On Jun 20, 2011, at 11:18 AM, Moritz Moeller wrote:

> On 06/20/2011 06:33 PM, Jonathan Egstad wrote:
>> The Render base class attempts to correlate GeoInfos between two separate 
>> time-sliced
>> GeometryLists by using the GeoInfo::out_id() hash, but as I recall it does 
>> nothing to
>> correlate individual Primitives inside each respective GeoInfo.  If I recall 
>> there may
>> be a test for point counts to make sure that they haven't changed between 
>> time slices.
> 
> Jon, thanks for taking the time but I think you didn't read my mail
> carefully. :)

No worries, mr. verbose  :p

So then it's unclear to me then whether you're seeing this issue outside the 
particle system with regular primitives, or whether the behavior is simply 
confined to particles.  This seems to be very specific to the particle 
generation logic.

Assuming nothing has changed the RenderScene base class does nothing out of the 
ordinary to create it's input GeoOps - it uses split_input() and inputContext() 
to create separate GeoOp input chains, then calls validate() and build_scene() 
on them.  It up to the GeoOps to generate the correct data for the requested 
time.

-jonathan

_______________________________________________
Nuke-dev mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev

Reply via email to