> I use almost only relatively slow computers. More than 100 of them. If
> they are only 1/5 as fast as what some people think is apropiate,
> it means
> that they can do 100 tests in the same time as an appropiate machine does
> 5. While a very few people think I should stop using those computers for
> testing, I'm no 23 on the GIMPS (not Primenet) top-100 list. (I can't
> find this Mr. "Go Away Loosers With Slow Computers" there, but he is
> probably working under some pseudonym.)
Egads...I'm not telling *anyone* to NOT use slow computers...I'm
"suggesting" they use < p166 for double-checks or factoring. I do, Primenet
will assign them that way if you let it, it's no big deal when you think
about it. Unless, that is, you're only motivation is prize money in which
case double-checking gets you none in all likelihood. Heck, now that the
$50K prize is gone, nobody is going to get the next prize anytime soon
anyway, so why not do double-checks?
BTW, my Primenet ID is madpoo if anyone is interested (and who would be).
Currently we're oscillating between 20-21 on Primenet. Strangely, I am 220
on the GIMPS list, but my brother Jeremy is 54...I don't know how that
happened since we both use the same madpoo primenet ID...hmmm...does he go
by email or something? As team "madpoo" we have about 4-5 different people
contributing...would we all show up separately in the GIMPS rankings?
Even more curiously, I checked my *old* Primenet ID recently, "blosser" and
lo and behold, what did I find? One of the US WEST machines had turned in a
result as recently as May of this year!! yikes! Looks like they found it
since the current exponenent assigned to that machine is way past due.
If anyone cares to join the mostly extinct team "blosser" the password is
"acidrain" (which the search warrant pages clearly show).
> This spring I retired a lot of Indys because there were no exponents left
> with FFT size at or below 128KB. (256KB FFT takes many months for each
> exponent and a few of the now rather old machines got RAM problems.) 18
> of them (less than half) participated in the NFS factoring of M619 (which
> should be done in a couple of weeks), and together they became one of the
> major contributors. 18 Indys, which were standard low-end workstations
> five years ago, are not worthless. They can still outperform a couple of
> Pentiums.
This is an excellent illustration of my point. Some machines are much
better suited to something *besides* GIMPS, such as the NFS factoring
challenges, or even the DES crackers.
I'd rather see computing power applied to a project with some thought given
to how that computing power can best be used. A 486, with it's rather
pitiful FP performance, should be given something to do that will stick with
integer ops, just for example. It makes little sense to have a 486 doing LL
testing when it would be flexing it's power best by doing factoring.
I hope I'm being clear that I don't consider slower computers useless
entirely...just useless for doing LL tests in the range of exponents
currently being assigned. There are PLENTY of other things they could be
doing that make much more sense, as you have figured out.
Aaron
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
