Message boards :
Number crunching :
Optimize your GPU. Find the value the easy way.
Message board moderation
Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · Next
Author | Message |
---|---|
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
Compile the entire program into a single executable file... I completely disagree (unless you mean SFX ZIP/RAR/7z) - this will make hard/impossible to add/change the apps/WUs But maybe more descriptive text or error code (if Fred implements it) may help, e.g.: ERROR: Unable to copy ...... Aborted (error code: XX returned by FFF function - <descriptive text>) 1) Do you have some read-only files in slot0 C:\Users\Jim\AppData\Roaming\eFMer\SetiPerformance\slot0\ Delete all files from slot0 (or Delete all slot0 slot1 ... folders, they will be recreated) 2) Did you try what Fred suggested? - Copy manually ALL files from: C:\videocardtest\SetiPerformance_32_64_1_4\zip\x41g\ ... to: C:\Users\Jim\AppData\Roaming\eFMer\SetiPerformance\slot0\ (to see if any error pops-up) 3) Alex Storey: "are you trying to run the tool straight from the zipped folder?" Meaning "did you extract the files to a real folder" or instead "did you just double-click SetiPerformance_32_64_1_4.zip and run the program/exe" (I remember in Vista this (the Vista's internal zip handling) was implemented very stupid, don't know how it is in Windows 7) (I think you use real folder else (I think) the message will say "Unable to copy from: C:\.....\TEMP\.....") But make it easy or im just not interested...ya know....like almost everyone else. "almost everyone else" easy succeeded to run this program. I recall only one else reporting this 'copy' problem: http://setiathome.berkeley.edu/forum_thread.php?id=69136&postid=1281569#1281569 ERROR: Unable to copy from: D:\My Documents\Downloads\BOINC\BOINC Tasks\x41g\ to: C:\Users\Cliff Harding\AppData\Roaming\eFMer\SetiPerformance, slot: 0 Aborted Probably because of spaces in both paths (which this program for some reason don't like) Notice how nearly 100% of this discussion involves Volunteer Testers 'Volunteer tester' means (I think) nothing more than they run tasks from SETI@home/AstroPulse Beta: http://setiweb.ssl.berkeley.edu/beta/ Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
ERROR: Unable to copy from: D:\My Documents\Downloads\BOINC\BOINC Tasks\x41g\ to: C:\Users\Cliff Harding\AppData\Roaming\eFMer\SetiPerformance, slot: 0 (was too late to edit again so I quote myself) ... but my target folder is this: D:\Documents and Settings\A & L\Application Data\eFMer\SetiPerformance\slot0\ ... and "this program" have no problem with 5 spaces (nor with "&") in: "D:\Documents and Settings\A & L\Application Data\" Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Snowmain Send message Joined: 17 Nov 05 Posts: 75 Credit: 30,681,449 RAC: 83 |
I completely disagree (unless you mean SFX ZIP/RAR/7z) - this will make hard/impossible to add/change the apps/WUs Are you serious?! He has the source code! The idea that it would limit future upgrades is ridiculous..... I don't see anyone other than Fred making any changes at all to the software? So if he continues to be the only one making changes, again he has the source code and could upgrade as he saw fit, than simply compile a new executable. I am simple suggesting that the final file to the "public" be a single executable. Making this directory of files into one file in no way limits the future possibility of upgrades. KISS |
Horacio Send message Joined: 14 Jan 00 Posts: 536 Credit: 75,967,266 RAC: 0 |
I completely disagree (unless you mean SFX ZIP/RAR/7z) - this will make hard/impossible to add/change the apps/WUs Its not the future upgrades which will be limited, its the option for the users to use different WUs for the test and different versions of the SETI apps (that, by the way, are not made by Fred). If all the required files/data is inside one "compiled" file, then Fred will need to compile a different version of the tool for each new version of the apps... And, worst, it will make the tool useless for other purposes like testing a failed WU in offline mode or to test an unreleased/Beta version of a new app... I dont know why you are having this issue, and IIRC you used this tool before, so I guess there is something different in one of your hosts, if you really want this tool to become more easy to use, the best thing you can do is to try to help Fred to find what makes it fail so he can fix/change whatever is needed to avoid this in the future. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13751 Credit: 208,696,464 RAC: 304 |
