Message boards :
Number crunching :
How Are Benchmarks Calculated?
Message board moderation
Previous · 1 · 2 · 3 · Next
Author | Message |
---|---|
Misfit Send message Joined: 21 Jun 01 Posts: 21804 Credit: 2,815,091 RAC: 0 |
> Thanks Rom. > > To all Celeron users !!! In order to complete your work in time, be sure to > run the benchmarks manually with your box fully loaded. Launch a defrag or > play doom while you BOINC benchmarking... > > Friendly > Marc > Im not sure how that would work since benchmarking suspends all other CPU activity. I cant even open up a new window let alone start an application. |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Hi all, Thanks Rom for the input on the Benchmarks. I appreciate all you do. >BTW, MLCudd.... Aim High! Go USAF! ;) (you know a zoomie's gotta poke a little fun at a squid) :) Hey Thunder, I amied higher than the Air Force could Fly. We flew F-14's, The old obselete F-4 Phantoms, and I was lucky enough to be part of the testing on the YF-17, Which became the Navy's Best Fighter, The F-18 Hornet. The Air Force only wanted me to work on the Engines. :-( no fun in that!! And Thanks everybody who had input on this thread. Have A Great Day. Warm Regards To All, Rocky Cudd |
Razorirr Send message Joined: 3 Apr 99 Posts: 92 Credit: 7,414 RAC: 0 |
want to know somthing weird i actually was able to get my to completion times to get into the 1 day mark. my computer usually says they will take 8-9 hrs but i ran a bench mark while running a bunch of computer sterss test at the same time( hot cpu tester lite edition, the tourture test that comes on gimps, etc.) that was bad. another thing i dont run any thing on this computer except boinc unless im on these forums. the last wu seti ran took 22 hrs when it said it would take 8.5. the scary thing is if its taking almost 3 times longer than it should how long will my cpdn one take its competion time is 2352:33:16. if it ran the same rate thats 7000 to 7100 hrs. :> |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Hi Roland, >>how long will my cpdn one take its competion time is 2352:33:16. if it ran the same rate thats 7000 to 7100 hrs. :> A short and Sweet Answer A Long Long Time!! :-) :-) :-) Happy crunching!! Regards, Rocky |
Razorirr Send message Joined: 3 Apr 99 Posts: 92 Credit: 7,414 RAC: 0 |
> Hi Roland, > A short and Sweet Answer > > A Long Long Time!! :-) :-) :-) > > > Happy crunching!! > > Regards, > > Rocky > <img> src="http://boinc.mundayweb.com/seti2/stats.php?userID=948&useCached=true"> > hey theres a good thing to it il go from worst to first in one wu except for the fact that as i was writing this it said this climateprediction.net - 2004-09-10 00:07:33 - Unrecoverable error for result 1y22_100111719_0 (CreateProcess() failed - The paging file is too small for this operation to complete. (0x5af)) climateprediction.net - 2004-09-10 00:07:33 - CreateProcess() failed - The paging file is too small for this operation to complete. (0x5af) im going to their message boards right now to post this if it already hasent happened |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Roland, >>Unrecoverable error for result 1y22_100111719_0 (CreateProcess() failed - The paging file is too small for this operation to complete. (0x5af)) That is the exact reason why I detached from CPDN until all the bugs are worked out of Boinc. From what I got from the other message posts, It has something to do with I believe Server tuning or timing. When I find the Threads, I will forward them Have A Better Tomorrow, Rocky Cudd |
Rachel Send message Joined: 13 Apr 02 Posts: 978 Credit: 449,704 RAC: 0 |
Hello my pc is a Dell 2.8ghz with pentuim 4 processors with ht and 520 ram with 80gb hard drive and I am doing two seti@home boinc wu's in 3 hours and 20 mins, so around one hour and 40 mins per unit. |
Razorirr Send message Joined: 3 Apr 99 Posts: 92 Credit: 7,414 RAC: 0 |
> Roland, > > > >>Unrecoverable error for result 1y22_100111719_0 (CreateProcess() > failed - The paging file is too small for this operation to complete. > (0x5af)) > > > That is the exact reason why I detached from CPDN until all the bugs are > worked out of Boinc. From what I got from the other message posts, It has > something to do with I believe Server tuning or timing. When I find the > Threads, I will forward them > > > Have A Better Tomorrow, > > Rocky Cudd > <img> src="http://boinc.mundayweb.com/seti2/stats.php?userID=948&useCached=true"> > i just did that i will try them again next year. i startedd with them on the 23 of august and havent ever got to a trickle but ive had five wus 1. boinc crashed and lost it 2. tryed to go to 4.08 and crashed it 3. same as two 4. paging error 5. the one it downloaded before i detached |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Rach, You have a nice fast box. I will gladly trade you 2 of mine for it. Are you on Dial-up? Being new, I would investigate the links that you were sent previously. They will be of tremendous help. Paul Buck is the master of documentation. Like I said just holler, someone will be there to help. Regards, Rocky |
