M40 Again!!!!

[cheesehead]

Originally posted by Xyzzy
IIRC, it was G@H, and I think the fiasco was they reset the stats or maybe came along later and "readjusted" them causing people to lose tons of work...

Remind me to add this example to my arsenal of arguments next time someone wants to update the "obsolete" P90 CPU-year.

[GP2]

Originally posted by Dresdenboy
Theoretically the Athlon could come close to a P4 per clock. It has to be implemented. Maybe my new implicit transpose (making use of modern processor features) could help here.

You mean Athlon 64, no?

[Dresdenboy]

It is true for both but I meant the K7-Athlons because they will still account for more than 95% of the new Athlon accounts.

However, their FPU design is the same and the "old" Athlons can do as much per clock as the P4 FPU can do with SSE2. The only disadvantage regarding the x87 FPU is that SSE2 actually allows to increase the distance between dependent instructions leading to a better throughput.

[ET_]

Optimizing a bit is not quite nothing when millions of bits are involved...

Luigi

[wfgarnett3]

Congrats to this new largest prime number!! and to all the hard work that everyone in GIMPS has contributed :) :) :)


william

P.S.--- We need a GIMPS party on ArsTechnica chat :)

[edorajh]

I just calculated 2^20996011-1 on my own using Mathematica software. It's great feeling when program finishes calculating it, to se sooo huge number. (I know we can find actual digits on the Net, but I just have to do that on my own... hm, maybe this is for You might ne addicted to GIMPS thread material! )

[liqi]

This Picture should be changed.

http://www.mersenne.org/primenet/ipsnet.gif

[Quacky]

*hopes for the Athlon optimization*

[flava]

Originally posted by Quacky
*hopes for the Athlon optimization*

me too

[PageFault]

That is the entire scope of the problem - the mainstream have a short attention span and no interest in running dc, hence client removal in a week's time. The real job is to convert those from the mainstream into hardcore crunchers. Even at a hardware oriented site the success rate is low.

Originally posted by GP2

But my guess is that the influx of new users is largely composed of people who read the mainstream media stories and who are new to distributed computing... not hardcore crunchers

I'm going to forget this ever happened after this reply. The stats fiasco was the last straw, the finishing touch to a litany of things defining bad management of a DC project. It won't happen with GIMPS. I chose GIMPS in part based on this.

I keep hearing indirect allusions to the stats fiasco that befell one of the other distributed computing projects... could you provide the details for our benefit? Can we try to ensure that such a thing doesn't happen in GIMPS?

[M29]

Originally posted by GP2
Another question we'd need to ask is, would a screensaver version of the program lead to better retention of new users... or maybe even a higher rate of users trying it in the first place?

Why does Prime95 itself have to be the screen saver?

An independent program could be written that queries Prime95.ini, for example. It could also query PrimeNet.

In fact, there could be any number of entirely different screen savers written by different people.

There may be requests of George to log additional information, but that's about the worst I see.