[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference rusure::math

Title:Mathematics at DEC
Moderator:RUSURE::EDP
Created:Mon Feb 03 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2083
Total number of notes:14613

667.0. "Costa-Meeks-Hoffman minimal surface" by ENGINE::ROTH () Thu Feb 12 1987 07:24

    I recently came across a paper that describes a newly discovered
    'minimal surface' that is a counterexample to some long-standing
    conjecture that the only complete minimal surfaces in 3-space
    were the plane, the catenoid, and the helicoid.

    I wrote an (admittedly inefficient) program to calculate and display
    it for the Vaxstation using UIS; if you want to try it, versions
    are in ENGINE::SYS$USRDISK:[ROTH]COSTA_MEEKS.EXE and COSTA_MEEKS_PIX.EXE.

    The first runs on a VS-II and uses some 3-D viewing controls
    I happened to have around.  You specify fineness of the mesh, about
    50 is a good number.

    The _PIX version runs on an 8-plane GPX and draws a depth cued
    image which is much easier to understand, though its slower to
    do the drawing.  You can optionally click on the "additional options"
    menu item and save the picture in a .UIS file.  I may get around
    to renderng the surface colored according to curvature, but now
    its just a wireframe.  It's interesting to try and see the
    8-fold symmetry it has; also it's topologically a torus with a few
    points deleted.  The Lynx workstation would be much better for
    examining an object like this.

    The paper was "A Complete Embedded Minimal Surface in R3 With
    Genus One and Three Ends", by Hoffman and Meeks, J. Differential
    Geometry 21, 1985.  The discoverer of the surface, Costa, didn't
    know what it looked like, just its topology, and the above authors
    extended his theory after looking at computer images of it.

    - Jim
T.RTitleUserPersonal
Name		DateLines
667.1bug report?CACHE::MARSHALLhunting the snarkMon Feb 16 1987 11:1014
    re .0:
    
    I copied the _PIX version. Tried to run it and got back:
    
    image file not found DUA0:[SYS0.][SYSLIB]HCUISSHR.EXE
    
    I am using a VSII/GPX with 8 planes, running uVMS v4.5.
    
                                                   
                  /
                 (  ___
                  ) ///
                 /
667.2Oops, not everyone has HCUIS installedENGINE::ROTHMon Feb 16 1987 11:4910
    Re HCUISSHR.EXE - That is the hard-copy UIS facility, and I thought
    it was part of the normal VWS distribution.  I placed a copy in
    my directory on ENGINE which you can copy into SYS$SHARE on your GPX.

    If HCUIS is installed, then you can save images and replay them with
    the RENDER DCL command, or send them to plotters and the like so it's
    sort of nice to have around.  If this is a problem, I can conditionally
    eliminate references to it (via the LIB$FIND_IMAGE_SYMBOL service).

    - Jim
667.3fantastic!CACHE::MARSHALLhunting the snarkMon Feb 16 1987 14:4811
    re .0:
    
    I found a worstation that has HCUIS installed. The program works
    wonderfully. Can't wait to get a LYNX.
    
                                                   
                  /
                 (  ___
                  ) ///
                 /
667.4Fixed unconditional ref to HCUISENGINE::ROTHTue Feb 17 1987 08:005
    I changed the _PIX program to check if HCUIS was in your SYS$SHARE
    directory, and if not, just disable the additional option to save
    the display, rather than hardwiring the reference at link time.

    - Jim
667.5problems with hardcopyCACHE::MARSHALLhunting the snarkTue Feb 17 1987 09:0721
    Me again, with another problem.
    
    Tried to get a hardcopy by hitting the ADDITIONAL OPTIONS line in
    the window menu. This generated a .UIS file alright, to which I
    then did:
    
    RENDER/DEV=LN03 COSTA_MEEKS_HOFFMAN
    
    This dutifully generated a .REN file, which I copied over to the
    system with the LN03. However, what came out of the LN03 was a page
    3/4 black with a scrambled image in the remaining 1/4.
    
    Any idea what I did wrong?
    
    Thanks,
                                                   
                  /
                 (  ___
                  ) ///
                 /
667.6No longer there ?KBOV07::J_WOLFFJulian WolffWed Feb 25 1987 03:376
    Maybe I did something wrong, but I can't copy from ENGINE::, 
    it tells me, INVLOGIN, invalid login information at remote node.
    
    	Julian.
667.7ENGINE::ROTHWed Feb 25 1987 08:227
�   Maybe I did something wrong, but I can't copy from ENGINE::, 
�   it tells me, INVLOGIN, invalid login information at remote node.

    No, operations here did something wrong.  Our cluster went down
    yesterday with a network problem, I'll have them fix this.

    - Jim
667.8files moved...CTCADM::ROTHIf you plant ice you'll harvest windThu Sep 22 1988 07:2411
    The programs mentioned above have been moved to VITUS::DUA1:[PUBLIC...]
    (somewhere in the directory tree - it's world readable.)  There are a
    few other odds and ends there as well.

    Disk quota on our clusteer was a bit short to keep these programs so
    they had to be moved to a workstation.

    Since I'm now using DECwindows, I'll be porting some of these so
    subdirectories should appear at some time with DW and VWS versions.

    - Jim