Making this directory of files into one file in no way limits the future possibility of upgrades. That's what archiving it does. Grant Darwin NT |
S@NL - eFMer - efmer.com/boinc Send message Joined: 7 Jun 99 Posts: 512 Credit: 148,746,305 RAC: 0 |
The problem is in the inconsistencies. \n is sometimes a new line \r\n is the Windows way. But writing it to file gets \r\r\n the \n -> \r\n I now remove all \r before writing, produces \r\n. TThrottle Control your temperatures. BoincTasks The best way to view BOINC. Anza Borrego Desert hiking. |
S@NL - eFMer - efmer.com/boinc Send message Joined: 7 Jun 99 Posts: 512 Credit: 148,746,305 RAC: 0 |
V 1.5 http://www.efmer.eu/forum_tt/index.php?topic=974.msg5946#msg5946 TThrottle Control your temperatures. BoincTasks The best way to view BOINC. Anza Borrego Desert hiking. |
MarkJ Send message Joined: 17 Feb 08 Posts: 1139 Credit: 80,854,192 RAC: 5 |
I ran the automatic test, which runs x41g thats included in the archive. In real life I use x41z cuda42 app. The cards are also running Einstein apps which would appear to be running through in 40 min (average) compared with 46 min (average) from the 560Ti that they replaced. Their app is a cuda32 one. So its not fantastically faster but then it uses 30 watts less and is still a bit faster than the 560Ti. BOINC blog |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13751 Credit: 208,696,464 RAC: 304 |
So its not fantastically faster but then it uses 30 watts less and is still a bit faster than the 560Ti. And notice it's faster than a 560Ti. The Ti's always perform better than the non Ti's, so a 660Ti would be (somewhat) faster still. Grant Darwin NT |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
V 1.5 http://www.efmer.eu/forum_tt/index.php?topic=974.msg5946#msg5946 From V 1.5 changelog: In the logging file line ends were \r\r\n now \r\n. - Yes, result.log fixed (and is now in Unicode) Added: Max count drop down box. - Not shown by default, to enable: <showMaxCount>1</showMaxCount> Fixed: Progress indicator dropped back to 0. - Partially fixed (for AP): it shows and do not drop back to 0 but now stays at 11% till task end Not fixed: Count actually run is one more than the selected by the drop down box: (the same problem happened (I reported it before) with previous versions with <maxAutoCount>3</maxAutoCount> Now I replaced this with <showMaxCount>1</showMaxCount> and selected 0.33 from the GUI) Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
S@NL - eFMer - efmer.com/boinc Send message Joined: 7 Jun 99 Posts: 512 Credit: 148,746,305 RAC: 0 |
V 1.5 http://www.efmer.eu/forum_tt/index.php?topic=974.msg5946#msg5946 1) If you can send me the file every time it's update I can check what's going on. All files that appear while processing. 2) OK. TThrottle Control your temperatures. BoincTasks The best way to view BOINC. Anza Borrego Desert hiking. |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
V 1.5 http://www.efmer.eu/forum_tt/index.php?topic=974.msg5946#msg5946 If you get 'my' last package: http://setiathome.berkeley.edu/forum_thread.php?id=69136&postid=1287852#1287852 ... it contains ready-to-use NVIDIA AP (if you don't want to spend time to prepare NVIDIA AP yourself) This will allow you to test in real time. The files to monitor are only 3: pulse.out0 pulse.out1 pulse.out They update in some 'unknown' order. Get 3x3 samples of them here: http://www.datafilehost.com/download-c89d93d1.html  - ALF - "Find out what you don't do well ..... then don't do it!" :)  |
Snowmain Send message Joined: 17 Nov 05 Posts: 75 Credit: 30,681,449 RAC: 83 |
With the lack of Nvidia work units I put some more time into this. EVGA GTX 570 Classified Overclocked. A standard GTX 570 = 732MHz Processor Stock 823MHz Processor 976Mhz Memory Current 839MHz Processor 1001MHz Memory Starting automatic test: (x41g) 03 October 2012 - 21:05:40 Start, devices: 1, device count: 1 (1.00) Device: 0, device count: 1, average time / count: 115, average time on device: 115 Seconds (1 Minutes, 55 Seconds) 03 October 2012 - 21:07:36 Start, devices: 1, device count: 2 (0.50) Device: 0, device count: 2, average time / count: 194, average time on device: 97 Seconds (1 Minutes, 37 Seconds) 03 October 2012 - 21:10:53 Start, devices: 1, device count: 3 (0.33) Device: 0, device count: 3, average time / count: 287, average time on device: 95 Seconds (1 Minutes, 35 Seconds) >> The best average time found: 97 Seconds (1 Minutes, 37 Seconds), with count: 0.50 (2) 03 October 2012 - 06:53:55 Start, devices: 1, device count: 1 (1.00) (Wu: FG00134_V7.wu) Device: 0, device count: 1, average time / count: 2975, average time on device: 2975 Seconds (49 Minutes, 35 Seconds) 