Thread View: sci.comp-aided
4 messages
4 total messages
Started by Ching-ju Lee
Mon, 06 Apr 1992 18:28
Contour lines
Author: Ching-ju Lee
Date: Mon, 06 Apr 1992 18:28
Date: Mon, 06 Apr 1992 18:28
31 lines
1640 bytes
1640 bytes
Newsgroups: sci.engr.civil, sci.engr, sci.math
Subject: Contour lines
Summary: References needed
Expires:
References:
Sender: Ching-ju Lee
Followup-To: leecj@cs.rpi.edu
Organization: Rensselaer Polytechnic Institute, Troy NY
Keywords:
Dear Netters:
I need recent(1980 or later) references on finding contour
lines(level sets) given the function in a closed functional
form, or only a set of function values on a grid. Most of
the algorithms I could find were published in the 1970's,
and they all concerned with issues like pen plotter move-
ment time, or virtual memory required. But now that we
have graphic output device other than pen plotters, and
supercomputers with enormous virtual memory and multi-
processors available, I wonder if there are any new approaches
that are designed to take advantage of these features.
I also wonder if there are new ideas in dealing with labeling
different branches of a level set. The ones I know of seem
to be too slow. Accuracy of the contour lines is also very
important to me. So far I only know methods which use
linear interpolations. References on higher order algorithms
are also very helpful. Please send all follow-ups through
e-mail to leecj@cs.rpi.edu. I appreciate your help very much.
Ching-ju Lee
Re: Contour lines
Author: Pat Fitzpatrick
Date: Tue, 07 Apr 1992 21:57
Date: Tue, 07 Apr 1992 21:57
24 lines
1821 bytes
1821 bytes
I, too, would like to find out what the latest scoop is in
contouring packages. Meteorologists use lots of data that is
usually contoured to deduce the meaning of the numbers. NCAR graphics
is typically used, but I have developed a strong dislike for this
software, even though it is public domain (NCAR graphics seems to be
just "thrown" together and is hard to learn because it is poorly
documented).
Any comments on this are appreciated. I'm sure I speak for lots of
scientists.
***********************************************************************
Pat Fitzpatrick (____) /\
* * * / \
Said one meteorologist * * * / \
to another: "I was afraid * * * / The \
of this when we started / Rocky \
mixing female hurricanes / Mountain \ / \
with male hurricanes. / Coon-ass \ CSU | |
Now there's lots of little _________________________________________
hurricanes popping up everywhere." | E-mail kjun@typhoon.atmos.colostate.edu |
-----------------------------------------
***********************************************************************
Re: Contour lines
Author: Charles A. Lind
Date: Wed, 08 Apr 1992 12:17
Date: Wed, 08 Apr 1992 12:17
59 lines
3315 bytes
3315 bytes
In article <1992Apr07.215737.53397@yuma.acns.colostate.edu>, kjun@typhoon.atmos.colostate.edu (Pat Fitzpatrick) writes: >I, too, would like to find out what the latest scoop is in >contouring packages. Meteorologists use lots of data that is >usually contoured to deduce the meaning of the numbers. NCAR graphics >is typically used, but I have developed a strong dislike for this >software, even though it is public domain (NCAR graphics seems to be >just "thrown" together and is hard to learn because it is poorly >documented). > >Any comments on this are appreciated. I'm sure I speak for lots of >scientists. > >*********************************************************************** > >Pat Fitzpatrick (____) /\ > * * * / \ >Said one meteorologist * * * / \ >to another: "I was afraid * * * / The \ >of this when we started / Rocky \ >mixing female hurricanes / Mountain \ / \ >with male hurricanes. / Coon-ass \ CSU | | >Now there's lots of little _________________________________________ >hurricanes popping up everywhere." | E-mail kjun@typhoon.atmos.colostate.edu | > ----------------------------------------- >*********************************************************************** My work centers around CFD (Computational Fluid Dynamics) so contouring is not only necessary but critical to my work. I have used IDL and Tecplot on the Unix system and Surface III, MatLab, and Deltagraph on the MAC. By far the easiest to use and most powerful is Tecplot. In our research group we started with 1 Tecplot license and within 1 month we just about abandened (sp?) idl and quickly purchased 2 more licenses (that 3 for a group of 10). You can allow Tecplot to choose which lines to contour or you can point and click the location of the contours. You are able to control just about as many parameters as you could think of when it comes to countour plotting, but that is not all it does. It will do lines, meshes, flooded and lined contours, vector, scatter , and composite plots. It will also allow you to automate text and geometry insertan, and to save styles. I could go on. If you have any particular questions I would be happy to entertain them. By the way, I am not associated with Tecplot except that I am a happy user. Their number is 1-800-676-7568 and they do have demos. PS Tecplot will work on Unix, Iris, IBM PC, VAX, DEC, and other machines. l8r. Charles ___________________________ Charles Lind Dept. of Aerospace Eng. University of MD College Park, MD 20742
Re: Contour lines
Author: John Prentice
Date: Wed, 08 Apr 1992 20:00
Date: Wed, 08 Apr 1992 20:00
28 lines
1855 bytes
1855 bytes
In article <bw1tvcl@rpi.edu> leecj@vega.cs.rpi.edu (Ching-ju Lee) writes: > >I need recent(1980 or later) references on finding contour >lines(level sets) given the function in a closed functional >form, or only a set of function values on a grid. > I would suggest looking in the ACM Transactions on Mathematical Software over the last 12 years or so. There have been numerous really excellent contour algorithms written up in there, including TRICP and its offspring. The code for these algorithms is available free over NETLIB. I have used the TRICP algorithm extensively over the last many years for producing contour plots of stress, density, etc... in solid impact dynamics problems. It has proven both accurate, affordable, and amazingly rugged. There have been numerous articles in more recent years that seem even better, including algorithms for three dimensional plotting. I would also suggest looking at what is available from GNU and from the National Center for Supercomputing Applications. Both have contour packages available. I don't know much about GNUplot, but the NCSA stuff is pretty high class. John -- John K. Prentice | "We must make good the demands of Department of Physics And Astronomy | reason and create a life worthy of University of New Mexico | ourselves and of the goals we only john@aquarius.unm.edu | dimly perceive." - Andrei Sakharov
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads