Message boards :
Number crunching :
Average Credit: 1,498 (12/31/2005) on 1 CPU!
Message board moderation
Author | Message |
---|---|
JRL Send message Joined: 6 Dec 01 Posts: 23 Credit: 4,206,402 RAC: 0 |
Ok, If one goes to the BOINCSTATS leaderboards located at this address: http://www.boincstats.com/stats/host_cpu_stats.php?pr=bo&st=0&to=100&or=10 I see there are a few CPU consistently generating over 1,000 average credits and, as one would expect, they are top of the line AMD CPU's. But, are they really that fast? Favor to ask of the owner of these monsters . . . tell us how you do it. Are the over-clocked, optimized, water-cooled, blessed by CPU Gods? Seriously, I want one of these machines! could you give us a hint on the configuration? Many thanks. John Lee John R. Lee, Jr. Akuratus Corporation, CEO |
Sir Ulli Send message Joined: 21 Oct 99 Posts: 2246 Credit: 6,136,250 RAC: 0 |
Ok, |
Hans Dorn Send message Joined: 3 Apr 99 Posts: 2262 Credit: 26,448,570 RAC: 0 |
Hi John, I don't know how the boincstats numbers were calculated. For a seti related statistic have a look over here The best performers (per CPU) are dual core P4s ATM Regards Hans |
JRL Send message Joined: 6 Dec 01 Posts: 23 Credit: 4,206,402 RAC: 0 |
Thanks for the info and the info on your machine. Have you optimized it or made any other modifications? John Ok, John R. Lee, Jr. Akuratus Corporation, CEO |
Sir Ulli Send message Joined: 21 Oct 99 Posts: 2246 Credit: 6,136,250 RAC: 0 |
Hi John, Hi Hans but i think that the good old Silicon Graphics, Inc.IP35 are not so bad i think http://setiathome.berkeley.edu/show_host_detail.php?hostid=1934524 the new DualCore, i say nothink to this, they are fast Greetings from Germany NRW Ulli |
Sir Ulli Send message Joined: 21 Oct 99 Posts: 2246 Credit: 6,136,250 RAC: 0 |
|
KB7RZF Send message Joined: 15 Aug 99 Posts: 9549 Credit: 3,308,926 RAC: 2 |
Sorry that this is off topic. @ Sir Ulli, which part of Germany are you from? My mom was born and raised in Marburg/Lahn, where my oma still lives. Just curious. Thanks. Sorry to go off topic all |
snorrre Send message Joined: 7 Feb 00 Posts: 46 Credit: 16,393 RAC: 0 |
for Info the second host ia a 32 CPU Host.................... Wow... Now I know what I really wanted for christmas. Too bad it's to late now. I better write it down on my wish-list for next year...: Computer with 32 CPUs and 32 GB RAM... That computer alone has almost 1500 workunits in que in its cache. But why is it claiming only about 5 credits per WU? Happy new year! Greetings from Norway. Crunching is fun! |
Landroval Send message Joined: 7 Oct 01 Posts: 188 Credit: 2,098,881 RAC: 1 |
But why is it claiming only about 5 credits per WU? Because claimed credit is based on benchmarked CPU speed * crunch time. A machine that crunches a unit very quickly claims less credit. Also, there are factors such as memory speed and L2 cache that affect crunch times that the benchmark doesn't measure. Finally, Linux machines will routinely have lower benchmark scores than Windows machines, even on the same hardware. The enhanced science application, due for release Real Soon Now (tm), will use an actual operations count to determine credit, rather than estimating based on a flawed benchmark. Happy crunching, Brian If you think education is expensive, try ignorance. |
Toby Send message Joined: 26 Oct 00 Posts: 1005 Credit: 6,366,949 RAC: 0 |
This host used to crunch for me and had a RAC of over 1,500. I was using the latest optimized client at the time that did a work unit in 40-45 minutes and it has 4 CPUs so... thats a whole lotta crunching :) At one point that host was the #8 host on all of seti. Unfortunately its burn-in time is over and it is now doing "real" work as a database server :) A member of The Knights Who Say NI! For rankings, history graphs and more, check out: My BOINC stats site |
Sleestak Send message Joined: 22 Jun 01 Posts: 779 Credit: 857,664 RAC: 0 |
|
Clyde C. Phillips, III Send message Joined: 2 Aug 00 Posts: 1851 Credit: 5,955,047 RAC: 0 |
It is better to use number of workunits per hour or day instead of recent average credit. The awarding of credit (cobblestones per workunit) is highly variable and not a reliable indicator. Some machines claim 50 cobblestones per workunit and others five. The time required to calculate a workunit varies but most stay within +- 10 percent of the average. Granted credit is more consistent than claimed credit due to the use of just the median of the claims of the (usual) first three per workunit. But I still think it's better to go by workunits per time unit. Later, with SetiEnhanced, cobblestones per time unit will probably be much more consistent than workunits per time unit because of the extreme variance of calculation times per workunit with Enhanced, so I've read. The way I take it, using the crunchers at the Guntec/Crunch3r website produces the fastest results. |
-= Vyper =- Send message Joined: 5 Sep 99 Posts: 1652 Credit: 1,065,191,981 RAC: 2,537 |
Hi My X2 4400+ is averaging about 1650 credits with Crunch3r's SSE3 optimized Seti client. So that's possible.. Kind Regards Vyper http://setiathome.berkeley.edu/show_host_detail.php?hostid=1358316 _________________________________________________________________________ Addicted to SETI crunching! Founder of GPU Users Group |
Lee Carre Send message Joined: 21 Apr 00 Posts: 1459 Credit: 58,485 RAC: 0 |
I've seen a host with an RAC of about 4,000,000, but it was all "legal" because i'm guessing the user merged multiple hosts into one or something, i can't remember the host, or the thread it was discussed in, sorry also this is possible if a host has been returning lots of work, and the validators have been off for a while (for example), when they start validating again, the host will be granted a lot of credit in a short space of time, thus one possible cause of a "spike" of high RAC (had this myself once or twice) |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.