03 October 2012 - 07:43:31 Start, devices: 1, device count: 2 (0.50) (Wu: FG00134_V7.wu) Device: 0, device count: 2, average time / count: 5649, average time on device: 2824 Seconds (47 Minutes, 4 Seconds) 03 October 2012 - 09:17:44 Start, devices: 1, device count: 3 (0.33) (Wu: FG00134_V7.wu) Device: 0, device count: 3, average time / count: 8444, average time on device: 2814 Seconds (46 Minutes, 54 Seconds) 03 October 2012 - 11:38:40 Start, devices: 1, device count: 4 (0.25) (Wu: FG00134_V7.wu) Device: 0, device count: 4, average time / count: 11089, average time on device: 2772 Seconds (46 Minutes, 12 Seconds) --------------------------------------------------------------------------- Results: (FG00134_V7.wu) |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
It will be interesting to know how did you overcome the "ERROR: Unable to copy ......" Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Snowmain Send message Joined: 17 Nov 05 Posts: 75 Credit: 30,681,449 RAC: 83 |
Directory permissions, I am not the admin on my own machine. Atleast I think thats what it was. |
S@NL - eFMer - efmer.com/boinc Send message Joined: 7 Jun 99 Posts: 512 Credit: 148,746,305 RAC: 0 |
V 1.6 with some minor fixes. http://www.efmer.eu/forum_tt/index.php?topic=974.0 According to the developer, the AP application can't give a progress.... But I made some changes anyway. TThrottle Control your temperatures. BoincTasks The best way to view BOINC. Anza Borrego Desert hiking. |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
New combined package (21 MB): All from original SetiPerformance_32_64_1_6.zip + 5 OpenCL tests: ATI (MB and AP) and NVIDIA AP + 2 my 'apps' to test SetiPerformance decisions for "The best average time found" (Edited config.xml accordingly) SetiPerformance_32_64_1_6__OpenCL_ATI_NV_Added-by-BilBg.7z Direct link: http://www.datafilehost.com/get.php?file=be9d6904 ! Edit: First click on the link will open a page in a new tab - just close the tab and use again the above 'Direct link' from here (I give 'Direct link' and not the usual: http://www.datafilehost.com/download-be9d6904.html because yesterday I found that datafilehost.com started to use/offer some stupid/adware "download manager" (NOD32 pop-up orange warning about PUP for this "download manager") Again yesterday Unchecking "Use our download manager and get recommended downloads" in Firefox did not have any effect (meaning the "Download: Click Here" lead to the same long link (http://www.storagenl.info/.....) despite the check box) Now in Opera the check box have effect (the "Download: Click Here" changes to the above 'Direct link' (which I in fact extracted yesterday manually from the long link)) For me the 'Direct link' works Directly ;) If it does not work for you: - first open the usual link (in case it sets some temporary cookie or other means to avoid direct linking): http://www.datafilehost.com/download-be9d6904.html - then use the 'Direct link' above ) OpenCL tests: MB6_win_x86_SSE3_OpenCL_ATi_HD5_r390 MB6_win_x86_SSE3_OpenCL_ATi_r390 (for ATI HD4xxx or to compare the speed difference to HD5_r390 app on ATI HD5xxx+ GPU) (New) Beta_MB7_6.99_OpenCL_ATi_r1643 (== setiathome_6.99_windows_intelx86__opencl_ati_sah.exe) (appears to be slower than r390 probably because of stock defaults aimed at low/no lag) AP6_win_x86_SSE2_OpenCL_ATI_r1316 AP6_win_x86_SSE2_OpenCL_NV_r1316 (Not tested - I don't have NVIDIA) (New) My 'apps': (written for fun using AutoHotkey (this is my first attempt to use AutoHotkey scripts, source .ahk + compiled .exe + Compiler Ahk2Exe included)): SetiPerformance Test4Test (Simple, Exit Manually) SetiPerformance Test4Test (Progress, Random times) (as usual with programs it appears that the 'Simple' is better, the other 'app' (Progress) is just more 'fancy' ;) ) === === === Note: I consider all these SETI@home apps 'Freely available' because: - Anyone that is attached to SETI@home (and uses stock apps, and have capable GPU) will get/download automatically from the SETI@home servers the same AstroPulse OpenCL ATI/NVIDIA apps as included here (compare the stock .exe files with the optimized if you don't believe me) - Anyone can attach to SETI@home Beta and get (the same way: from SETI@home servers) the Beta apps: setiathome_6.99_windows_intelx86__opencl_ati_sah.exe (== MB OpenCL ATi r1643) (which is a newer version of MB6_win_x86_SSE3_OpenCL_ATi_r390) x41z (not included in 'my' combined package) (the .exe properties say: "Copyright (c) 2012, Jason Richard Groothuis bSc, All Rights Reserved", Original File name: Lunatics_x41z_win32_cuda42.exe) - Some of the apps are still available on the Lunatics