Rachel Send message Joined: 13 Apr 02 Posts: 978 Credit: 449,704 RAC: 0 |
> Rach, > You have a nice fast box. I will gladly trade you 2 of mine for it. > > Are you on Dial-up? > Hello Got my pc last week.Got broadband last week.Before that I had an 8 year old pc which was 333mhz and only 2.9gb hard drive and 250 ram.It took 24 hours to do one seti@home classic screensaver ,lol.It would crash 3 times a day.I bought my new pc as the other one was unusable.It's nice having something that actually works.My dial up I had was so slow and it was only £3 a month more to have 10x broadband.Well worth an extra £3 a month I say. |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Hey Roland, I don't think it will be necessary to wait until next year to restart, I have a feeling that all the bugs will be resolved in the next 30 days. By the way, the new program LHC opens to the public around the 29th of Sept. It is on the Boinc Platform. I will look for you there. Regards, Rocky |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
Rocky, > Being new, I would investigate the links that you were sent previously. They > will be of tremendous help. Paul Buck is the master of documentation. Come on, you are making me blush! Thank you though... If I just help one, that is the reward. Rachel, not to scare you off, but I do have about 230 pages of stuff you can read. Lots of examples. And my friends John and John keep me honest and right (well, I think they are my friends, I am not sure if the feeling is mutual, I never asked!) :) <p> Click Me! |
mlcudd Send message Joined: 11 Apr 03 Posts: 782 Credit: 63,647 RAC: 0 |
Paul, Your Documentation has been a tremendous help for me, and I AM SURE OTHERS!! Thank You. Warm Regards, Rocky Cudd |
Razorirr Send message Joined: 3 Apr 99 Posts: 92 Credit: 7,414 RAC: 0 |
> Hey Roland, > I don't think it will be necessary to wait until next year to restart, I > have a feeling that all the bugs will be resolved in the next 30 days. > > By the way, the new program LHC opens to the public around the 29th of Sept. > It is on the Boinc Platform. I will look for you there. > > Regards, > > Rocky > yah i found out about LHC the day after they were open but they all ready filled up. ill have to wait for the final front ear to start a team there though. |
dazphotog Send message Joined: 13 Mar 02 Posts: 73 Credit: 99,224 RAC: 0 |
I have a 2.0 Ghz Celeron with 256meg of Ram...and the benchmarks say a workunit should take 6.5 hours and most units take 6.5 to 6.75 hours. Don |
Christopher Hauber Send message Joined: 10 Feb 01 Posts: 196 Credit: 71,611 RAC: 0 |
Hmm.. I knew the Celeron was the "value" line but I didn't realize just how inferior to the Pentiums they really were. My P3 1.0 GHz with 512 MB RAM only takes about an hour longer than your 1.0 GHz Celeron. I'm surprised there isn't a bigger difference there. As an added comparison my "2800+" (2400+ overclocked) processes work in about 2:50 on the low end of the 2800+ range (I say "range" because I have lowered the speed a couple times recently without it changing the 2800+ designation). Chris > I have a 2.0 Ghz Celeron with 256meg of Ram...and the benchmarks say a > workunit should take 6.5 hours and most units take 6.5 to 6.75 hours. > > Don > > |
Razorirr Send message Joined: 3 Apr 99 Posts: 92 Credit: 7,414 RAC: 0 |
i got the same processor and ram as he did but my computer must be slow they used to say 8 hours. they wuold take 22 though. now after increasing my virtual paging file from 200 mgs to 3 gigs of memory i dropped to 6.5 hrs but i havent goten to run one from start to finish. when i get enough cash im going to take the inspiron 1100 notebook in to get a bigger ram chip and swap out the celeron with a mobile p4 with hyperthreading tech. that should make it faster. |
Celeron300A@450 Send message Joined: 16 May 99 Posts: 2 Credit: 318,306 RAC: 0 |
