From swampler@noao.edu (Steve Wampler)
Subject: Responses: Khoros vs AVS
Message-ID: <1993Jun1.224026.19922@noao.edu>
Sender: news@noao.edu
Nntp-Posting-Host: orpheus.tuc.noao.edu
Organization: National Optical Astronomy Observatories, Tucson, AZ, USA
Date: Tue, 1 Jun 1993 22:40:26 GMT
Lines: 50

A while ago I posted a request for comparisons of Khoros and AVS.
There were not many responses (other than people asking for the
results!), so what follows is my understanding of the differences.
I'm hoping this will prompt others to contribute (best to email
me and I'll collect and summarize...)


------------------------
Khoros is a complete software development environment, with
visual programming, GUI building, code generators, source-code
configuration managers, and so on.).  There are 3D visual
toolboxes available for use with Khoros.  Khoros is designed
to permit integration of unrelated components.

AVS is a visual programming system with a 3D emphasis.

As such, AVS can be viewed as a subset of Khoros.
------------------------
------------------------
Both (now) are commercially supported.  Khoros software is free, AVS
is reasonably priced.
------------------------
The user community for Khoros is at least twice as large as that for
AVS, with strong support available (for free) over the Internet.
------------------------
Both support distributed processing, but Khoros' cost (free) is
considerably less restrictive than the cost of distributed licensing
of AVS.
------------------------
Khoros 2.0 provides for process control support.  This is not provided
for in AVS.
------------------------
Khoros has code freely available, allowing you to customize/fix things without
requiring service or maintenance agreements.
------------------------
AVS has better facilities for 'upstream' communication between modules
(eg. a display module feeding mouse locations back to an input module
(say)).
------------------------



As you can see, there is a a one-sided-ness to the responses I've gotten.
Since I'm *really* not trying to start a dog-fight, let me ask that people
email me their comments!  If anything interesting comes along, I'll post
a second summary...
-- 
Steve Wampler
swampler@noao.edu
Gemini Project (under AURA)


From buhle@wharton.upenn.edu
Newsgroups: comp.graphics.avs
Subject: netCDF&others + AVS
Message-ID: <1993Jun1.211916.1@wharton.upenn.edu>
Date: 2 Jun 93 02:19:16 GMT
Sender: news@netnews.upenn.edu
Organization: University of Pennsylvania
Lines: 22
Nntp-Posting-Host: wilma.wharton.upenn.edu

For those of you attending the AVS '93 Conference in Orlando and were amused by
three panelists in the Data Management Panel mentioning the "netCDF public 
domain" software . . . I would like to direct you to an article in the May/June 
1993 issue of "Computers in Physics", pp. 304-308, entitled: "Software for 
Portable Scientific Data Management". This article discusses "...the general 
philosophy of portable scientific-data-management systems and compare four of 
the systems that address these kinds of problems: CDF, HDF, netCDF, and PDB."

The ftp addresses of these software packages (they are all PD) are listed in 
the back. I am biased towards netCDF. The authors of this paper do a pretty 
good job of discussing the general aspects of the packages . . . it is a good 
place to start.

Note: netCDF is available on the INTERNET from unidata.ucar.edu in file:
     pub/netcdf/netcdf.tar.Z    (this will be softlinked to V2.3, April '93)

Yes, it runs on Ultrix, SGI, AIX, Stardent-Titan (let me know if you need the 
Fortran jackets routines), VMS, MSDOS, etc....

Dr. E. Loren Buhle, Jr.  INTERNET: BUHLE@XRT.UPENN.EDU
University of Pennsylvania School of Medicine         Phone: 215-662-3084
Rm 440A, 3401 Walnut St., Philadelphia, PA 19104-6228   FAX: 215-349-5978


From reedt@ece.ucdavis.edu (Todd Reed)
Subject: Stardent 2000 disk array
Message-ID: <C80652.65n@ucdavis.edu>
Originator: reedt@madrone.eecs.ucdavis.edu
Sender: usenet@ucdavis.edu (News Administrator)
Nntp-Posting-Host: madrone.eecs.ucdavis.edu
Organization: U.C. Davis - Department of Electrical Engineering and Computer Science
Date: Wed, 2 Jun 1993 16:36:37 GMT
Lines: 19


I am currently in the process of trying to bring up a Stardent 2000, which
was recently donated to us. Unfortunately, we are a bit short of documentation,
and the company which produced these machines (as many of you no doubt know)
no longer does hardware.

My problem is that there is a disk array in a rack separate from the system
unit, which must be cabled to the unit. There are four cables which connect 
the two. The cables are hardwired to the disk array, but there are eight
identical connectors (arranged in two rows of four each) on the back of the 
system unit, to which the cables could be connected. The cables are numbered 
I,II, III, and IV, the connectors J13 though J20. 

I would very much appreciate hearing from anyone who can shed some light on
this problem. Please contact me by email at trreed@ucdavis.edu, if you can.

With many thanks in advance,
Todd R. Reed
Center for Image Processing and Integrated Computing


From rao@autsrv.iitsg.nrc.ca (Rao Palacharla)
Subject: paper from AVS'93
Message-ID: <1993Jun2.185347.15502@nrcnet0.nrc.ca>
Originator: rao@autsrv.iitsg.nrc.ca
Sender: root@nrcnet0.nrc.ca (Operator)
Nntp-Posting-Host: autsrv.iitsg.nrc.ca
Organization: National Research Council, Ottawa, Canada
Date: Wed, 2 Jun 93 18:53:47 GMT
Lines: 15

I am looking for a copy of the following paper which was presented at
AVS'93. 

Pascal Landi, "Optical device modeling: debugging and visualizing with
AVS", Paper No. TL3-4. Pascal Landi is with Hoechst Celanese Corporation,
Summit, NJ.

I would appreciate if anyone who has the conference proceedings
can fax the paper or mail it to me.

My fax number is 613-952-0215.

thanx in advance
paparao



From jboyle@med.unc.edu (Jim Boyle)
Newsgroups: comp.graphics.avs
Subject: How to AVS MRI??
Date: 3 Jun 1993 01:15:51 GMT
Organization: UNC-CH School of Medicine
Lines: 16
Distribution: world
Message-ID: <1ujjc7$2iv@samba.oit.unc.edu>
NNTP-Posting-Host: dalton.med.unc.edu
Originator: jboyle@dalton

Hello AVSers,

I'd like to try to get an MRI data set I have into AVS.  Anyone know
how?? The format is each slice is 256x256 pixels, each of unsigned int
variety.  Slices are 3mm and inplane resolution is .78125 mm.  Memory
layout is 
	[column,row,slice]

	[0,0,0] [1,0,0] .....[255,0,0] [0,1,0] .....[255,255,0] [0,0,1]...

or as memory addresses increase we first increment column, then row, then
slice.  Any suggestions on how to get AVS to read it??

Thanks,

Jim Boyle


From fsmarc@tristero.lerc.nasa.gov (Marc Cooper)
Newsgroups: comp.graphics.avs
Subject: Problems compiling contour_field
Date: 4 Jun 1993 17:24:31 GMT
Organization: NASA Lewis Research Center [Cleveland, Ohio]
Lines: 27
Message-ID: <1uo0gf$cp0@eagle.lerc.nasa.gov>
NNTP-Posting-Host: tristero.lerc.nasa.gov


After dowloading the contour_field module and its libraries form the avs center
I went to compile them on ouyr SGI 480 running IRIX 4.0.5

The libraries seem to do themselves up with no problem, (but a lot of warnings).

I modified the sample AVS Makefile to find the libraries and made it up.
It failed on the final link, with the following:

Undefined:
TRIANGULAR_INTERP
i_nint
*** Error code 1


Now, I've grepped the libraries AND the source for these fellows and can't
seem to find them.  Anyone got any suggestions?


-- 
Marc Cooper - Graphics Programmer - Sverdrup Tech.| "Misfeasance, malfeasance,
fsmarc@lerc.nasa.gov                              |  nonfeasance, and mopery  
NASA Lewis Research Center  MS 5-11               |  with intent to gawk." 
21000 Brookpark Dr.                               |  
Cleveland, OH  44135               (216) 433-8898 |               -Mike Royko
-------------------------------------------------------------------------------
                Disclaimer:  "It's mine! All mine!"  -D. Duck


From m.s.goodrich@larc.nasa.gov (Mike Goodrich)
Newsgroups: comp.graphics.avs
Subject: Problems with ipt_warp?
Date: 4 Jun 1993 19:31:20 GMT
Organization: NASA LaRC
Lines: 21
Distribution: world
Message-ID: <1uo7u8INNnhm@rave.larc.nasa.gov>
Reply-To: m.s.goodrich@larc.nasa.gov
NNTP-Posting-Host: lynx.larc.nasa.gov


Greetings:
	This is an inquiry about possible bugs in the ipt_warp module.
It seems to exhibit a historisis where the warped image looks incorrect, then 
forcing the module to rexecute
(after its initial execution in the network) by changing its parameter
(e.g. point to bicubic) gives the correct display, and going back to the 
original parameter setting gives the display you were originally anticipating.
Is this known bug? Thanx...


-Mike


________________________________________________________________________________
Mike Goodrich						Phone: (804)864-1790
NASA/LaRC, M.S. 490					FAX:   (804)864-7975
Hampton, VA  23665				m.s.goodrich@larc.nasa.gov





From glennd@george.arc.nasa.gov (Glenn Deardorff)
Subject: Question on Geometry API
Message-ID: <1993Jun4.195715.13871@news.arc.nasa.gov>
Sender: glennd@pioneer.arc.nasa.gov (Glen Deardorff GDP)
Organization: NASA Ames Res. Ctr. Mtn Vw CA 94035
Date: Fri, 4 Jun 1993 19:57:15 GMT
Lines: 17

I am presently working on an interpreter to convert our inhouse
graphics metafile format (ARCGraph) to AVS geometries.  

A lot of the geometry types (such as polyhedra or spheres) can be
defined as assemblages all at once.  For instance, one
GEOMcreate_sphere call can be used to define many spheres.

My question is:  Will this be "visibly" quicker (perform significantly
faster) than if I were to issue a seperate GEOMcreate_sphere call
for each sphere?  Or does this "multiple-element-at-once" feature
exist mainly for programming convenience?  My question pertains to all
geometry types for which this feature exists, not just spheres.

Thanks in advance.

- Glenn



From ringer@spf.trw.com (Brian Ringer)
Subject: Canny Edge Detection Module???
Message-ID: <2C1383DA.5E75@deneva.sdd.trw.com>
Keywords: Canny Edge Detector
Sender: news@deneva.sdd.trw.com
Organization: TRW Sensor Data Processing Center, Redondo Beach, CA
Date: Mon, 7 Jun 93 18:07:21 GMT
Lines: 12

Hi,
        I'm attempting to locate an AVS module to perform Canny
edge-detection.  Unfortunately, no such module seems to exist either
with the basic distribution or at the IAC.  Does anybody know if there
is such a module in existance?

Any pointers/help would be greatly appreciated.

Thanks,

Brian



From ianh@resmel.bhp.com.au (Ian Hoyle)
Newsgroups: comp.graphics.avs
Subject: control widgets not visible
Date: 9 Jun 93 06:45:22 GMT
Organization: BHP Research - Melbourne Laboratories
Lines: 25
Message-ID: <ianh.739608322@monster>
NNTP-Posting-Host: monster.resmel.bhp.com.au

Help

Using the "Color Legend" module, only half of the control widgets for it
are visible. I only have those for tick marks, border color etc. Those
controlling legend position, size etc disappear out the bottom of the
pane holding the widgets. This is non-resizable and I can't get to them.