site, e.g.: AP 6.01 r555 ATI OpenCL Cuda Multibeam x41g >> Evidence: http://lunatics.kwsn.net/index.php?module=Downloads;catd=39 http://setiathome.berkeley.edu/apps.php http://setiathome.berkeley.edu/forum_thread.php?id=68948 http://setiweb.ssl.berkeley.edu/beta/apps.php http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=1964 http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=1956 === === === So using my 'apps' I tested SetiPerformance - the decisions for "The best average time found" are really weird (I'm confident there is a bug in that code) Better to do the decision by yourself (as you probably already do) looking the values, feeling the lag, judging the used RAM, GPU load, temperature, ... "The best average time found" test using my 'apps' which permit to simulate different run times: Starting automatic test: (SetiPerformance Test4Test (Simple, Exit Manually)) Device: 0, device count: 1, average time / count: 59, average time on device: 59 Seconds (0 Minutes, 59 Seconds) Device: 0, device count: 2, average time / count: 50, average time on device: 25 Seconds (0 Minutes, 25 Seconds) Device: 0, device count: 3, average time / count: 104, average time on device: 34 Seconds (0 Minutes, 34 Seconds) Device: 0, device count: 4, average time / count: 39, average time on device: 9 Seconds (0 Minutes, 9 Seconds) >> The best average time found: 34 Seconds (0 Minutes, 34 Seconds), with count: 0.33 (3) Starting automatic test: (SetiPerformance Test4Test (Simple, Exit Manually)) Device: 0, device count: 1, average time / count: 16, average time on device: 16 Seconds (0 Minutes, 16 Seconds) Device: 0, device count: 2, average time / count: 35, average time on device: 17 Seconds (0 Minutes, 17 Seconds) Device: 0, device count: 3, average time / count: 36, average time on device: 12 Seconds (0 Minutes, 12 Seconds) Device: 0, device count: 4, average time / count: 30, average time on device: 7 Seconds (0 Minutes, 7 Seconds) >> The best average time found: 12 Seconds (0 Minutes, 12 Seconds), with count: 0.33 (3) Starting automatic test: (SetiPerformance Test4Test (Progress, Random times)) Device: 0, device count: 1, average time / count: 83, average time on device: 83 Seconds (1 Minutes, 23 Seconds) Device: 0, device count: 2, average time / count: 50, average time on device: 25 Seconds (0 Minutes, 25 Seconds) Device: 0, device count: 3, average time / count: 75, average time on device: 25 Seconds (0 Minutes, 25 Seconds) Device: 0, device count: 4, average time / count: 41, average time on device: 10 Seconds (0 Minutes, 10 Seconds) >> The best average time found: 25 Seconds (0 Minutes, 25 Seconds), with count: 0.33 (3) Starting automatic test: (SetiPerformance Test4Test (Progress, Random times)) Device: 0, device count: 1, average time / count: 46, average time on device: 46 Seconds (0 Minutes, 46 Seconds) Device: 0, device count: 2, average time / count: 84, average time on device: 42 Seconds (0 Minutes, 42 Seconds) Device: 0, device count: 3, average time / count: 84, average time on device: 28 Seconds (0 Minutes, 28 Seconds) Device: 0, device count: 4, average time / count: 85, average time on device: 21 Seconds (0 Minutes, 21 Seconds) >> The best average time found: 28 Seconds (0 Minutes, 28 Seconds), with count: 0.33 (3) Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Nick Send message Joined: 17 May 99 Posts: 96 Credit: 17,356,094 RAC: 0 |
BilBg Thanks much for this handy program. Very informative! Nick |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
Oh, No - the programs (SetiPerformance and apps) are not mine, I only collected them, edited config.xml, tested if all works on my system and uploaded the 'combined package'. (only the 2 small 'apps' (AutoHotkey scripts 'SetiPerformance Test4Test') are mine) The author of SetiPerformance is Fred (efmer), the link is in this his post: http://setiathome.berkeley.edu/forum_thread.php?id=69136&postid=1294078#1294078 If you didn't already - check also the programs/links in his signature (TThrottle, BoincTasks) Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
SupeRNovA Send message Joined: 25 Oct 04 Posts: 131 Credit: 12,741,814 RAC: 0 |
Hello Fred, thank you very much for the program i use it from the first version you release. I want to ask you which driver you use to complete 130sec for WU for GTX 295 and which WU from the list because my top time with 3 drivers tested same clock everything the same the best time for Core755/Shaders1655/Mem1255 is 188-190 sec with x41g for wu. and my max settings are 805/1733/1255 the time for PG0444_v7 unit need 179 sec. |
©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.