A couple of largish numbers

[debrouxl]

In fact, someone already hinted squalyl and I at raising the amount of credits granted to users, but I didn't do that largely for the reason mentioned by xilman...
It's true, however, that most users consider that stats are important.

[Syd]

Originally Posted by xilman

Sounds like an invitation to an inflationary arms race to me...

Paul

Maybe less than average was not the right term - its less than intended by boinc. If I do the benchmarks and do the math "cobblestones" on my 2 systems I get ~33 credits/h intended, ~26/h granted by RSALS and the other one ~12/h, granted ~9/h. Not much - but maybe 10% more credits and its still below the the intended credits.

[em99010pepe]

Originally Posted by debrouxl

In fact, someone already hinted squalyl and I at raising the amount of credits granted to users, but I didn't do that largely for the reason mentioned by xilman...
It's true, however, that most users consider that stats are important.

That was me.

BOINC is a computer distributed platform where with a few mouse clicks ( next, next next ) you start running a project without understanding the goal and the concepts of the project. BOINC is for ignorants. 90 % of BOINC users run for the stats, where they can return more points per GPU and CPU crunching, even if they are running a GPU that it is less energy efficient than a CPU doing the same work. They want points, team challenges, etc..In my opinion you should give to them what they want, more points for team and members rankings. You will see by then that activity will increase in RSALS...

Anyway, who will be post-processing C162_3678_3076?

Carlos

[Syd]

Originally Posted by em99010pepe

Anyway, who will be post-processing C162_3678_3076?

I´d like to do it. With 6 Cores/8GB memory it will finish in a few days.

[em99010pepe]

Originally Posted by Syd

I´d like to do it. With 6 Cores/8GB memory it will finish in a few days.

It will be faster than running on my current free Q6600.

[Syd]

Postprocessing is done, factors are:

prp72 factor: 212470575606466999863125410295739646908983528854995689565759949064649919
prp90 factor: 504129124864697847512307773754975771053180340277666703354781986620877305015457915076649889

I'll try to push this sequence forward for a few more days

[debrouxl]

Thanks for the post-processing

[fivemack]

3270.632 is now sieved (ten days on 48 K10/1900 cores); I'll do the linear algebra and carry on with that sequence. It's still vaguely interesting as the shortest <10^4.

[fivemack]

BLanczosTime: 211517 (48 cores)
sqrtTime: 16152
prp80 factor: 13244901306481804073783791555906189988726341073744338510945049063254989773644747