On Wed, 24 Feb 1999 04:04:51 GMT, par### [at] mailfwicom (Ronald L.
Parker) wrote:

>It appears to be a little of both the parametric object and your 
>bad coding. :)  First, your bad coding:
>
>>       function (cos(u) ^ _E1) * (_Rj + _Rn * (cos(v) ^ _E2)),
>>                     (sin(u) ^ _E1) * (_Rj + _Rn * (cos(v) ^ _E2)),
>>                     (sin(u) ^ _E2) *  _Rn
>
>The z term is supposed to be based on sin(v), not sin(u).
>
>Now the bad news: there seems to be an issue with the parametric
>stuff that I don't fully understand.  It seems like only things
>in the first quadrant, where both the sin and the cos are positive,
>get rendered correctly.  I found an ABS in the code for exponentiation
>that could have caused the problem, but removing it didn't seem to 
>make the problem go away.  It's possible that I'm missing something
>somewhere else in the code.  Unfortunately, I didn't write it, so
>understanding it fully could be a bit of a treat.
>
Thanks for the debug, Ron.  I thought that seemed odd, but I looked at
the original equation several times and saw "theta" instead of "phi"
every time.  Maybe I should sue my opthalmologist.  Good luck on the
parametric code.  FWIW, I will dig up the isosurface source and have a
look, but my compiler is broken so I can do no testing at the moment.
Incidentally, I turned your SuperPatch doc into a Windows help file
for my own use.  If you want to add it to the distribution, let me
know and I'll clean up the index and send it along.  If you want to
send info on changes for the next version, I'll gladly add it.

Jerry Anning
clem "at" dhol "dot" com

Post a reply to this message