move/draw versus v* drawing

Brent Bates ViGYAN AAD/TAB blbates at AERO36.LARC.NASA.GOV
Wed Apr 24 22:07:33 AEST 1991


   I recently modified a program I originally wrote for our 3130 so it
would use the v3d routine instead of move/draw.  It worked fine on our
4D/320 VGX using move/draw, however, it doesn't work as well using v3d.
   I use the program to examine 3D surface definitions as wireframes.
With move/draw I could zoom in VERY close to the line drawings with out
any problems.  However, now with v3d, if I get too close, lines start
coming out of the edges of the screen and as I get closer the original
lines disappear and reappear, but not neccessarily where they should be.
Has any one seen anything like this?  The only changes I made were in
changing move/draw calls to bgnlin/v3d/endlin calls.  OS is 3.3.2 and I
am using FORTRAN.

   FYI on another note.  A long while back I posted a note about some
software I had that was written for the 3130 but core dumped on the very
first call to graphics on a 4D and I couldn't figure out what was the problem.
Well, I found out.  The people who wrote the C source had written their
own version of strcat.  Their strcat took 3 arguments, the 3rd argument
was for the returned value, and the function type was integer.  Apparently,
the 3130 didn't care, however, on the 4D the graphics library tried to
use their routine instead of the one in the system.  A call to winopen
apparently would call strcat and core dump.  What I found strange was that
a call to foreground() also called strcat.  I changed the name of their
routine and all their calls to strcat and now everything is working fine.
   Just goes to show you that you should be careful about giving routines
the same name as ones already in the system, there is no telling what can
happen.

	Brent L. Bates
	NASA-Langley Research Center
	M.S. 361
	Hampton, Virginia  23665-5225
	Phone:(804) 864-2854          FAX:(804) 864-6792
	E-mail: blbates at aero36.larc.nasa.gov or blbates at aero8.larc.nasa.gov



More information about the Comp.sys.sgi mailing list