Message boards :
Number crunching :
4th: Compiling faster Windows client with Intel C++ compiler (and fftw?)
Message board moderation
Author | Message |
---|---|
![]() Send message Joined: 6 Jul 99 Posts: 1600 Credit: 391,546 RAC: 0 ![]() |
A little summary (cut & paste from the other threads): <blockquote>1. P3 seti-p3.zip (for AMD XP and 64 use this one!) 2. P4 w/o sse3 seti-p4-no-sse3.zip 3. P4 w sse3 seti-p4-sse3.zip 4. And reference work unit and result unit are in a zip referenceunits.zip The usage (benchmark and validation) of reference work unit is in the lower part of Ned's page for gcc/linux and this is the validation limits. In my personal experience with Linux Intel compiler/fftw, I've never got my results invalidated with them (few weeks.) generic SSE2 To crunch the "old" workunits from 4.09 with the new optimized 4.11 replace the app_info.xml file in the BOINCprojectssetiathome.berkeley.edu folder with this one: app_info.xml (right click and "save link to...") Otherwise you will lose all already downloaded workunits!!</blockquote> ...Maybe you can find some of the binaries here: <blockquote>1. for Pentium 3 and AMD Athlon seti-p3.zip 2. for Pentium 4 w/o SSE3 support seti-p4-no-sse3.zip 3. for Pentium 4 w SSE3 support seti-p4-sse3.zip</blockquote> NOTE: All the binaries listed above are currently (as of this writing) unavailable. If someone have them, please upload them to another web site, so other people also can get them! Some binaries for Athlon 64 are here: <blockquote>http://home.arcor.de/jojo4ubin/seti/</blockquote> Installation: BUT FIRST: MAKE BACKUP! <blockquote>1. SHUT BOINC DOWN 2. DETERMINE WHICH OPTIMITZED FILE TO DOWNLOAD FOR YOUR COMPUTER 3. UNZIP FILE 4. PUT APP_INFO.XML AND 4.11 APPLICATION IN setiathome.berkeley.edu 5. PUT BOINC CLIENT IN THE BOINC FOLDER Program FilesBOINC 6. START BOINC</blockquote> And here are the other threads in this series: Thread 1 Thread 2 Thread 3 |
![]() Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 ![]() |
Just to clean up this minor issue, the correct app_info.xml file did indeed fix the problem; I'm up and flying on 4.11. Thanks again go to Digger ! Hiya Gomeyer, I'm glad that you're up and running now with the optimized app! Don't thank me though, I'm just a follower here... Tetsuji and others did all the work. I do wish that everyone who distributes the optimized apps would include AndyK's app_info.xml file, since that is the one that works with your current cache. Even if you don't have remaining 4.09 units, that file works well in either case. Be sure to let us know what your new crunch times are. ;) Dig |
Pascal, K G ![]() Send message Joined: 3 Apr 99 Posts: 2343 Credit: 150,491 RAC: 0 ![]() |
I think all those links are dead or almost dead...... Semper Eadem So long Paul, it has been a hell of a ride. Park your ego's, fire up the computers, Science YES, Credits No. |
![]() ![]() Send message Joined: 5 Jul 99 Posts: 4548 Credit: 35,667,570 RAC: 4 ![]() |
A little summary (cut & paste from the other threads :)</blockquote> Hi Prognatus --- ![]() Thank you very much for your excellent Post this is the information I have been looking for ___ :) Greetings _ from _ [url=http://www.westerncanadahomestay.ca/Vancouver-01.jpg ] </B> the Pacific West Coast _of__ Vancover / Canada</B>[/url] Greetings _ to _[b] [url=http://home.no.net/arildbg/homelink/images/Tonsberg/tonsberg1a.jpg ]</B>_ Beautifully _ Tonsberg _ Norway's _oldest _ town </B> [/url] ........ happy crunching friendly and respectful byron |
![]() Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 ![]() |
I can offer at least one of the optimized clients. To be honest I have no idea what my bandwidth limits are so it might not last long. Intel P4 w/SSE3 (includes app_info.xml file to crunch remaining 4.09 work units) seti-p4-sse3.zip Dig |
Nova_nz_w Send message Joined: 26 May 02 Posts: 12 Credit: 198,265 RAC: 0 ![]() |
Anyone got the Pentium M optimised one? I'm running a 1700Mhz Dothan with MMX, SSE and SSE2 (no SSE3). Currently crunching a unit around 140mins. -edit- Oooh and the Athlon XP one would be nice for my computer at home as well. Would be interested to see how much the optimized client speeds things up! |
Graham Hamilton ![]() Send message Joined: 11 Jun 99 Posts: 12 Credit: 1,201,816 RAC: 0 ![]() |
More download links to download optimized clients from. Simple cut & paste from first post from Prognatus onto a single web page with working links to download seti-p3.zip, seti-p4-no-sse3.zip and seti-p4-sse3.zip. If i get a chance i'll improve the web page. ![]() |
![]() ![]() Send message Joined: 24 Feb 01 Posts: 44 Credit: 84,362 RAC: 0 ![]() |
Anyone got the Pentium M optimised one? I'm running a 1700Mhz Dothan with MMX, SSE and SSE2 (no SSE3). Currently crunching a unit around 140mins. SETI-PM |
Nova_nz_w Send message Joined: 26 May 02 Posts: 12 Credit: 198,265 RAC: 0 ![]() |
Anyone got the Pentium M optimised one? I'm running a 1700Mhz Dothan with MMX, SSE and SSE2 (no SSE3). Currently crunching a unit around 140mins. I think that Yahoo account is over it's limit (or the files are gone) |
![]() Send message Joined: 3 Apr 99 Posts: 1603 Credit: 2,700,523 RAC: 0 ![]() |
The solution hosting these apps seems clear. Berkeley should adopt and host these optimized versions. The size is almost the same the 'official' versions so there'll be little long-term difference to their net usage. They'll also benefit from more science being done. We'll automatically get the right apps, and a somewhat improved degree of security. ![]() |
