[geeks] Slow Indy

[Kurt Huhn]

Bjorn Ramqvist <brt at g.haggve.se> wrote:

> Kurt Huhn wrote:
> > I learned that Seti at home purposefully crippled the executable for
> > fast architectures for some reason (so that they didn't run out of
> > units to give out or something) - so I decided to switch to protien
> > folding.
> 
> Mmm... You don't mean the fact that they had to increase the
> datablocks just to get faster architectures to do their work a little
> longer? Apparently they had bandwidth problems, too many hits/second,
> so their solution was to increase the datablock size.

There's also that, but I recall someone (perhaps on list here)
complaining of code that purposefully put wait times and delays into the
loops while the client was runnning, in addition to unoptimized routines
that looked like they were designed that way on purpose, and a couple
other interesting things.  I think someone or a group of people got
together and created a Seti at home client that *was* optimized, and
performed calulations extremely fast (an order of magnitude faster) with
greater precision.  I can't find the email and supporting evidence, so
take what I say with a grain of salt...

-- 
Kurt                "I remember that I've got a solid base of worms and
kurt at k-huhn.com      ant eggs.  I puke like a hero all night long."
                                                      -- Tony Bourdain