[Sidefx-houdini-list] dPds, dPdt troubleshooting...

Ben Schrijvers Ben.Schrijvers at framestore-cfc.com
Thu Aug 31 10:20:56 EDT 2006


It's a real pitty you can't query P in (textured) uv space to calculate 
you own derivatives. Or can you?
What about the good old baking P in texture space and then query that as 
texture map during the "beauty pass".. ?

hmmm...
benS



Dan Seddon wrote:
> hoknamahn at cgtalk.ru wrote:
>> > In the case of mantra s and t is the primitive  parameterization - 
>> so this can be discontinuous from one face to another in the case of 
>> poly/subd's.
>>
>> I'm trying to figure out when the polygonal geometry behaves as 
>> parametric surface and when as polygonal (with discontinuity).
>> In some cases the whole piece of polygonal geometry looks like 
>> parametric surface (i.e. with nice anisotropy) in other case the 
>> whole geometry or the part of geometry looks weird.
> The polygonal geometry doesn't have to behave as a parametric surface 
> to have continuity ( though it may join large numbers of quad faces 
> together in to a grid internally if they can be represented that way) 
> - to appear continuous the derivatives just need to match on 
> neighbouring faces - you're more than likely seeing this when you have 
> quads arranged in a grid like fashion. Try rendering s and t out 
> rather than dPds and dPdt on different surfaces and you'll get an idea 
> of what the problem involves.
> _______________________________________________
> Sidefx-houdini-list mailing list
> Sidefx-houdini-list at sidefx.com
> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>



More information about the Sidefx-houdini-list mailing list