[XviD-devel] GPU Fragment Processor Acceleration of XviD encoding

[Dan]

Xvid Developers,

For the last year or so, I have been engaged in research in (among other 
things) using Graphics Processors to accelerate signal processing and other
scientific calculations.  The results, in some cases are quite stark.  In some 
cases, as much as 100x or more speedup has been observed.  30x+ over, say, a 
Athlon64 ~2.4GHz, is not uncommon for some of the 2D signal processing 
calculations that I've worked with.  I am also an XviD user and I believe that 
there may be portions of the XviD encoding process that coule be accelerated
using these techniques.  I'd like to investigate whether this is the case.

I loooked through the forums at xvid.org, and browsed a bit through
the mailing list archives.  I saw that this has been brought up before, but
not by anyone that appeared to be prepared to do actual development, and it
does not appear to have been followed through with.

I am, unfortunately, totally new to the programmatic workings of XviD.  I
will ramp up on this, but in the meantime, if anyone is able and willing to
give me any points on where the most computationally intensive part of the 
code is, it would help me tremendously.  I don't yet know how much time I can
dedicate to this, but I think it would be a very interesting result if 
significant speedups became available.

Thanks for any tips, and I look forward to contributing some working
prototypes ASAP.