In fact the initial positioning of the control widgets is way too low
in the pane anyway. Is this a bug or can I control this somehow??

	ian

PS oh yes, running AVS 5.0 on an SGI Onyx/VTX (Irix 5.0)
-- 
   /\/\     :  Ian Hoyle,  Senior Research Scientist
  / / /\    :  Image Analysis Group
 / / /  \   :  BHP Research
/ / / /\ \  :  245 Wellington Rd, Mulgrave, 3170, AUSTRALIA
\ \/ / / /  :  Phone    +61-3-560-7066
 \  / / /   :  E-mail   ianh@resmel.bhp.com.au
  \/\/\/    :
	    :
"If the idea of a protocol behaving like a rabid, diseased sex-crased 
bunny rabbit appeals to you, AppleTalk is for you." 
      -- John Kennedy, comp.dcom.sys.cisco


From ianh@resmel.bhp.com.au (Ian Hoyle)
Newsgroups: comp.graphics.avs
Subject: Re: control widgets not visible
Date: 9 Jun 93 07:05:57 GMT
Organization: BHP Research - Melbourne Laboratories
Lines: 28
Message-ID: <ianh.739609557@monster>
References: <ianh.739608322@monster>
NNTP-Posting-Host: monster.resmel.bhp.com.au

ianh@resmel.bhp.com.au (Ian Hoyle) writes:

>Using the "Color Legend" module, only half of the control widgets for it
>are visible. I only have those for tick marks, border color etc. Those
>controlling legend position, size etc disappear out the bottom of the
>pane holding the widgets. This is non-resizable and I can't get to them.

Damn, it was so obvious.

Color legend had obviously changed between releases so my
old saved network was at fault. I killed the existing
color legend module then dragged in a new copy and all is well.

S'pose I should have RTFM (assuming this is noted somewhere :-)

	ian
-- 
   /\/\     :  Ian Hoyle,  Senior Research Scientist
  / / /\    :  Image Analysis Group
 / / /  \   :  BHP Research
/ / / /\ \  :  245 Wellington Rd, Mulgrave, 3170, AUSTRALIA
\ \/ / / /  :  Phone    +61-3-560-7066
 \  / / /   :  E-mail   ianh@resmel.bhp.com.au
  \/\/\/    :
	    :
"If the idea of a protocol behaving like a rabid, diseased sex-crased 
bunny rabbit appeals to you, AppleTalk is for you." 
      -- John Kennedy, comp.dcom.sys.cisco


From larryg@avs.com (Larry Gelberg)
Subject: Is anyone using AVS for acoustical modeling?
Message-ID: <1993Jun10.134742.1512@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
Date: Thu, 10 Jun 1993 13:47:42 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 13

We are interested in learning if any AVS users out there are doing
any work using AVS for visualizing concert-hall acoustics.  If you
or any of your colleagues are involved with this, please drop a line 
either to me or to Ian Reid (ianr@avs.com), our VP of Marketing.

Thanks in advance,
larryg

--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From cmft@doc.ic.ac.uk (C M F Tsang)
Newsgroups: comp.graphics.avs
Subject: Camera Model wanted
Date: 10 Jun 1993 15:59:23 +0100
Organization: Department of Computing, Imperial College, University of London, UK.
Lines: 11
Distribution: world
Message-ID: <1v7i8bINN4j7@oak29.doc.ic.ac.uk>
NNTP-Posting-Host: oak29.doc.ic.ac.uk
Keywords: Camera model


I urgently want a program which can generate a well focused image and the
defocused image (modelling of the point spread function) of a object.
It is desirable for the camera model with the raytracing algorithm included.

Thank for attention

Please reply to : k.chow@ee.ic.ac.uk

 



From rao@autsrv.iitsg.nrc.ca (Rao Palacharla,,)
Subject: Writing AVS modules to run on DECmpp 12000/ Maspar MP-2
Message-ID: <1993Jun10.150121.6179@nrcnet0.nrc.ca>
Originator: rao@alpha.ps.iit.nrc.ca
Sender: root@nrcnet0.nrc.ca (Operator)
Nntp-Posting-Host: alpha.ps.iit.nrc.ca
Organization: National Research Council, Ottawa, Canada
Date: Thu, 10 Jun 93 15:01:21 GMT
Lines: 16

We have a DECmpp 12000 (same as Maspar MP-2, with 1024 processors) 
with DEC workstation 5000/240 as the front end. We are running AVS 4.0 on 
the DEC workstation. I want to know if anyone has tried writing modules in 
AVS which will run on the Data Parallel Unit (DPU) in parallel. For example, 
to perform a Fourier Transform of a 512x512 image, it would be better to 
do the calculations in parallel on the DPU and then display the results 
thru AVS's image viewer.

I would appreciate to hear from anyone with experience on writing AVS modules
which will perform calculations on the Data Parallel Unit.

thank you
bye
rao

e-mail: rao@autsrv.iitsg.nrc.ca


From skr@rousku.kau.vtt.fi (Seppo Rantala)
Newsgroups: comp.graphics.avs
Subject: Hardware renderer with Sparcstation 10/GS
Date: 11 Jun 93 11:19:57
Organization: Technical Research Centre of Finland, Machine Automation
	Laboratory
Lines: 18
Distribution: comp
Message-ID: <SKR.93Jun11111957@rousku.kau.vtt.fi>
NNTP-Posting-Host: kau.vtt.fi


I have tried to install AVS 5.0(50.25 sparc) to our SPARCstation 10(41)/GS so
that we could use hardware renderer, but without any luck. I have contacted
support@avs.com, but they have no idea why it is not working. We have done
all tasks which are mentioned in the manual and even installed again both
AVS and OpenWindows.

So if you have managed to install AVS/harware renderer to sparc 10/GS, please
inform me how.

Many thanks in advance,

Seppo Rantala
--
 ____         ----------------------------------------------------------------
 \  /_______   Seppo Rantala     VTT, Machine Automation Lab
 \\//  | |     skr@kau.vtt.fi    PL 192, 33101 Tampere, FINLAND
  \/   | |     Tel: 931-163616   Fax: +358-31-163694


From hsu@crl.dec.com (William Hsu)
Subject: Re: Writing AVS modules to run on DECmpp 12000/ Maspar MP-2
Message-ID: <1993Jun11.174259.13739@crl.dec.com>
Sender: news@crl.dec.com (USENET News System)
Organization: DEC Cambridge Research Lab
References:  <1993Jun10.150121.6179@nrcnet0.nrc.ca>
Date: Fri, 11 Jun 1993 17:42:59 GMT
Lines: 35

In article <1993Jun10.150121.6179@nrcnet0.nrc.ca>, rao@autsrv.iitsg.nrc.ca (Rao Palacharla,,) writes:
> We have a DECmpp 12000 (same as Maspar MP-2, with 1024 processors) 
> with DEC workstation 5000/240 as the front end. We are running AVS 4.0 on 
> the DEC workstation. I want to know if anyone has tried writing modules in 
> AVS which will run on the Data Parallel Unit (DPU) in parallel. For example, 
...etc...
> bye
> rao
> 
> e-mail: rao@autsrv.iitsg.nrc.ca

Yeah, I have. I have a parallel volume renderer which runs on the
DECmpp 12000Sx Model 100 (MP-1, MP-2 for the purpose of this discussion
is essentially the same thing). It uses AVS as the interface to the
volume renderer. The module is a remote module so I can run it from
any workstation. I have a paper that describes the volume renderer,
that will appear in the Parallel Rendering Symposium, but it doesn't
discuss AVS in it.

I also believe the people from CSIRO have done a fair amount of
DECmpp/MP-n within AVS work. CSIRO is an Australian industrial
research consortium. I believe they've used it for terrain modeling.

The process is fairly easy. Just write the AVS wrapper as you
normally would, around your mpl (dpl) code. Compile and link
with both sets of libraries and you're set.

The only problem is if you want to share data within the DECmpp
between multiple modules, without shipping it back and forth between
the modules. I think CSIRO has something that's similar to the
shared memory business for the DECmpp, though.

Enjoy,
William
Standard Disclaimer...


From roy@ccsf.caltech.edu (Roy Williams)
Newsgroups: comp.graphics.avs
Subject: Large Images
Date: 11 Jun 1993 18:33:27 GMT
Organization: Caltech CCSF
Lines: 9
Distribution: world
Message-ID: <1vaj5nINNa6i@gap.caltech.edu>
Reply-To: roy@ccsf.caltech.edu
NNTP-Posting-Host: willow.ccsf.caltech.edu

We are thinking of buying a high-resolution color printer, so there is
a natural desire to be able to make high-resolution images from the
Geometry Renderer. The only way I know to do this is by putting it on the
screen first, then getting Postscript or AVS image format: the size of the
result is determined by the size of the screen image. Does anyone know
how to get a, say, 4000x4000 image from the geometry renderer?
Thanx
Roy



From mcheng@dunlop.cs.wisc.edu (Michael Cheng)
Subject: How to get AVS?
Message-ID: <1993Jun12.183930.13341@daffy.cs.wisc.edu>
Sender: news@daffy.cs.wisc.edu (The News)
Organization: University of Wisconsin, Madison -- Computer Sciences Dept.
Distribution: usa
Date: Sat, 12 Jun 1993 18:39:30 GMT
Lines: 12

Hi

Would someone tell me how to get a copy of AVS for the DECstations?
How much would it cost?

While we're at it, does any one have a bibliography of papers
on AVS, or any other related tools? I only have the one in the July 89 issue
of IEEE Computer Graphics and applications.

Thanks in advance for any help.

Mike


From avs@doppler.ncsc.org (AVS account)
Subject: AVS '93 Proceedings now available
Message-ID: <C8M6up.4ss@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Mon, 14 Jun 1993 13:59:13 GMT


We have copies of proceedings from the AVS '93 conference available for
$50.00 each. Shipping is free within the US; $10.00 for outside the US.
Please send check or money order payable to MCNC to:

MCNC
Corporate Communications
Attn.: Roger Lewis
PO Box 12889
Research Triangle Park, NC 27709-2889

Please include a mailing address with payment.
Thank you for your interest.
-- 
International AVS Center
North Carolina Supercomputing Center
avs@ncsc.org


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: AVS 93 Proceedings Available
Message-ID: <C8MGLI.K9M@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Mon, 14 Jun 1993 17:29:42 GMT

Hi Folks,

   -------------AVS '93 Proceedings Available----------

AVS '93 Conference Proceedings are now available from the
International AVS Center.  This 3 volume set includes over 100
papers covering AVS lectures, workshops, and tutorials.

To obtain all three volumes of the AVS '93 Proceedings, please
send US $50 (plus $10 if outside the US) with your request to:

        MCNC - Corporate Communications
        P.O. Box 12889
        3021 Cornwallis Road
        Research Triangle Park, NC 27709-2889
        USA

   ----------------------------------------------------

----------------------------------------------------------------
   Steve Thorpe, Application Visualization System Specialist
International AVS Center, North Carolina Supercomputing Center
PO Box 12889   3021 Cornwallis Rd, RTP, NC 27709   avs@ncsc.org
----------------------------------------------------------------


From larryg@avs.com (Larry Gelberg)
Subject: Re: Large Images
Message-ID: <1993Jun14.175122.7041@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
References: <1vaj5nINNa6i@gap.caltech.edu>
Date: Mon, 14 Jun 1993 17:51:22 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 30

roy@ccsf.caltech.edu (Roy Williams) writes:
: Does anyone know
: how to get a, say, 4000x4000 image from the geometry renderer?

Typically, if you use a computer's hardware renderer, you cannot 
get an image larger than the screen because the rendering is tied
to the system's framebuffer size.

In AVS, you can get bigger images by using the software renderer.
If you don't want to use the CLI, you can do all this interactively
in the Geometry Viewer's Camera menu: first set the software renderer
and then set the width and height.

Alternatively, You can use the CLI to set the renderer to the 
software renderer:
	geom_set_renderer "Software Renderer"

and then you can either (a) change the size of the of the camera:
	geom_resize_camera 4000 4000

or (b) you can create a new scene with a specified size:
	geom_create_scene 0 0 4000 4000

I hope this helps!
larryg
--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From news@berlioz.crs4.it (USENET News System)
Subject: Remote display of local network
Sender: news@crs4.it (USENET News System)
Message-ID: <1993Jun15.110058.73015@crs4.it>
Date: Tue, 15 Jun 1993 11:00:58 GMT
Organization: CRS4, Center for Adv.Studies, Research and Development in Sardinia
Lines: 32


In order to take some snapshots of my visualizations I'm trying to open 
a "display image" window on a remote machine that is holding a frame 
grabber board.