> about 3.8 hours, and the WU's recieved fit that box. However my Intel Celeron > 2.8 box, says that I should be crunching WU's in 4.22 hours. Problem is, it > takes exactly double that time. When I ran a new set of Benchmarks yesterday, > the WU's I just got, are for the exact same amount of time 4.22 hours. I did well, my celeron 2.4ghz is crunching wu's in under 4 hours, when boinc says it should take more than 8 hours. im not sure if these are factors, but ill list them for analysis : 1) using single stick of cheap unbranded ram. 512mb shared with on board agp chip 2) mobo chipset is single channel via 3) swap file is set to 1024mb 4) usage load is basically for surfing, email, streaming radio. i do have concerns that i may run out of work. not so much that i wont finish work in time. |
Benher Send message Joined: 25 Jul 99 Posts: 517 Credit: 465,152 RAC: 0 |
Hi all, I'm the Ben Rom was referring to. As for how benchmarks are calculated... Whetstone is the name of the benchmark that is reported on your [Show computers] web page as "Measured floating point speed" Dhrystone is the name of the benchmark used for "Measured integer speed". Float = 1.48283 or 3.141592, Integer = 1 or 2 or 938283 or 2004. Floats have a decimal point, integers are also called whole numbers. Whetstone Whetstone does 8 different groups of tests (repeatedly of course), times how long they took to finish, and pops out a number. [ops performed]/[time]. These tests all use floating point math operations of the CPUs being tested. Some of them are simple math plus, minus, times, divide and others check trigonometry feaures sine, cosine, tangent, exponent. Dhrystone Dhrystone checks repeated integer operations, and several operating system file handling operations. Neither of the tests really checks how well/fast a system can access memory, and seti (for example) acesses memory a lot. Here is an example of memory introducing a delay. A pentium 4 CPU of any speed, can calculate the sine of an angle in approximately 170 tics of its internal clock. It could have performed 170 regular integer additions in this time. But if it wanted to do an integer addition on a number somewhere out in memory (say it was working on a table of numbers), the CPU might have to wait for this memory integer to be delivered to the CPU as much as 260 tics. So a badly timed integer+memory operation would take far longer than a sine calculation. This is where Celeron CPUs can really slow down. Pentium has many more features to predict when the CPU might be getting memory, and begins getting it long before the CPU actually calculates with it. Thus much less delay for most memory operations = faster. Predicted time vs actual Each WU delivered to your machine includes an estimated number of floating point (FP) calculations. BOINC divides this by the FP benchmark number to estimate completion time. Seti's WUs estimate number is currently allways 27.9 trillion (american), however the actuall number of FP ops varies greatly which is why they don't all take the same ammount of time. A better estimate might be calculated on the servers, but it would probably take so long looking through the WU data that much fewer WUs would be generated and host machines wouldn't get WUs. Seti vs Benchmarks Seti uses almost all "single" precision floating point math, whetstone is all "double" precision math. On Intel x86 processors the speed difference in calculating single vs. double isn't very large. Seti uses mostly add, sub, multiply and divide. About 20% of its time is spent in trigonometry. Almost all the time in Whetstone is used for trigonometry. Memory access speed and trigonometry are the two major reasons that the benchmark results and seti processing speed don't match up on many systems. |
Ned Slider Send message Joined: 12 Oct 01 Posts: 668 Credit: 4,375,315 RAC: 0 |
Ben, Nice explanation! I've been thinking of adding a small faq to my site relating to common questions concerning some of the more technical aspects of the credit system as it seems to relate to my optimised boinc clients (benchmark scores). I've noticed you've posted some nice explanations such as the one above together with a previous explanation of the formula to calculate claimed credit that also caught my eye. I was wondering if you'd mind if I lifted/borrowed your excellent explanations and used them on my site, acknowledging yourself, of course, as the source? This is the type of info you don't come across to often and I feel having it available on my site may be helpful to others. Thanks, and I await your thoughts, Ned *** My Guide to Compiling Optimised BOINC and SETI Clients *** *** Download Optimised BOINC and SETI Clients for Linux Here *** |
©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.