[XviD-devel] zero-biased dequant/idct

[Christoph Lampert]

> > Hm, that's of course hard to estimate, but I think 
> one can be at least sure
> > that there won't be any huge advantages of this 
> approach. However there
> > might be some mild speed improvement, so if 
> you'd like to try this out, I
> > would appreciate it. btw: did the paper where 
> you read about this idea
> > present any numbers about speed 
> improvements? Maybe this approach is more
> > successful for c-code only? We already learned 
> that an early-stop check
> > within sad[8,16] was not useful...
> 
> i looked again for the paper but couldn't find it 
> through google.  it would have been C-code only 
> (from 1995), but the idea might still work out.
> 
> i don't have high hopes for this improving speed 
> by much, but even 5% faster decoding would be 
> nice.

I doubt it. Speeding up C code and speeding up MMX
assembler are two different pairs of shoes... 
One branch prediction might stall everything longer than 
50 extra MMX instruction. 

But please test, if you have the time.

gruel

-- 
Christoph H. Lampert chl@math.uni-bonn.de | Diese Signature wurde maschi-     
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da    | keiner Unterschrift. AZ 27B-6