The networks needs to be run on an Indigo2 and only the display module
should run on a remote SGI host.

In principle also the display module could be run locally if I could
change the DISPLAY variable only for it.

I can open most other modules and run then on the remote machine with
the "Read remote module(s)" option but I can't manage to load any
display module on it.

Any help will be appreciated.

Thanks.

/lf

------------------------------------------------------------------------------
      _/_/_/_/ _/_/_/_/ _/_/_/_/ _/    _/  email  : Luigi.Filippini@crs4.it
     _/    _/ _/    _/ _/    _/ _/    _/   phone  : +39-70-2796-253 
    _/       _/    _/ _/       _/    _/             +39-70-2796-254
   _/       _/_/_/_/ _/_/_/_/ _/_/_/_/     fax    : +39-70-2796-283
  _/       _/   _/        _/       _/      snail  : Via Nazario Sauro, 10
 _/    _/ _/    _/ _/    _/       _/                09123 CAGLIARI
_/_/_/_/ _/    _/ _/_/_/_/       _/                 ITALY
------------------------------------------------------------------------------
     CENTER FOR ADVANCED STUDIES, RESEARCH AND DEVELOPMENT IN SARDINIA
------------------------------------------------------------------------------


From bolstad@blad.rtpnc.epa.gov (Mark Bolstad)
Subject: Re: Remote display of local network
Message-ID: <1993Jun15.120533.16489@rtpnc.epa.gov>
Sender: usenet@rtpnc.epa.gov
Nntp-Posting-Host: blad.rtpnc.epa.gov
Organization: United States Environmental Protection Agency
References:  <1993Jun15.110058.73015@crs4.it>
Date: Tue, 15 Jun 1993 12:05:33 GMT
Lines: 30

In article <1993Jun15.110058.73015@crs4.it>, news@berlioz.crs4.it (USENET News System) writes:

>In order to take some snapshots of my visualizations I'm trying to open 
>a "display image" window on a remote machine that is holding a frame 
>grabber board.
>
>The networks needs to be run on an Indigo2 and only the display module
>should run on a remote SGI host.
>
>In principle also the display module could be run locally if I could
>change the DISPLAY variable only for it.
>
>I can open most other modules and run then on the remote machine with
>the "Read remote module(s)" option but I can't manage to load any
>display module on it.
>
>Any help will be appreciated.
>
>Thanks.

The 5.0 version of the geometry viewer has a display field.  We have
used it to distributed the viewer between 2 SGIs, so this should
work for you.  As a side note AVS core dumped when I tried to
send the viewer to a SUN Sparc2 (I did set it for S/W renderer).

Any Comments?

Mark Bolstad
bolstad@vislab.rtpnc.epa.gov



From merlin@neuro.usc.edu (merlin)
Newsgroups: comp.graphics.avs
Subject: QUERY:  How much does AVS cost?
Date: 15 Jun 1993 06:02:20 -0700
Organization: University of Southern California, Los Angeles, CA
Lines: 24
Sender: merlin@neuro.usc.edu
Message-ID: <1vkh8s$ntm@neuro.usc.edu>
NNTP-Posting-Host: neuro.usc.edu

Could someone outline the rate structure for AVS?  I am presently
at an academic site -- however my background is in management of
large academic, commercial, and government computer centers.  I'd
like a general sense of what it costs to incorporate AVS (vs any
other packave -- i.e. khoros) into each of these environments?

Is AVS licensed on an institution wide basis?  Does the computing
performance of a processor modify the rate?  Does the license make
any distinction between single and multiple processor systems?
If site licenses are available does a site have to be physically
or merely administratively contiguous?

Thanks, AJ

------------------------------------------------------------------------------
Alexander-James Annala
Principal Investigator
Neuroscience Image Analysis Network
HEDCO Neuroscience Building, Fifth Floor
University of Southern California
University Park
Los Angeles, CA 90089-2520
------------------------------------------------------------------------------



From cstom@swan.pyr ( t simpson)
Subject: Re: Large Images
Message-ID: <1993Jun15.135509.25310@swan.pyr>
Organization: Swansea University College
References: <1vaj5nINNa6i@gap.caltech.edu>
Date: Tue, 15 Jun 1993 13:55:09 GMT
Lines: 58


      /*** SERIOUS HELP NEEDED ***/

Does anyone out there have any experience of programming
AVS modules that manipulate UCD data?  I have just about
given up on the module that I'm attempting to write as I
can't get the UCD structure routines to work. My code works
roughly thus :-

            UCD input transfered to
            my own datatype for manipulation

                        |
                        V

            One or two tranformations on the
            Data

                        |
                        V

            Data transfered back to UCD_structure
            for output.

The last stage where I convert the data back into a 
UCD structure is where it all falls apart, I can't get it
right.  The data is a very simple finite element mesh, 
triangular cells, no mid-edge-nodes, 1 scalar node-component
of data.  Is there a specific order in which you must call
the UCD routines, does anyone have any good tips or hints?
Of particular curiousity is that when the module is connected
to UCD_Write in a network, the first half cells and nodes are
numbered correctly, and the second half are all zero. (ie).

  # example wrong UCD file.
  6 2 1 0 0
  1  <x> <y> <z>
  2  <x> <y> <z>
  3  <x> <y> <z>
  0  <x> <y> <z>
  0  <x> <y> <z>
  0  <x> <y> <z>
  1 1 2 3
  0 0 0 0
  1 1
  streamfn, scalar
  1 0.0
  2 0.0
  3 0.0
  0 0.0
  0 0.0
  0 0.0

_PLEASE_, can someone help, it's driving me nuts.

   Tom Simpson : post-grad, UC Swansea CFD group.
        < T.Simpson@uk.ac.swansea >



From bolstad@blad.rtpnc.epa.gov (Mark Bolstad)
Subject: Coroutine Syncing
Message-ID: <1993Jun15.164207.19715@rtpnc.epa.gov>
Sender: usenet@rtpnc.epa.gov
Nntp-Posting-Host: blad.rtpnc.epa.gov
Organization: United States Environmental Protection Agency
Date: Tue, 15 Jun 1993 16:42:07 GMT
Lines: 13

I have a coroutine that is at the end of a network, i.e.
it has no outputs.  The input to the module is AVSImage
data from the geometry viewer.  I have set the
AVScorout_set_sync value to 1 (execute synchronously).

The problem is that when I run my network, the input
to my module sometimes skips a frame of data.

How do I get the network to behave synchronously?

Mark Bolstad
bolstad@vislab.rtpnc.epa.gov



From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: Re: Coroutine Syncing
Message-ID: <C8oHDK.ny@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Tue, 15 Jun 1993 19:41:44 GMT

Mark recently posted:

>I have a coroutine that is at the end of a network, i.e.
>it has no outputs.  The input to the module is AVSImage
>data from the geometry viewer.  I have set the
>AVScorout_set_sync value to 1 (execute synchronously).
>
>The problem is that when I run my network, the input
>to my module sometimes skips a frame of data.
>
>How do I get the network to behave synchronously?
>
>Mark Bolstad
>bolstad@vislab.rtpnc.epa.gov

Maybe this December posting from Wes Bethel will help solve the problem?
Good luck with it!

-Steve

From: wes@ux6.lbl.gov (Wes Bethel)
Newsgroups: comp.graphics.avs
Subject: Re: Help with Synchronization
Organization: Lawrence Berkeley Laboratory

In article <1992Dec20.054705.8951@ccnews.nctu.edu.tw+ blshyu@nctu.edu.tw () writes:
+...
+                                 boolean
+                                    |
+			 ----------------------
+			|                      |
+			|                animate_file
+	                |                      |
+			|                 read_field
+			|                      |
+	           animate_file           iso_surface
+			|                      |
+			|               render_geometry
+			|                      |
+			 ---------    --------- 
+			          |  |
+                               write_image
+
+  The "write_image" module needs two inputs from the two branches of
+the network indevudually.  Since the right branche of the network
+needs more time to complete the calculation than the left branch, I
+need a method to do the synchronization of the two branch or I can have
+many images saved with different names but with the same content.

(sorry for including so much of the original article - it was unavoidable)

Back when I was in school, the EE's would use a T-flop (trigger
flip-flop) to solve this problem. At the IAC is a module called
"shutter" which provides the functionality that you need. 

It takes as input an image, and when a "true" signal is applied to
it's "fire" parameter, it sends the input image to its output port.

So, your modified network would look like:

+                                 boolean
+                                    |
+			 ----------------------
+			|           |          |
+			|           |    animate_file
+	                |           |          |
+			|           |     read_field
+			|           |          |
+	           animate_file     |     iso_surface
+			|           |          |
+			|           |   render_geometry
+			|           |          |
+			 ---------  | --------- 
+			          | ||
+			          | shutter
+			          |  |
+                               write_image
+

Happy holidays!

wes


----------------------------------------------------------------
PS  Would you be interested in sharing your AVS work with the
scientific visualization community worldwide via a short article 
and / or slides in an upcoming issue of AVS Network News?  This 
is the IAC's quarterly magazine made up of user contributed 
articles.   Thanks for considering this!
----------------------------------------------------------------
   Steve Thorpe, Application Visualization System Specialist
International AVS Center, North Carolina Supercomputing Center
PO Box 12889   3021 Cornwallis Rd, RTP, NC 27709   avs@ncsc.org
----------------------------------------------------------------



From goodrich@lynx.noname (Mike Goodrich)
Newsgroups: comp.graphics.avs
Subject: Re: Coroutine Syncing
Date: 15 Jun 1993 19:39:07 GMT
Organization: Sun Microsystems, Inc.
Lines: 22
Distribution: world
Message-ID: <1vl8grINN50c@rave.larc.nasa.gov>
References: <1993Jun15.164207.19715@rtpnc.epa.gov>
Reply-To: goodrich@lynx.noname
NNTP-Posting-Host: lynx.larc.nasa.gov

In article 19715@rtpnc.epa.gov, bolstad@blad.rtpnc.epa.gov (Mark Bolstad) writes:
>I have a coroutine that is at the end of a network, i.e.
>it has no outputs.  The input to the module is AVSImage
>data from the geometry viewer.  I have set the
>AVScorout_set_sync value to 1 (execute synchronously).
>
>The problem is that when I run my network, the input
>to my module sometimes skips a frame of data.
>
>How do I get the network to behave synchronously?
>
>Mark Bolstad
>bolstad@vislab.rtpnc.epa.gov
>


I'm kinda curious. If its at the end of a network, why is it
a coroutine? I f you look in the Developers Guide, it shows you
how to synchronize the way you want to, I beleive...

-Mike



From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: Re: How to AVS MRI??
Message-ID: <C8oJuv.4Hx@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Tue, 15 Jun 1993 20:35:18 GMT

Jim Boyle recently wrote:

> I'd like to try to get an MRI data set I have into AVS.  Anyone know
> how?? The format is each slice is 256x256 pixels, each of unsigned int
> variety.  Slices are 3mm and inplane resolution is .78125 mm.  Memory
> layout is 
> 	[column,row,slice]
> 
> 	[0,0,0] [1,0,0] .....[255,0,0] [0,1,0] .....[255,255,0] [0,0,1]...
> 
> or as memory addresses increase we first increment column, then row, then
> slice.  Any suggestions on how to get AVS to read it??

One option would be to use the example code that comes with the
AVS distribution in /usr/avs/examples/read_image.c.  This could easily
be modified to be a read_MRI module, which would read in the data
format you describe above.  This would have the added benefit of helping
you to better understand what's going on "behind the scenes" with the
data structures in AVS.

There are a bunch of data input modules on the International AVS Center's 
public domain anonymous ftp site, and no doubt at least a few of these
could be easily modified to read your dataset (if there isn't on there
already).  These and others can be found in 
		avs.ncsc.org:avs_modules/data_input

In particular, the read_rectiliner module might be just the ticket...


