[ydl-gen] No video hardware acceleration on PS3
Derick Centeno
aguilarojo at verizon.net
Tue Nov 14 11:53:58 MST 2006
ugo wrote:
> Hello again :)
>
> Still out of topic, but i guesse that it's not a trouble ^^
> As i see ps3 could be used for scientific research thanks to its
> Cell's cpu.
> I understood that first it would be for applications development, then
> maybe also for calculs light enough to do not need big calculators.
> I imagine that linked together several ps3 could provide more calcul
> power.
> Did you imagine a kind of folding at home use to take benefit of this
> kind of use ?
Hi Ugo:
Folding @ Home (http://folding.stanford.edu/download.html) and similar
applications which run on BOINC
(http://boinc.berkeley.edu/system_requirements.php) are designed to
function as screen savers while your computer is not busy doing what you
want. It is intentionally designed for the person who is willing to
allow some computer time to be dedicated to assisting in some way to the
total processing/calculation effort of serious scientific work. Anyone
can contribute computer time, at any level of interest or skill. Many
offices and communities of all kinds have these programs running on
their systems and it does help.
However, currently none of them run on any version of Linux which runs
on PowerPC systems. SETI @Home used to, but ever since they switched to
BOINC, they no longer do.
However, as the power within the PS3 is already massive -- it is more
than enough for anyone willing to use it while making the time to learn
and do science throughout their studies up to and maybe even include a
Ph.D. Of course, by that time a new generation will be out unless we
are speaking of an occasional wunderkind who has completed their
doctoral work in Quantum Chemistry by 18. For the wunderkinds out
there, the PS3 could not happen fast enough, and whatever they do with
it will be unknown until they get it.
However, I don't believe it has to be restricted to students. It can
and should be used by anyone to brush up on any area that interests
them. As their interests become more intricate and involved this is one
system which is designed to move data no matter how large. And few
individuals are going to be moving Terabytes of data; what is
interesting is that even if one or a group of individuals did -- this
system should be able to spit out results as fast as completed models to
be tested via programming routines have been designed.
It's going to be interesting to find out if the rest of us can even
understand what those wunderkinds will have done. It wouldn't be the
first time professors totally underestimated or misunderstood their
students. We'll see.
More information about the yellowdog-general
mailing list