[Bf-committers] Sunday meeting minutes, 25th feb 2007

[Ton Roosendaal]

Hi Giuseppe,


> - fluids
> - softbody
> - game engine (splitting physics and logic from rendering)
> - compositor
> - sequences playback, either as Imaging as well as the Sequencer. This
>    could be also made in some distributed way (e.g. preparing frame
>    x+1, x+2, ...,  x+M [where M is the # of cores] ready for playback).

(snip)

> - how much time could take a parallelization task of each of the
>   modules above (listing from easiest|shortest to longest).

- Fluids already has threads, not sure about switching to use more than  
2.
- Softbody I dont know, I think its not worth to do it even. Better is  
to find a way to execute/solve/build multiple non-depending objects  
together in parallel (so subsurf, armature deform, modifiers, etc etc  
all become parallel)
- Compositor is well doable (easy), apart from memory issues. Some  
nodes (currently all) require the full bufffers to be alloced (vector  
blur, defocus, etc), and allocating all buffers for each thread is  
costy

> - can the google summer of code be considered for some of the above
>   tasks?

It can be, it could not. Depends on too many variables... find a  
developer first i'd say!

> - which one of programming model would be most affordable to do this?
>   pthread, OpenMP (gcc 4.2/gomp/...), etc.

I only know pthread, which we happily use now.

-Ton-


>
> Bye
> Giuseppe.
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-committers
>
>
>
------------------------------------------------------------------------ 
--
Ton Roosendaal  Blender Foundation ton at blender.org  
http://www.blender.org