Name        : read_rectilinearVersion      : 1.000     Mod Number : 1296
Author      : Wes Bethel, Lawrence Berkeley Laboratory
Submitted   : 08/04/92        Last Updated : 08/04/92  Language   : C
Ported to   : Convex IBM HP DEC Sun
Description : This module reads data in ascii format into an "rectilinear
              field" data structure. The disk file is structured to
              contain information about the field at the beginning of the
              file, followed by the data.

Name        : read_uniform    Version      : 2.000     Mod Number : 1075
Author      : Wes Bethel, Lawrence Berkeley Laboratory
Submitted   : 10/10/91        Last Updated : 08/30/92  Language   : C
Ported to   : Convex HP Sun
Description : This module reads data in ascii format into a "uniform
              field" data structure. The disk file is structured to
              contain information about the field at the beginning of the
              file, followed by the data.


Good luck with it!

-Steve

PS  Would you be interested in sharing your AVS work with the
scientific visualization community worldwide via a short article 
and / or slides in an upcoming issue of AVS Network News?  This 
is the IAC's quarterly magazine made up of user contributed 
articles.   Thanks for considering this!
----------------------------------------------------------------
   Steve Thorpe, Application Visualization System Specialist
International AVS Center, North Carolina Supercomputing Center
PO Box 12889   3021 Cornwallis Rd, RTP, NC 27709   avs@ncsc.org
----------------------------------------------------------------

  


From bolstad@blad.rtpnc.epa.gov (Mark Bolstad)
Subject: Re: Coroutine Syncing
Message-ID: <1993Jun16.120915.578@rtpnc.epa.gov>
Sender: usenet@rtpnc.epa.gov
Nntp-Posting-Host: blad.rtpnc.epa.gov
Organization: United States Environmental Protection Agency
References:  <C8oHDK.ny@doppler.ncsc.org>
Date: Wed, 16 Jun 1993 12:09:15 GMT
Lines: 12

Thanks for the responses.

The reason I am using a coroutine is that I am processing X events
and the only way I saw to handle them was to use a coroutine with
AVScorout_X_wait.

I will give shutter a try.

Thanks,

Mark Bolstad



From ajp39368@uxa.cso.uiuc.edu (James_Bond)
Subject: NEEDED: Targa 24-bit picture file decoder
Date: Thu, 17 Jun 1993 01:06:50 GMT
Summary: needed: Targa 24-bit picture file decoder
Message-ID: <C8qr3E.FC5@news.cso.uiuc.edu>
Sender: usenet@news.cso.uiuc.edu (Net Noise owner)
Keywords: needed, targa, 24-bit, decoder, picture file, source
Organization: University of Illinois at Urbana
Lines: 14

I posted a while ago regarding this need of mine and i received
some good info about it and also some source code regarding
the 24-bit TARGA file decoder...  
but, by accident, i have LOST all that information, so i am going
to try this again and not lose it again...
so, if anyone has any information as to the 24-bit TARGA picture
file format decoder, please let me know, and if you have
any source for it, please contact me...
thank you...

--AJ>
ajp39368@uxa.cso.uiuc.edu
An ideal wife is the woman who has an ideal husband.



From sauer@iitcsun3.med.miami.edu.iitc (Frank Sauer)
Subject: HELP NEEDED: sliced 3D MRI data
Message-ID: <1993Jun15.174814.10622@newssun.med.miami.edu>
Sender: news@newssun.med.miami.edu
Nntp-Posting-Host: iitcsun3.med.miami.edu
Reply-To: sauer@iitcsun3.med.miami.edu.iitc
Organization: Sun Microsystems, Inc.
References: <1993Jun15.164207.19715@rtpnc.epa.gov>
Date: Tue, 15 Jun 1993 17:48:14 GMT
Lines: 23

Hi folks,

I am a new user of AVS and I like to know
how to read data of the following type into
AVS to produce a 3D model:

My data is an MRI of a human brain, and consists
of 90 files of raw binary data, each file being a
256x256 12-bit image.

I have a feeling someone must have done this before,
so before I start re-inventing the wheel...

Besides this, are there any ftp-sites containing user
contributed AVS stuff? Is there a FAQ?

Any help greatly appreciated,

Frank Sauer
sauer@iitcsun3.med.miami.edu





From mohan@tulip (Mohan Pakkurti)
Date: 20 Jun 93 06:56:37 EST
Reply-To: mohan@tulip.es.utoledo.edu
Nntp-Posting-Host: tulip.es.utoledo.edu
X-Newsreader: TIN [version 1.1 PL9]
Lines: 13

Hello,

I am interested to know if anyone here is using AVS for visualizing the
behavior of semiconductor devices and electronics circuits. 

Thanks for the info.
:) mohan


===============================================================================
+ Mohanakrishna Pakkurti            + mohan@jupiter.cse.utoledo.edu           +
+ HOME: 2239 University Hills Blvd #204, Toledo OH 43606. Phone:(419)536-9073 +
===============================================================================


From kj@cs.brown.edu (Kopal Jha)
Subject: 3d particle distribution
Message-ID: <1993Jun21.163027.8101@cs.brown.edu>
Summary: Need help to make 3D scatter plot
Keywords: scatter dots, particles, 3D
Sender: news@cs.brown.edu
Organization: Brown University Department of Computer Science
Date: Mon, 21 Jun 1993 16:30:27 GMT
Lines: 28

Hi,

I need help making what is basically a 3D scatter plot.  I have data written
as 

x y z val
. . . .
. . . .
. . . .
.
.

where the variable val has a value at x,y,z within a 3D volume.  I want 
to plot a point at each x,y,z location, and give the point a colour that
depends on the value of val.

I'm not that adept at AVS, so I am having a lot of trouble doing this though
it seems to me that it should be a trivial exercise.  

Complicating factors (?) are that 
a)  I wrote the numbers out in integer*2 to save space
b)  I used fortran unformatted for the same reason

Please help!  Thanks very much,

Kopal Jha




From madhatta@plume.mit.edu (Tom Yates)
Newsgroups: comp.graphics.avs
Subject: tracer background setting
Date: 21 Jun 1993 18:38:11 GMT
Organization: CMPO at MIT
Lines: 23
Message-ID: <204v6j$nie@senator-bedfellow.MIT.EDU>
NNTP-Posting-Host: plume.mit.edu

i am using the tracer module in AVS 4 to render temperature data from an
ocean convection model.  the idea is to present our results in a similar
form to another group who have simulted the process by floating a slab of
blue ice on a rotating water bath.

i have arrived at the conclusion that some form of volume rendering is
necessary and tracer seems to be the module for the job, which gives
bearable results.  the problem is, the background is black (rays which
intercept nothing of significant opacity and disappear off to infinity) and
in the photographs, the background is white.  i would very much like some
way of persuading tracer to "do my volume rendering against a white
background".  short of actually defining an opaque white wall at the back
of my data set, is there a good way of doing this analogous to the
background colour settings i can get with "render geometry"?

can i do better with a different rendering module?

sorry if this is a trivial question, our AVS documentation is not very
good.

tom
-- 
>From the machine of Tom the Mad - madhatta@teaparty.mit.edu - DoD#0135 - KoHO


From larryg@avs.com (Larry Gelberg)
Subject: Re: HELP NEEDED: sliced 3D MRI data
Message-ID: <1993Jun22.124048.1482@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
References: <1993Jun15.174814.10622@newssun.med.miami.edu>
Date: Tue, 22 Jun 1993 12:40:48 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 28

sauer@iitcsun3.med.miami.edu.iitc (Frank Sauer) writes:
: My data is an MRI of a human brain, and consists
: of 90 files of raw binary data, each file being a
: 256x256 12-bit image.

The module read_scans.c in /usr/avs/examples shows how to do this
almost exactly.  I think you may need to fiddle with some of the 
file parameters (like skip factor, etc.) Is the data 12-bit data 
stored as 16-bit shorts or is it 12 bits (3 bytes) followed directly
by the next 12 bits?  If the latter, you may need to read it in 
3 bytes at a time and collect them in either a short or long integer.

: Besides this, are there any ftp-sites containing user
: contributed AVS stuff? Is there a FAQ?

The International AVS Center at the North Carolina Supercomputing 
Center maintains the anonymous ftp site of AVS modules.  They can 
be reached via email at avs@ncsc.org or via ftp at avs.ncsc.org
(128.109.178.23)

Good luck!
larryg

--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From larryg@avs.com (Larry Gelberg)
Subject: Re: 3d particle distribution
Message-ID: <1993Jun22.125152.2111@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
References: <1993Jun21.163027.8101@cs.brown.edu>
Date: Tue, 22 Jun 1993 12:51:52 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 42

kj@cs.brown.edu (Kopal Jha) writes:
: I need help making what is basically a 3D scatter plot.  
: I have data written as 
: 
: x y z val
: . . . .
: 
: where the variable val has a value at x,y,z within a 3D volume.  I want 
: to plot a point at each x,y,z location, and give the point a colour that
: depends on the value of val.
: 
: Complicating factors (?) are that 
: a)  I wrote the numbers out in integer*2 to save space
: b)  I used fortran unformatted for the same reason
: 

The easiest way to visualize this data is to use the SCATTER DOTS module
as it is expecting data in almost exactly that format.  From the man page:

The scatter dots module generates spheres of various radii at the
coordinate locations in a specified field. For a scalar field, each
sphere's radius is proportional to the scalar value, and the sphere is
always colored white.  If the field is a 4-vector float (such as that
produced by the bubbleviz module), only the first element of the
vector determines the sphere's radius. The other three elements are
interpreted as red-green-blue color values (normalized to the range 0..1).

You probably want to read in your data as a "field 1D short 3-space
irregular" and then use the "field to float" module prior to sending it
to scatter dots.

Also check out scatter to ucd to turn your scattered data into tetrahedra
which you can compute isosurfaces, slices, etc. against.

Good luck!
larryg

--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From larryg@avs.com (Larry Gelberg)
Subject: Re: tracer background setting
Message-ID: <1993Jun22.130303.2328@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
References: <204v6j$nie@senator-bedfellow.MIT.EDU>
Date: Tue, 22 Jun 1993 13:03:03 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 127

Here is a network which demonstrates a hack to accomplish the task
of putting a colored background behind your tracer image.  The hack
is based on the fact that tracer also outputs data in the opacity
channel so the image it produces can be composited against a 
background image.

