[Open-graphics] decoding video in real-time
Nicolas Boulay
nicolas.boulay at gmail.com
Sun Mar 18 16:04:49 EDT 2007
2007/3/18, Attila Kinali <attila at kinali.ch>:
> On Sun, 18 Mar 2007 16:18:40 +0100
> "Nicolas Boulay" <nicolas.boulay at gmail.com> wrote:
>
> > > Yes, but 4GHz is about the limit how fast chips with
> > > 2x2cm can be clocked. And even that needs very carefull
> > > design of the clock distribution and signal routing.
> > > Keep in mind that 2cm is about the distance a signal
> > > can travel within one clock cycle (depends a lot on
> > > the actually used wire model). And then you still
> > > need some safety margin to reliably operate the chip
> > > under all specified conditions.
> > >
> >
> > A pentium 4 use double frequency ALU (this permit to simulate 2 ALU).
> > Pentium 4 at 3 Ghz exist sine more than 4 years. 3 Ghz cpu imply 6 Ghz
> > ALU in the pentium 4.
>
> Video decoding is hard to parallelize on general purpose
> CPUs. Thus even if it has two ALUs, you will not be able
> to use both to their full potential. Specialized video
> decoding hardware is much better in that case.
>
> Also keep in mind, that having two processors does not
> mean you have doubled your processing power. How much
> you get depends a lot on:
> 1) How well you can parellize the work tasks.
> 2) How much communication overhead you have.
> 3) How much locking you have between the two CPUs.
> 4) How much cache/memory contention you have due to two
> CPUs work in parallel.
>
> For more information on this, please read a book
> on high performance and/or parallel computing.
>
I just speak about the fact that 4 Ghz logic is not impossible.
Because 6 Ghz (or 130 ps of propagation time) still exist. Usualy the
problem are in the clock propagation time, but globaly asynchronous
design could work at that speed.
I never speak about the difficulties or not of parallel computing. I
work for the f-cpu project long enough to know about it.
> Attila Kinali
>
> --
> Linux ist... wenn man einfache Dinge auch mit einer kryptischen
> post-fix Sprache loesen kann
> -- Daniel Hottinger
>
More information about the Open-graphics
mailing list