PfChannel and frustum

New Message Reply Date view Thread view Subject view Author view

AnitaKishore (kishore++at++electrogig.com)
Tue, 20 Jun 1995 10:09:21 -0700


First of all, I want to thank all who gave me pointers on the usage of
pfMakePerspFrust.

Next, I realize that I need to make some corrections to my earlier mail.
I missed to include pfChanNearFar before calling pfMakePerspFrust. YES, I
am doing that with the same parameters as before. And, I am not making
any matrix corresponding to h,p,r 'coz they are all zeros.

So all together it is:

WORKING:

>
> pfChanNearFar(chan, 1.0f, 10000.0f)
> pfChanFOV(chan, 140.0f, -1.0f)
> pfSetVec3(view.hpr, 0, 0, 0)
> pfSetVec3(view.xyz, 0.0f, -50.0f, 60.0f)
  pfChanView(chan, view.xyz, view.hpr) ---- I forgot to write this too before

NOT WORKING:

pfChanNearFar(chan, 1.0f, 10000.0f)

> Next, I convert the above representation so as to use pfMakePerspFrust
> as follows:
> x = near * tan(140/2) where, near = 1.0
> z = near * tan(131.0724/2) (this angle I got out of pfGetChanFOV for the
> first type of setting)
> Then,
> pfMakeTransMat(m1, 0.0f, -50.0f, 60.0f)
> pfMakePerspFrust(chan, -x, x, -z, z)
> pfChanViewMat(chan, m1)
>

I have also tried setting a matrix for view direction with each of h, p, r
being zero, multiply this with the translation matrix and set pfChanViewMat
with the result, but with no luck. Anyone with more suggestions?

>
> There also appear to be more problems when using NearFar with non-simple
> frustrums and with general pfChan* FOV's, though SGI's response is so
> slow they have yet to confirm these problems (or otherwise).
>
>
> roy
>-- End of excerpt from ROY RUDDLE

Can somebody from SGI please confirm the above??

Thanks
-anita

-----------------------------------------------------------------------------
Anita Kishore
kishore++at++electrogig.com
-----------------------------------------------------------------------------


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:51:36 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.