earth's surface is not a true sphere, but
more like a lumpy pear-shaped object. This means that it takes a
couple hundred lines of computer code to do the conversion. The
result is that the conversion is too slow to do on-the-fly while
plotting or doing anything that requires speed.
> 2) Since trying to keep "Compass" for windows compatible with DOS and
> 3.1 limits the power of compass in many ways (and its expansion),
> wouldn't you consider separating the versions; just as you have a DOS
> version and a windows version, you would have a DOS, a 3.1 and a
> 98/2000 version which would permit all sorts of possibilities.
The general plan is to eventually move everthing over to the 32 bit
environment without maintaining two versions. I might freeze the 16
bit version and leave it available for anyone who can't live without
Win3.1. The DEM Reader and CaveX are already 32 bit applications and
the other would be easy to do, mostly just a few keystrokes. However
there is really not that much advantage for most programs in 32 bit
environment. It wouldn't make much difference for the Project
Manager, Compiler, Loop Closer, and Editor. The Viewer is the only
program that would be enhanced using 32 bits. I figure that the
drawing speed could be 5 to 10 times faster.
> 3) I agree that the information section in the printout should be
> more extensive. I know you want to make a nice feature of this. But
> maybe in the meanwhile one could have the option of having, say, 6 to
> 8 cells in which one could type in any info one wishes (as text). No
> interpretation necessary by "Compass".
I wish there was a way to do this with "no interpretion necessary."
Things always sound simple when you describe a feature like
this, but in practice they get pretty complicated. First, nobody is
going to be satisfied without having at least a few options. For example,
everyone is going to want to be able to choose the fonts, colors and
sizes. If the text is going to go in a "cell," the cell will have to
be resizable and the text will have to be able wrap so it fills the
cell. The best way to do this is with the mouse like you can do in a
drawing program like Publisher