Let me know if this works for you:

--------------------- CUT HERE --------------------------------------
#!/usr/bin/avs -network
version 5.0 (50.33 gs)
module "read field.user.0" -xy 398,82 -ex $Path/avs_library/mongo
module "generate colormap.user.1" -xy 68,82
module tracer.user.2 -xy 238,172 -ex $Path/avs_library/tracer
module "euler transformation.user.3" -xy 228,82 -ex $Path/avs_library/mongo
module background.user.4 -xy 118,232 -ex $Path/avs_library/mongo
module composite.user.5 -xy 238,282 -ex $Path/avs_library/mongo
module "image viewer.user.6" -xy 238,352
port_connect "read field.user.0":0 tracer.user.2:0
port_connect "generate colormap.user.1":0 tracer.user.2:2
port_connect tracer.user.2:0 composite.user.5:0
port_connect tracer.user.2:0 background.user.4:0
port_connect "euler transformation.user.3":0 tracer.user.2:1
port_connect background.user.4:0 composite.user.5:1
port_connect composite.user.5:0 "image viewer.user.6":0
parm_set "read field.user.0":"Read Field Browser" $Path/data/field/hydrogen.fld
parm_set tracer.user.2:Width 128
parm_set tracer.user.2:Height 128
parm_set "euler transformation.user.3":phi 40.09144974
parm_set background.user.4:"Lower Left Sat" 0
parm_set background.user.4:"Lower Left Value" 1
parm_set background.user.4:"Lower Right Sat" 0
parm_set background.user.4:"Lower Right Value" 1
mod_image "image viewer.user.6"
image_set_scene -scene "image viewer.user.6"
image_set_view_size 739 121 492 492
image_set_view_transformation -tx 0.000000 -ty 0.000000 -sx 1.000000 -sy 1.000000
image_set_color 0.000000 0.000000 0.000000
image_create_image -image "composite.user.5"
image_set_image_transformation -image "composite.user.5" -tx 0.000000 -ty 0.000000 -sx 2.234375 -sy 2.234375
image_set_visibility -image "composite.user.5" 1
}
shell "ui" shell
 panel Application -w app_panel -p ui -xy 0,0 -wh 260,1024
  panel "Top Level Stack" -w master_stack -p Application -xy 2,100 -wh 256,754
   panel "read field.user.0" -w panel -p "Top Level Stack" -xy 0,152 -wh 256,324
    manipulator "read field.user.0:Read Field Browser" -w browser -p "read field.user.0" \
        -xy 10,10 -wh 236,194
    manipulator "read field.user.0:Data Conversion" -w radio_buttons -p "read field.user.0" \
        -xy 10,204 -wh 118,44
    manipulator "read field.user.0:Read Status" -w textblock -p "read field.user.0" \
        -xy 10,248 -wh 236,66
   panel "generate colormap.user.1" -w panel -p "Top Level Stack" \
       -xy 0,86 -wh 256,566
    manipulator "generate colormap.user.1:colormap" -w color_editor -p "generate colormap.user.1" \
        -xy 10,10 -wh 236,416
    manipulator "generate colormap.user.1:lo value" -w dial -p "generate colormap.user.1" \
        -xy 10,426 -wh 90,130
    manipulator "generate colormap.user.1:hi value" -w dial -p "generate colormap.user.1" \
        -xy 100,426 -wh 90,130
   panel tracer.user.2 -w panel -p "Top Level Stack" -xy 0,152 -wh 256,326
    manipulator "tracer.user.2:alpha scale" -w dial -p tracer.user.2 \
        -xy 10,10 -wh 90,130
    manipulator tracer.user.2:perspective -w dial -p tracer.user.2 \
        -xy 100,10 -wh 90,130
    manipulator tracer.user.2:Width -w typein_integer -p tracer.user.2 \
        -xy 10,140 -wh 118,22
    manipulator tracer.user.2:Height -w typein_integer -p tracer.user.2 \
        -xy 10,162 -wh 118,22
    manipulator tracer.user.2:interpolate -w toggle -p tracer.user.2 \
        -xy 10,184 -wh 118,22
    manipulator "tracer.user.2:Performance Stats" -w textblock -p tracer.user.2 \
        -xy 10,206 -wh 236,110
   panel "euler transformation.user.3" -w panel -p "Top Level Stack" \
       -xy 0,152 -wh 200,280
    manipulator "euler transformation.user.3:theta" -w dial -p "euler transformation.user.3" \
        -xy 10,10 -wh 90,130
    manipulator "euler transformation.user.3:phi" -w dial -p "euler transformation.user.3" \
        -xy 100,10 -wh 90,130
    manipulator "euler transformation.user.3:rho" -w dial -p "euler transformation.user.3" \
        -xy 10,140 -wh 90,130
    manipulator "euler transformation.user.3:scale" -w dial -p "euler transformation.user.3" \
        -xy 100,140 -wh 90,130
   panel background.user.4 -w panel -p "Top Level Stack" -xy 0,152 -wh 256,602
    manipulator "background.user.4:Upper Left Hue" -w dial -p background.user.4 \
        -xy 10,10 -wh 90,130
    manipulator "background.user.4:Upper Right Hue" -w dial -p background.user.4 \
        -xy 100,10 -wh 90,130
    manipulator "background.user.4:Lower Left Hue" -w dial -p background.user.4 \
        -xy 10,140 -wh 90,130
    manipulator "background.user.4:Lower Right Hue" -w dial -p background.user.4 \
        -xy 100,140 -wh 90,130
    manipulator "background.user.4:Upper Left Sat" -w slider -p background.user.4 \
        -xy 10,270 -wh 236,32
    manipulator "background.user.4:Upper Left Value" -w slider -p background.user.4 \
        -xy 10,302 -wh 236,32
    manipulator "background.user.4:Upper Right Sat" -w slider -p background.user.4 \
        -xy 10,334 -wh 236,32
    manipulator "background.user.4:Upper Right Value" -w slider -p background.user.4 \
        -xy 10,366 -wh 236,32
    manipulator "background.user.4:Lower Left Sat" -w slider -p background.user.4 \
        -xy 10,398 -wh 236,32
    manipulator "background.user.4:Lower Left Value" -w slider -p background.user.4 \
        -xy 10,430 -wh 236,32
    manipulator "background.user.4:Lower Right Sat" -w slider -p background.user.4 \
        -xy 10,462 -wh 236,32
    manipulator "background.user.4:Lower Right Value" -w slider -p background.user.4 \
        -xy 10,494 -wh 236,32
    manipulator "background.user.4:X Resolution" -w typein_integer -p background.user.4 -hide \
        -xy 10,526 -wh 236,22
    manipulator "background.user.4:Y Resolution" -w typein_integer -p background.user.4 -hide \
        -xy 10,548 -wh 236,22
    manipulator background.user.4:Dither -w toggle -p background.user.4 \
        -xy 10,570 -wh 118,22
 panel "image viewer.user.6!image" -w container -p ui -xy 719,141 -wh 532,564\
   -P zoom_coords string "0 0 0 0 0 <$NULL> 0 0 0 0"
manipulator "image viewer.user.6":"Update Always" -w none
manipulator "image viewer.user.6":"Update Image" -w none
# End of file
--------------------- CUT HERE --------------------------------------


--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From siegel@mcclb0.med.nyu.edu
Newsgroups: comp.graphics.avs
Subject: HELP! AVS .geom to DXF, IGES, VDAFS, Arc+..... or Arc+ formats
Message-ID: <22JUN93.18514134@mcclb0.med.nyu.edu>
Date: 22 Jun 93 23:51:41 GMT
Sender: notes@cmcl2.nyu.edu (Notes Person)
Organization: NYU Medical Center, New York, 10016
Lines: 21
Nntp-Posting-Host: mcclb0.med.nyu.edu

I am in need of a module (or the file description) for ....to convert my AVS .geom
output into one of the following 3D model formats: DXF, IGES VDAFS or Arc+.
I have generated both a surface (vie.a isosurface) and a decent wireframe from
dt..my data, and would like to export it.

Also, I noc.ticed a module cao.lled write_rectilinear, whcih ....ich outputs what wul..ould
probably be a ber...very useful ascii file s.describing a surface; the problem is
that its input port is looking for "field rectii.linear float" whilb..le my output
is geom....a "geometry"; I.and I can't seem to find a module to convert a geometry
back to a field rectilinear float.

It seem .s to me that I h.can't be the first preson......er..person wich..shing to export my .geom
files; thse..ese modules muys..st exist somewhere..; but can't find them at AVS .
Interan..nta..ational FTP site.

Tah..hanks for any help.

Chris Siegel
NYU School of Medicine
email (PLEASE!!): chris@mchip01.med.nyu.edu
or: *.(2120.)263-5744


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: For AVS/MRI users
Message-ID: <C91KC8.1p6@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Tue, 22 Jun 1993 21:14:31 GMT

Hi Folks,

Just a quick posting on behalf of Doug Smith, please respond
directly to him at dsmith@uoft02.utoledo.edu.

Thanks!

Dear AVS/MRI users:

	I am very interested in finding out who out there is working in MRI
using AVS, what level of software and GUI exists, and what needs there are
for further development.  I am particularly interested in finding out your
opinions on 3D rendering of MRI data.

	I am also looking for developers who might have an interest in 
technology transfer of their work which might lead to commercialization.

Douglas A. Smith, Ph.D.
President and CEO
DASGroup

voice:	419-537-2116 or
	419-472-9160
fax:	419-537-4033
email:	dsmith@uoft02.utoledo.edu

mailing address:  3807 Elmhurst Road
		  Toledo, OH  43613


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: RE: HELP! AVS .geom to DXF, IGES, VDAFS, Arc+..... or Arc+ formats
Message-ID: <C92vvt.9FA@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Wed, 23 Jun 1993 14:21:28 GMT

Chris recently posted:

>I am in need of a module (or the file description) for ....to convert my AVS .geom
>output into one of the following 3D model formats: DXF, IGES VDAFS or Arc+.
>I have generated both a surface (vie.a isosurface) and a decent wireframe from
>dt..my data, and would like to export it.
>
>Also, I noc.ticed a module cao.lled write_rectilinear, whcih ....ich outputs what wul..ould
>probably be a ber...very useful ascii file s.describing a surface; the problem is
>that its input port is looking for "field rectii.linear float" whilb..le my output
>is geom....a "geometry"; I.and I can't seem to find a module to convert a geometry
>back to a field rectilinear float.
>
>It seem .s to me that I h.can't be the first preson......er..person wich..shing to export my .geom
>files; thse..ese modules muys..st exist somewhere..; but can't find them at AVS .
>Interan..nta..ational FTP site.
>
>Tah..hanks for any help.
>
>Chris Siegel
>NYU School of Medicine
>email (PLEASE!!): chris@mchip01.med.nyu.edu
>or: *.(2120.)263-5744

Perhaps the following module could be adapted to solve this problem.
If so, please resubmit it to the IAC!  (FYI, this can be found
at avs.ncsc.org:avs_modules/data_output/print_verts

Thanks,

-Steve

Module Name : print verts

Input	    : any geometry

Output	    : none

Parameters  : none

This module inputs any geometry, and outputs to the avs console window
all of the vertices in the geometry objects present in the geometry
edit list.

This module initially can serve as a debugging tool for modules that
output geometries.  It also serves as an example module on how to 
process geometries on a module's input port.  It most useful purpose
however, is probably to extract quantitative information regarding
surfaces, from visualization displays.


PS  Would you be interested in sharing your AVS work with the
scientific visualization community worldwide via a short article 
and / or slides in an upcoming issue of AVS Network News?  This 
is the IAC's quarterly magazine made up of user contributed 
articles.   Thanks for considering this!
----------------------------------------------------------------
   Steve Thorpe, Application Visualization System Specialist
International AVS Center, North Carolina Supercomputing Center
PO Box 12889   3021 Cornwallis Rd, RTP, NC 27709   avs@ncsc.org
----------------------------------------------------------------


From wild@embl-heidelberg.de
Newsgroups: comp.graphics.avs
Subject: string on output port - Fortran example needed.
Message-ID: <1993Jun23.171919.95914@embl-heidelberg.de>
Date: 23 Jun 93 17:19:19 +0100
Organization: EMBL, European Molecular Biology Laboratory
Lines: 12

Hello,

I am writing a Fortran module which has a string as an output port.

The string variable is passed as CHARACTER*(*) to the computation routine,
but I am not clear as to how one actually copies the string to the output
port. In the C examples I have found, the string is passed as 'char **'.
What is the equivalent Fortran trick?

Many thanks,

David Wild


From jlee@cs.ulowell.edu (John Peter Lee)
Subject: CFP II Workshop on Database Issues for Visualization
Message-ID: <C93DwF.5yy@ulowell.ulowell.edu>
Sender: usenet@ulowell.ulowell.edu (News manager - ulowell)
Organization: UMass-Lowell Computer Science
Date: Wed, 23 Jun 1993 20:50:39 GMT
Lines: 253


Hi:

This is a second Call, as some changes were made to the paper submission
and registration sections.

Turnkey Visualization Systems such as AVS, Khoros, and Explorer are
putting the power of Visualization onto many desktops, and are converting
many researchers into advanced data visualizers.

The issues of dealing with data models, their access, and querying services
for data visualization are of great importance to the research community.
We would like to solicit position papers and statements from qualified
individuals using or developing applications on these turnkey visualization
systems, who are involved with the data management aspects. They do not
have to be working with traditional DBMSs (Postgres, ObjectStore, etc.).

I have noticed on these systems that many of the modules are related
to reading, writing, and transforming among various data formats. Perhaps
needs would be better served by the incorporation of a DBMS !

We'd like to hear from you. Deadline is July 15th, 1993.

thanx,

JP Lee and Georges Grinstein
UMass Lowell 
Institute for Visualization and Perception Research





	************** 	Call for Participation	************** 



Title:	"Workshop on Database Issues for Data Visualization"


Place:	Visualization '93
	Red Lion Inn
	San Jose, CA

Date:	Tuesday, October 26, 1993
	8:30 am - 5:00 pm


A full-day workshop on the application of database technology to data 
visualization will be held during the 1993 IEEE Visualization Conference.
The purpose of the workshop will be to provide a forum for the presentation,
discussion, and evaluation of designs that apply database technology to the
data visualization problem. It seeks to bring together a diverse group of
researchers in the database and visualization fields, along with application
developers towards creating more supportive tools to enable the exploration,
analysis, and visualization of complex data. It is imperative that such 
cross-fertilization occur for visualization tools to better fit the task they
are designed for.



IEEE VISUALIZATION'93 CONFERENCE

Scientific visualization is an important research frontier shared by a
variety of computational science and engineering fields. Visualization work
is both interdisciplinary and a field in its own right.  This conference
focuses on interdisciplinary methods and supports collaboration among the
developers and users of visualization methods across all of science,
engineering, and commerce.

The conference is sponsored by the IEEE Computer Society Technical Committee
on Computer Graphics, in cooperation with ACM SIGGRAPH


WORKSHOP DESCRIPTION


Data Visualization deals with the effective protrayal of data with a goal
towards insight about the data. Typically, the data is of high volume, 
multidimensional in nature, and does not lend itself to easy display. The
data is also often spatial and temporal in nature. 

Data visualization software systems are very popular with end-user domain
scientists who require visual tools to explore and analyze their data.
Whereas the output issues for visualization production, i.e., image synthesis,
have received much attention, input and data management issues have not.
The visualization community has not embraced database technology to handle
data access for reasons of performance and inadequacies of the available
data models (such as relational). This workshop hopes to address these issues,
and make a case for database technology integration into visualization.

Database issues for data visualization have become increasingly important as
ever-larger data set sizes and dimensionality create problems that present-day 
systems cannot handle. Collaborative computing requires multiple users
accessing multiple, heterogeneous databases. To explore relationships between
many data variables, researchers must be able to query data in meaningful
ways. Currently, visualization system users are only given minimal querying
support, limited in most cases to interaction with output data represent-
ations, or single-valued data probes. This workshop intends to deal with
issues specifically related to the integration of database management sys-
tems with data visualization in hopes of determining methods for supplying
end users with the data interaction support they require, and overcoming
performance impediments related to the integration.

Participation is limited to 20 people.

Two previous Workshops (Siggraph '90 and Visualization '91) have dealt with
areas such as high-level requirements for data structures and access soft-
ware, and data visualization environments. This workshop intends to deal with
issues specifically related to the integration of database management systems
with data visualization. This could also lead to low-level implementation
issues as well. Researchers who have working applications are encouraged to
present their work, for critique and evaluation.


Position Statements and papers are welcome on the following issues as they 
relate to data visualization - we would like to keep discussions focused on 
the end result, which is improving visualization with database management
systems:


	- Requirements Visualization places on Databases
	- Data Models and Access Structures
	- Modeling the User - Tasks, Processes, Support Issues
	- Advanced User Interfaces for databases
	- Visual Languages for Visualization
	- System Integration Issues
	- Querying for Visualization 
	- Distributed, Heterogeneous Data Set Issues - Data Sharing

	

WORKSHOP FORMAT

The Workshop format allows for the formal presentation of the papers in the
morning session (accepted papers included in preprints). These presentations
are designed to serve as catalysts for discussion. In addition, there will
be a few invited papers and several afternoon workshop discussion sessions
with designated discussion topics. Participants will have an opportunity to
edit and resubmit their papers for possible post-workshop publication.


PAPER SUBMISSIONS (Due 15 July 1993)

Papers are solicited that present research results in the related areas of
database technology as applied to visualization. Position papers will also
be accepted.  Papers should be limited to 5,000 words and may be accompanied
by NTSC video. These should describe some original research on the
particular subject, and how it fits in with the overall theme of the
workshop. Proper references should be cited.

Position Statements (1-2 pages) are also accepted.
Position Papers might preceed a longer paper to be given at the Workshop.


Submissions may be post-mailed or e-mailed to:

J.P. Lee
Institute for Visualization and Perception Research
University of Massachusetts at Lowell
Computer Science Department
Lowell, MA 01854

jlee@cs.ulowell.edu

Hard Copy: 2 copies of original, preferrably Formatted, double spaced
Electronic Copy: Ascii, TeX, or Postscript: preferrably formatted

email or call with specific requests:

(617) 271-7417 days
(508) 433-8848 nights


ACCOMMODATION

Location: The Workshop will be held at the Red Lion Hotel, the Visualization'93
Conference Hotel.

*** VERY IMPORTANT NOTE ***

There is a Park and Recreation conference that is using most of the hotel rooms
the week before our workshop, and has 417 (of the 500) rooms booked on Friday
and 345 booked on Saturday.  It is the largest conference that San Jose has
ever had!  Hotel space will get very tight so please reserve rooms early and
guarantee these for late arrival (if arriving after after 6:00 pm) with a
credit card.  Rooms are being held until October 3, 1993. When making
reservations please mention IEEE.

Accommodation inquiries:
      Red Lion Hotel - Reservations
      2050 Gateway Place
      San Jose CA 95110
      +1-408-453-4000


MEALS AND REFRESHMENTS

Lunch as well as coffee and tea breaks will be provided.
 

REGISTRATION FEE

Registration forms will be sent to the accepted participants by the
Visualization '93 Registration co-chairs.

There is a single registration fee of US $100 which covers the technical
session, the Workshop preprints, lunch, and morning and afternoon coffee.

Payment: the registration fee can be paid in any convertible currency by either
check or credit card (American Express, Mastercard or Visa).  


WORKSHOP ORGANIZERS:

J.P. Lee
Institute for Visualization and Perception Research
University of Massachusetts at Lowell
Lowell, MA 01854
and
The MITRE Corporation
202 Burlington Avenue
Bedford MA 01730-1420

email:	jlee@cs.ulowell.edu
fax:	+1-617-271-2964
phone:	+1-617-271-7417

Georges Grinstein
Institute for Visualization and Perception Research
University of Massachusetts at Lowell
Lowell, MA 01854
and
The MITRE Corporation
202 Burlington Avenue
Bedford MA 01730-1420

email: 	 ggg@mitre.org, fax: +1-617-271-2964, phone: +1-617-271-7714

Workshop Manager:

Maria Zimmerman
IEEE Database and Visualization Workshop Manager
The MITRE Corporation
202 Burlington Road
Bedford MA 01730-1420
Fax:          +1-617-271-2964
Phone:        +1-617-271-2513
Email:        rmz@mitre.org



======================================================================


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: help with avs
Message-ID: <C94oKF.wD@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Thu, 24 Jun 1993 13:38:38 GMT

Hi Folks,

This is being posted for Jamie Adams (jadams@uoft02.utoledo.edu).

-Steve

I am trying to compile some "old" avs 3.0 chemistry modules and received the 
following error message. I am these on a titan for use with avs 4.0.
==============================================================================
	cc  -c -I/usr/avs/include/chemistry  CHEMcon_r.c 
"CHEMcon_r.c", line 476: warning: implied return at end of function returns no value
"CHEMcon_r.c", line 495: warning: implied return at end of function returns no value
	cc -o CHEMcon_r CHEMcon_r.o -L/usr/avs/lib -lchem_c -lflow_c -lgeom -lutil -lm  
/lib/loadshlib:
Undefined:
UTILcheck_expiration
Undefined references, no loading.
*** Error code 2

Stop.
===============================================================================

any suggestions would be appreciated since I am a novice at this.



                 		                    Jamie Adams
						DAS RESEARCH GROUP
					          Univ. Of  Toledo
					email:jadams@uoft02.utoledo.edu
					voice: (419) 537-4028



From hagedorn@betsy.gsfc.nasa.gov (John Hagedorn)
Newsgroups: comp.graphics.avs
Subject: Output port of type "choice"?
Date: 24 Jun 1993 13:33:57 GMT
Organization: NASA Goddard Space Flight Center -- InterNetNews site
Lines: 20
Distribution: world
Message-ID: <1993Jun24.093353@betsy.gsfc.nasa.gov>
Reply-To: hagedorn@betsy.gsfc.nasa.gov (John Hagedorn)
NNTP-Posting-Host: betsy.gsfc.nasa.gov
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit



Hi,

I'm using AVS version 3.9 on a Convex.  I would like to create a module
with an output port of type "choice" so that I can drive the value of
a choice parameter on another module.  

I've successfully created the output port, but I haven't been able to 
correctly set the output value.  This is all undocumented in the Convex
AVS manuals and I have been unable to find an example of this among
the modules at the IAC.

Any suggestions would be appreciated.


Thanks,

John Hagedorn   hagedorn@betsy.gsfc.nasa.gov        (301) 286-7374



From mcheng@dunlop.cs.wisc.edu (Michael Cheng)
Subject: Is there C library support for writing graphics output module?
Message-ID: <1993Jun24.144125.265@daffy.cs.wisc.edu>
Sender: news@daffy.cs.wisc.edu (The News)
Organization: University of Wisconsin, Madison -- Computer Sciences Dept.
Date: Thu, 24 Jun 1993 14:41:25 GMT
Lines: 9

Hi

I'm investigating to see if I can incorporate my own graphics into AVS.
For this purpose, I would need a low level C library to open windows,
draw lines, polygons, and text, etc.  I read the AVS Developer's 
manual,  but couldn't find any such library. If would appreciate any pointer
if one exists. Thanks in advance.

Mike


From larryg@avs.com (Larry Gelberg)
Subject: Re: Is there C library support for writing graphics output module?
Message-ID: <1993Jun24.180400.24030@sol.ctr.columbia.edu>
Sender: nobody@ctr.columbia.edu
Organization: Advanced Visual Systems Inc.
X-Newsreader: Tin 1.1 PL4
References: <1993Jun24.144125.265@daffy.cs.wisc.edu>
Date: Thu, 24 Jun 1993 18:04:00 GMT
X-Posted-From: aurora.avs.com
NNTP-Posting-Host: sol.ctr.columbia.edu
Lines: 22

mcheng@dunlop.cs.wisc.edu (Michael Cheng) writes:
: I'm investigating to see if I can incorporate my own graphics into AVS.
: For this purpose, I would need a low level C library to open windows,
: draw lines, polygons, and text, etc.  I read the AVS Developer's 
: manual,  but couldn't find any such library. If would appreciate any pointer
: if one exists. Thanks in advance.

Well, if you really want to do things like that, you can always use 
the X Window System.  One of the intents of AVS was to free programmers
from that level of programming.  Another approach would be for you
to allocate an image (field 2D 4-vector byte) and just output that image
to the various display tools which are built into AVS: display image, 
image viewer, etc.  This is how the other renderers in AVS are 
implemented: tracer, cube, etc.

larryg

--
=== Larry Gelberg ============================ larryg@avs.com =======
      Advanced Visual Systems Inc. (AVS Inc.)
      300 Fifth Ave, Waltham, MA 02154
===== Tel: 617-890-4300 = Fax: 617-890-8287 =========================


From haabn@nye.nscee.edu (Frederick J. Haab)
Subject: Re: Is there C library support for writing graphics output module?
Message-ID: <1993Jun24.210439.7887@nevada.edu>
Sender: news@nevada.edu (USENET News System)
Nntp-Posting-Host: nye.nscee.edu
Organization: National Supercomputer Center for Energy and the Environment
References: <1993Jun24.144125.265@daffy.cs.wisc.edu> <1993Jun24.180400.24030@sol.ctr.columbia.edu>
Date: Thu, 24 Jun 1993 21:04:39 GMT
Lines: 11


On a similar note, I am also using AVS 3.9 on a Convex, but am using
an SGI for the display.  Is there any examples of using the rglserver
to take advantage of the SGI's graphics capabilities?

    -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
       Frederick J. Haab
       haabn@nye.nscee.edu
       National Supercomputing Center for Energy and the Environment
       University of Nevada at Las Vegas
    -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-


From rcpshdb@dutrun2.tudelft.nl (Han de Bruijn)
Subject: rglserver (WAS: Is there C library support for writing graphics output module?)
Message-ID: <1993Jun25.071619.9605@dutrun2.tudelft.nl>
Originator: rcpshdb@dutrun2.tudelft.nl
Sender: news@dutrun2.tudelft.nl (UseNet News System)
Nntp-Posting-Host: dutrun2.tudelft.nl
Reply-To: rcpshdb@dutrun2.tudelft.nl (Han de Bruijn)
Organization: Delft University of Technology
Date: Fri, 25 Jun 1993 07:16:19 GMT
Lines: 28

In article <1993Jun24.210439.7887@nevada.edu> Frederick J. Haab:
> On a similar note, I am also using AVS 3.9 on a Convex, but am using
> an SGI for the display.  Is there any examples of using the rglserver
> to take advantage of the SGI's graphics capabilities?

1. Go to the Convex directory "/usr/avs/bin/SGI" and "ftp" the executable
   called "rglserver" to your SGI. Be sure that a "$path" exists to it
   (which should be defined in your ".cshrc" file on the SGI).

2. Create an ".rhosts" file at the SGI with the following line in it:

             <hostname Convex> <your login name at BOTH machines>

   Example:  dutrex.tudelft.nl  rcpshdb

   Be sure that your login name at the SGI sounds the same ("rcpshdb").
   (At least, we don't know if it's possible to arrange it otherwise.)

3. Run AVS at the Convex. The message "Avs is unable to use the native
   SGI hardware. [...]" should be absent now, and everything runs MUCH
   faster (at least for modest geometries).

Good luck!
-- 
* Han de Bruijn; Applications&Graphics | "A little bit of Physics * No
* TUD Computing Centre; P.O. Box 354   | would be NO idleness in  * Oil
* 2600 AJ  Delft; The Netherlands.     | Mathematics" (HdB).      * for
* E-mail: Han.deBruijn@RC.TUDelft.NL --| Fax: +31 15 78 37 87 ----* Blood


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: Passing strings in Fortran and C
Message-ID: <C96MAM.Ky2@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Fri, 25 Jun 1993 14:44:46 GMT

Hi Folks,

A quick posting on behalf of Marco Picasso (EPFL-DMA) picasso@masg24.epfl.ch: 

-Steve

>From what I know, passing strings between Fortran and C is rather machine dependent. On Silicon Graphics workstation you have to pass the length of the string together with the string itself. The reason is probabln the fortran string.




From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: Passing strings in Fortran and C
Message-ID: <C96MAM.Ky2@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Fri, 25 Jun 1993 14:44:46 GMT

Hi Folks,

A quick posting on behalf of Marco Picasso (EPFL-DMA) picasso@masg24.epfl.ch: 

-Steve

>From what I know, passing strings between Fortran and C is rather machine dependent. On Silicon Graphics workstation you have to pass the length of the string together with the string itself. The reason is probabln the fortran string.




From avs@doppler.ncsc.org (AVS account)
Subject: AVS NETWORK NEWS ON FTP SITE
Message-ID: <C96oGr.10p@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Fri, 25 Jun 1993 15:31:38 GMT


********************************************************************************


	The following back issues of AVS Network News are available
	from the International AVS Center's anonymous ftp site in the
	directory avs_net_news:

	Inaugural Issue (Fall 1991)
		
	Volume 1, Issue 1 (Winter 1992)
		o AVS Overview
		o New Software Products
		o Using AVS for Prototyping a Distributed Computational
		  Molecular Biology and Genetics System
		o Interpolating Gridded Data from Scattered Data in AVS
		o Visualizing Molecular Dynamics Simulations
		o AVS Version 4.0
		o AVS Newsgroup Update

	Volume 1, Issue 2 (Spring 1992)
		o An Interactive Design Tool for the Magnetic Head Bearing 
		  Slider
		o User Group Update
		o SIGGRAPH '92
		o WAIS and AVS
		o AVS '92 - A Review
		o Obtaining Modules
		o IAC Module Highlights
		o What's New?
		o Development of a VHF/UHF Broadcasting Planning and 
		  Visualization System on the Basis of AVS
		o Animating Time Dependent 3D Data Using AVS
		o The AVS to Video Connection

	Volume 1, Issue 3
		o Cityscape Visualization of Telecommunications Network
		  Performance Data Using AVS
		o New Khoros Modules for AVS
		o A First Experience with AVS
		o Using AVS at EPA's Scientific Visualization Center
		o AVS Tips and Tricks
		o What Is AVS?
		o Next-Generation User Interface

	Volume 1, Issue 4
		o AVS Consortium Japan
		o AVS5 Update
		o Visualizing the Influence on Groundwater Flow When Driving
		  a Tunnel in an Environmentally Sensitive Area
		o AVS Platform Update
		o Volume Rendering in AVS5
		o Using AVS to Visualize Highly Viscous Liquid Simulation Data
		o AVS Module to Create Quick Time Movies for the Mac 
		o Top 10 Module Packs Available
	
	Volume 2, Issue 1
		o How I Do an AVS Kickstart
		o IAC Module Contest Winner
		o Crystal Lattice Viewer for AVS
		o AVS Support on the United Kingdom
		o Using UCD for the Visualization of Semiconductor Device
		  Simulation Data
		o AVS Courses Offered
	
	
 	These issues are available in postscript format.  The more recent
	issues have the images incorporated with the text just as they
	appeared in the hardcopy version.  These images are also available
	separately and are located in the color_images subdirectory.  The
	images are ordered by volume#, issue#, and author's name.
	
	
	
-- 
International AVS Center
North Carolina Supercomputing Center
avs@ncsc.org


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: IAC listserv system
Message-ID: <C96ss4.6rv@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Fri, 25 Jun 1993 17:04:52 GMT

Hi AVSers,

Due to the AVS '93 conference, the International AVS Center's
listserv service was suspended for about a month, and is now
operational again.  For those of you who don't know anything
about this service, you can read all about it in more detail
below.  

I find the most useful feature of this system is that it allows
me to automatically subscribe to every posting to this newsgroup.
All postings to comp.graphics.avs are sent to me via email 
on an hourly basis.

If you're interested in using it this way, send the following
message to listserv@avsusers.ncsc.org (any subject line will do):
subscribe avs_newsg.list <your email address>

You should shortly receive an automatic confirmation of your
subscription, and then begin receiving the postings. 

Enjoy!

-Steve

PS  Would you be interested in sharing your AVS work with the
scientific visualization community worldwide via a short article 
and / or slides in an upcoming issue of AVS Network News?  This 
is the IAC's quarterly magazine made up of user contributed 
articles.   Thanks for considering this!
----------------------------------------------------------------
   Steve Thorpe, Application Visualization System Specialist
International AVS Center, North Carolina Supercomputing Center
PO Box 12889   3021 Cornwallis Rd, RTP, NC 27709   avs@ncsc.org
----------------------------------------------------------------


-----------------------------------------------------------------------------
    Application Visualization System User Group Registration Form
		      Last Modified:  8/5/92
 Please use this form if you are interested in any of the following:
	a) registering for various electonic mailing lists, in
           which case please submit a message electronically to
           listserv@avsusers.ncsc.org in a format described below
	b) registering for the International AVS User's group,
	   in which case please submit this form via regular mail
           with your payment to the International AVS Center
	c) signing up for the IAC's (nonelectronic) mailing list
	   in which case please submit this form electronically with 
	   a message stating this to avs@ncsc.org  This is not 
	   necessary if you already have registered for the 
	   International AVS User's group, have subscribed to
	   AVS Network News, or attended an AVS Conference.


                       Usage Notes:
1. If you want to get added to automated mailing lists for the 
   appropriate groups listed at the bottom of this form, send a message
   to listserv@avsusers.ncsc.org, with the *only* lines in the message
   body being of the form "subscribe <group_name>.list <email adddress>"
   (WITHOUT THE QUOTES).  For example, if the user test@ncsc.org wants
   to subscribe to the following 3 groups:

	cfd       Computational Fluid Dynamics AVS Special Interest Group
	anim      Animation AVS Special Interest Group
	avs_newsg Electronic delivery of postings to comp.graphics.avs 

   Then the following message sent to listserv@avsusers.ncsc.org would
   do the trick:

	subscribe cfd.list test@ncsc.org
	subscribe anim.list test@ncsc.org
	subscribe avs_newsg.list test@ncsc.org

2. If you would like to join the International AVS User's group,
   please fill this out on paper and submit it with a check for $30
   (plus $10 if you're out of the U.S.) to:
   		International AVS Center
		Attn: David Bennett
		North Carolina Supercomputing Center
	        PO Box 12889
        	3021 Cornwallis Road
        	Research Triangle Park, NC 27709
   Membership in the International AVS User's group includes a subscription
   to the AVS Network News publication.

3. If you want to start another LUG or SIG, please send
   e-mail to the International AVS User Groups President at
   iug_pres@avsusers.ncsc.org

4. Phone numbers are optional items on the registration form.

5. Please keep all entries to one line - NO CARRIAGE RETURNS

6. DO NOT use colons (:) in any fields

7. If any field is unknown or not applicable, leave blank line after colon.
-----------------------------------------------------------------------------
    LAST_NAME:
FIRST_NAME,MI:
        TITLE:
 ORGANIZATION:
     ADDRESS1:
     ADDRESS2:
         CITY:
        STATE:
          ZIP:
      COUNTRY:
EMAIL_ADDRESS:
   WORK_PHONE:
     WORK_FAX:
   HOME_PHONE:
-----------------------------------------------------------------------------
GRP NAME  GROUP
austria   Austria AVS User Group
us_md_ed  Mid Atlantic (Maryland) AVS User Group
us_ca_sf  San Francisco AVS User Group
us_ny_ro  Rochester NY AVS User Group
us_tx_au  Austin Texas AVS User Group
us_ca_la  Los Angeles AVS User Group
germany   Germany AVS User Group
us_il_ur  Urbana Illinois AVS User Group
us_wa_ri  Richland Washington AVS User Group
us_nm_al  Albuquerque New Mexico AVS User Group
us_mi_an  Ann Arbor Michigan AVS User Group
france    France AVS User Group
uk        United Kingdom AVS User Group
us_tx_da  Dallas Texas AVS User Group
ca_crivr  Chalk-River Canada AVS User Group
medimg    Medical Imaging AVS Special Interest Group
envsci    Environmental Science AVS Special Interest Group
chemmod   Chemical Modeling AVS Special Interest Group
cfd       Computational Fluid Dynamics AVS Special Interest Group
anim      Animation AVS Special Interest Group
avs_newsg Electronic mail delivery of postings to comp.graphics.avs newsgroup
-----------------------------------------------------------------------------


From scastill@nmsu.edu (Steve Castillo)
Newsgroups: comp.graphics.avs
Subject: SGI for running AVS
Date: 25 Jun 93 14:13:29
Organization: NMSU Computer Science
Lines: 15
Distribution: world
Message-ID: <SCASTILL.93Jun25141329@tesla.nmsu.edu>
NNTP-Posting-Host: tesla.nmsu.edu


What level of graphics hardware for the Indigo2 line will give
somewhat equivalent graphics performance to say a Stardent GS2000 for
running AVS.  We use AVS for viewing data produced by finite
difference and finite element codes for solving electromagnetic
scattering problems.  We are currently looking at the XZ graphics
option.  Please reply via email.
--
*****************************************************************
Steve Castillo
New Mexico State University
Las Cruces, NM

scastill@nmsu.edu
(505)646-3214


From rvenable@helix.nih.gov (Richard M. Venable)
Subject: Re: Passing strings in Fortran and C
Message-ID: <1993Jun25.202932@helix.nih.gov>
Sender: postman@alw.nih.gov (AMDS Postmaster)
Reply-To: rvenable@helix.nih.gov (Richard M. Venable)
Organization: National Institutes of Health
References:  <C96MAM.Ky2@doppler.ncsc.org>
Date: Sat, 26 Jun 1993 00:29:32 GMT
Lines: 22

In article <C96MAM.Ky2@doppler.ncsc.org>, thorpe@doppler.ncsc.org (Steve
Thorpe) writes:
|> 
|>From what I know, passing strings between Fortran and C is rather
machine
|>dependent. On Silicon Graphics workstation you have to pass the length
of
|>the string together with the string itself. The reason is probabln the
fortran
|>string.

Fortran passes the length of character arrays as an "invisible"
argument
later in the argument list; Fortran-to-Fortran calls handle this
cleanly,
but you must supply (or accept) these integers when mixing Fortran and
C.

Also, you must handle the NULL string terminator (C) or padding with
blanks
(Fortran) explicitly.  Most Unix platforms work this way, not just SG.



From michele@concave.cs.wits.ac.za (Michele Anne Collie)
Newsgroups: comp.graphics.avs
Subject: Extracting data from an AVS geom
Message-ID: <1993Jun26.080240.14659@shannon.ee.wits.ac.za>
Date: 26 Jun 93 08:02:40 GMT
Article-I.D.: shannon.1993Jun26.080240.14659
Sender: news@shannon.ee.wits.ac.za
Organization: Wits University Electrical Engineering
Lines: 8

Hi, we have a 3D skull onto which  we are trying to build a face. The problem
is that once we have passed the file through a filter and converted it to 
a polyhedron we have a 5 Meg geom file. This file takes about 15 minutes to
be displayed and even longer to rotate. Is there any way that we can extract the data that we need ? (Besides using trial and error). Is there not a way that while in AVS you can mark the points you want and delete the rest?

THanks in advance

Michele


From haabn@nye.nscee.edu (Frederick J. Haab)
Subject: Re: rglserver (WAS: Is there C library support for writing graphics output module?)
Message-ID: <1993Jun28.054602.6082@nevada.edu>
Sender: news@nevada.edu (USENET News System)
Nntp-Posting-Host: nye.nscee.edu
Organization: National Supercomputer Center for Energy and the Environment
References: <1993Jun25.071619.9605@dutrun2.tudelft.nl>
Date: Mon, 28 Jun 1993 05:46:02 GMT
Lines: 29

In article <1993Jun25.071619.9605@dutrun2.tudelft.nl> rcpshdb@dutrun2.tudelft.nl (Han de Bruijn) writes:
>In article <1993Jun24.210439.7887@nevada.edu> Frederick J. Haab:
>> On a similar note, I am also using AVS 3.9 on a Convex, but am using
>> an SGI for the display.  Is there any examples of using the rglserver
>> to take advantage of the SGI's graphics capabilities?

[Description on how to get AVS to use rglserver deleted]

>Good luck!
>-- 
>* Han de Bruijn; Applications&Graphics | "A little bit of Physics * No
>* TUD Computing Centre; P.O. Box 354   | would be NO idleness in  * Oil
>* 2600 AJ  Delft; The Netherlands.     | Mathematics" (HdB).      * for
>* E-mail: Han.deBruijn@RC.TUDelft.NL --| Fax: +31 15 78 37 87 ----* Blood

I'm sorry if you misunderstood the problem, the original post was
about WRITING a MODULE that outputs graphics, I would like to
WRITE a MODULE that makes calls to the rglserver, some of the commands
I'd like to do, I don't believe, are readily available.  Turning
on stereo-graphics, for instance.  I'll probably end up writing
my own module that invokes a homemade program on the SGI, similar
to rglserver, if something better doesn't come along.

    -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
       Frederick J. Haab
       haabn@nye.nscee.edu
       National Supercomputing Center for Energy and the Environment
       University of Nevada at Las Vegas
    -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-


From rel@mtu.edu (Robert Landsparger)
Subject: read_ucd_points...strange behavior for me
Message-ID: <1993Jun28.184539.14911@mtu.edu>
Keywords: read ucd points broken
Organization: Michigan Technological University
Date: Mon, 28 Jun 1993 18:45:39 GMT
Lines: 58

After reading through comp.graphics.avs today, I decided to try and get
some data running in AVS as an exercise.  I have some 3d points with
a value attached to them.  The lead me to believe that read_ucd_points
was my answer because I would like an iso-surface from it.

Here is what the network looks like:

				     read ucd points
	generate colormap		   |	   
		|		           |
	   ucd legend			   |
		|			   |
		|______________________	   |
				       |   |
				    ucd to geom
	        			 |
			      		 |
				  geometry viewer

and the first few lines from my data file are:

	3 6
	46 67 11 100 2 3 4 5 6
	48 67 11 100 2 3 4 5 6
	50 68 11 100 2 3 4 5 6
	51 67 11 100 2 3 4 5 6

I don't see any geometry, and after hooking up "ucd print" I get some
information that I don't quite think is right:


Ucd Name: UCDpoints
data_veclen: 0
name_flag: 5f =(material_ids node_names cell_names cell_types mid_edges int )
ncells: 402 cell_veclen: 0 
nnodes: 402 node_veclen: 6 
util_flag: 0
X extent: [13.000000,86.000000]
Y extent: [31.000000,68.000000]
Z extent: [11.000000,88.000000]

I don't see what the data_veclen should be zero and the node_veclen is 6.
In the documentation, it says this module will work with ucd_iso, but
when I add it, I get:

		Error in ucd_iso: no scalar components

Using the module read_points, I get spheres at the points with variable
radii.

Thanks for any insight one can offer on this....


							rel

-- 
Robert E. Landsparger, rel@mtu.edu
"Fall behind early, it gives you more time to catch up."  -rel


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: C/Fortran passing strings
Message-ID: <C9CnnJ.4x6@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Mon, 28 Jun 1993 20:59:43 GMT

Hi Folks,  a quick posting for Marco Picasso:

>From what I know,
passing strings between Fortran and C is rather machine dependent.
On Silicon Graphics workstation you have to pass the length of the string
together with the string itself. The reason why is because fortran strings
don't contain the "\0".

On the contrary all types of arrays can be passed between C and Fortran
within a UNIX environment. For instance a fortran declaration like

integer a(4,*)

should be passed as

int (*a)[4]

in C and "mallocated" with

a = ( int (*)[2] )calloc( 100 , sizeof(int [2]) );

The fortran ligns

   do 10 i=1,n
    do 10 j=1,4
     a(j,i) = something
10 continue

should then be written in C

   for (i=0;i<n;i++) {
    for (j=0;j<4;j++) {
      a[i][j] = something;
    }
   }

Common blocks in Fortran can be passed using structures on SGI
but this is machine dependent and should be avoided.

Marco Picasso
Departement de Mathematiques
Ecole Polytechnique Federale de Lausanne
1015 Lausanne, Switzerland
email picasso@dma.epfl.ch


From thorpe@doppler.ncsc.org (Steve Thorpe)
Subject: Posting from Mike Rangitsch
Message-ID: <C9Cpsy.91F@doppler.ncsc.org>
Organization: North Carolina Supercomputing Center
Date: Mon, 28 Jun 1993 21:46:09 GMT

A note to all KPC Titan machine owners/users from mike rangitsch
(mrangitsch@dow.com):

At the 2nd AVS Conference in Orlando, there was an ugly rumor that
AVS 5.0 was not going to be ported to the Titan platform by Kubota
Pacific.  When we asked them, the Kubota representatives said that
this was truely the case at the time, but the decision was not
locked in stone.

As at Titan owner I find this to be very bad (to say the least).
When I bought my Titan, one of the principle reasons was that it
had AVS on it.  Without the Titan, I probably would never have
purchased the product and gotten so much use out of it.  The
Titan may not be 'state-of-the-art' but it is still respectible
(and very useable) both computationally and graphically.

The KPC representatives (Joe K. in particular) said that they
would be responsive to the wishes of their users in this matter,
so if you are a Titan owner/user please tell KPC that the port
of AVS 5.0 to the Titan is important (if you think it is!!)
Without input on the users' part (that's us), we will be stuck
with AVS 4.0 (and its bugs).

Thanks all.


mike rangitsch
mrangitsch@dow.com
(504)-389-1752


From bdickens@elara.mitre.org (Brian Dickens)
Subject: Marking Active Coroutines
Message-ID: <1993Jun29.200041.736@linus.mitre.org>
Keywords: coroutine, active, flow executive
Sender: news@linus.mitre.org (News Service)
Nntp-Posting-Host: elara.mitre.org
Organization: Research Computer Facility, MITRE Corporation, Bedford, MA
Date: Tue, 29 Jun 1993 20:00:41 GMT
Lines: 11

I have written a coroutine module which I would like to
"highlight" in the way which the flow executive marks
subroutine modules when they are running.  However, it
seems that coroutines are never lit up.  Is there a
function that I have overlooked that will do this, or is
there some kind of trick I can use to make this happen?

Brian Dickens
<bdickens@elara.mitre.org>




From lipman@oasys.dt.navy.mil (Robert Lipman)
Newsgroups: comp.graphics.avs
Subject: SGI AVS 5.0 modules to read images ?
Message-ID: <38611@oasys.dt.navy.mil>
Date: 30 Jun 93 19:07:38 GMT
Reply-To: lipman@oasys.dt.navy.mil (Robert Lipman)
Organization: Carderock Division, NSWC, Bethesda, MD
Lines: 18

I'm looking for SGI executabel AVS modules to read any or
all of the following image file formats: tiff, targa, sgi rgb.
I am aware of the read_*_image modules at the IAC, but I
always seem to have a problem making the executable to read
an image on an SGI for one reason or another.

If you've got the module that I need, can I get a copy of it?

Thanks in advance, e-mail please,

Robert Lipman                     | Internet: lipman@oasys.dt.navy.mil
David Taylor Model Basin - CDNSWC |       or: lip@ocean.dt.navy.mil
Computational Signatures and      | Voicenet: (301) 227-3618
   Structures Group, Code 2042    | Factsnet: (301) 227-5753
Bethesda, Maryland  20084-5000    | Phishnet: stockings@long.legs
				   
The sixth sick shiek's sixth sheep's sick.



From wgurney@irisbta.er.usgs.gov (Wes Gurney (ISD/BTA) x7120)
Subject: IGS File Format
Message-ID: <1993Jun30.190745.11775@rsg1.er.usgs.gov>
Sender: news@rsg1.er.usgs.gov
Organization: U.S. Geological Survey, Reston VA
Date: Wed, 30 Jun 1993 19:07:45 GMT
Lines: 11

I'm looking for a module that converts IGES or IGS (International 
Graphic Export Standard) to the geometry viewer.  I've checked
down in North Carolina and I didn't see anything there.  Does
anyone have this module or know where I might be able to find
it?  Thanks a lot.


Wes Gurney
Comp. Scientist
U.S. Geological Survey



From beresh@superior.egr.msu.edu (Steven Beresh)
Newsgroups: comp.graphics.avs
Subject: Surface Rendering
Date: 30 Jun 1993 20:27:07 GMT
Organization: College of Engineering, Michigan State University
Lines: 14
Distribution: world
Message-ID: <20ssur$tcc@msuinfo.cl.msu.edu>
NNTP-Posting-Host: superior.egr.msu.edu



Does anyone have any suggestions on performing a surface rendering?  I have a field
consisting of discrete data indicating whether each point is part of the surface or not. 
I tried feeding this through isosurface, but it didn't seem to like discrete data.  I
tried using 0's and 1's, 0's and 255's, and even tried replacing the 0's randomly with
other values (excepting the value representing my surface) so that the field was no longer
discrete.  In all cases, isosurface either produced nothing or a surface bearing no
resemblance to my surface.

If anyone has any suggestions for producing a surface from this data, I'd appreciate it.

--Steve