![]() ![]() Send message Joined: 31 Aug 03 Posts: 848 Credit: 2,218,691 RAC: 0 ![]() |
Our hero Tetsuji Maverick Rai said that since he made the optimize client with a trial version, that we could only use it till May 29 2005. Has anyone out there been able to remake an optimized client for use after that date? Time is running out and it will be very hard to go back to the old one. |
Ulrich Metzner ![]() Send message Joined: 3 Jul 02 Posts: 1256 Credit: 13,565,513 RAC: 13 ![]() ![]() |
Would it be possible to compile the seti cruncher with the MinGW compiler? I use that compiler at work and it has many options to optimize for Athlon, AthlonXP and Athlon 64. The gnu compiler collection and it's generated code is freely distributable also. I would like to get a hand on an 'all in one' package 'ready to compile' for any platform. I'm in a hurry here at work and have the complete compiler environment available. If i could get it to compile and validate i will share the resulting cruncher with the community. Aloha, Uli |
![]() ![]() Send message Joined: 15 May 99 Posts: 875 Credit: 4,386,984 RAC: 0 ![]() |
Don't know why I got the wrong one, but I did - oh my did windows not like that one, wanting to report errors etc *lol* Anyway, while I got a number of computational errors and ran my quota down to 95, I finally got seti-p3 working on my athlon xp 3200+ (barton) - so know I only wait to see if it'll finish and get validated ![]() |
![]() Send message Joined: 6 Jul 99 Posts: 1600 Credit: 391,546 RAC: 0 ![]() |
Hi Byron! <blockquote>this is the information I have been looking for ___ :)</blockquote> Yeah, from my own searching through many hundreds of messages in the previous threads, I guessed other people also had the same problem. Also, the 3rd thread was getting huge, so therefore I compiled a summary in a new thread. It's by no means exhaustive or even satisfying, but better than nothing at all. :) Thanks for the picture of my town! Vancouver is not bad either! ;) Many thanks also to those who have made other download links available! |
![]() ![]() Send message Joined: 14 Jul 99 Posts: 335 Credit: 1,178,138 RAC: 0 ![]() |
I think all those links are dead or almost dead...... I mirrored some of these files on my server www.marisan.nl/seti/ If you feel some important files are missing just mail the file or link to me and i will put them online. If you don't want me mirroring your files, send me a mail and i will stop providing them. (speedy67@gmail.com) Greetings, Speedy67 ![]() ![]() |
![]() Send message Joined: 17 Nov 04 Posts: 182 Credit: 68,878 RAC: 0 ![]() |
unfortunately, some of the latest links posted do not have the fixes mentioned earlier so that a 4.09 would still gets crunched. Just installed boinc on a new computer, it downloaded 20 4.09 wu's, started crunching them, I stopped it, and believing that I could blindly put the stuff in, and it had the lastest xml file, that it would finish the 4.09, well, it didn't have the change, and I just lost my 20 wu's. Whoever is hosting the .xml files, please make aure that all of the .xml files included the changes required to run the 4.09 stuff to completion instead of deleting them. The above process worked 100% on several machines, because it had the correct .xml file that did 4.11 and 4.09 work to completion. This computer was on the net, so I found the latest link in this list, and downloaded, it, boy was I burned. So please, everyone that is hosting a link, please 1 - update your zips and tars to include the change 2 - repost the link so that peeps know it contains this change. Also, once this is done, we can't simply cut/paste the links, because it was a cut/paste link from a day ago that I got this download from. Thanks, and sorry, but I just got burned on something that was fixed a week ago. Thanks, and Keep on crunchin' John Henry KI4JPL Sevierville TN I started with nothing, and I still have some of it left. <img src="http://www.boincstats.com/stats/banner.php?cpid=989478996ebd8eadba8f0809051cdde2"> |
![]() ![]() Send message Joined: 14 Jul 99 Posts: 335 Credit: 1,178,138 RAC: 0 ![]() |
Fixed the xml file in the archives mirrored on www.marisan.nl/seti/ Thanks for the suggestion, Speedy67 ![]() ![]() |
![]() Send message Joined: 24 Oct 03 Posts: 27 Credit: 14,251,492 RAC: 0 ![]() |
Hello all, I upgraded my P-4 hpt boxes with the p4 apps and they are doing great, 35 to 55 percent increase depending on work unit. Thanks to all who compiled and tested thes programs. I used the link in Greg Hamiltons post below, be sure to download the seperate apt.info file and use it. As a side note I ahve seen no temp rises on my boxes these aps are very well done. So far all work units done have validated or are pending. I will report avgs when I have enough data. Again thanks for the hard work everyone. |
![]() ![]() Send message Joined: 15 May 99 Posts: 875 Credit: 4,386,984 RAC: 0 ![]() |
|
![]() Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 ![]() |
You can easily tell if you have the correct app_info.xml file by opening it up in Notepad or any text editor. You will note an additional section which references 4.09 and tells the client to process these workunits with the 4.11 app. I agree that this should be the only version that is provided by folks, which is why I only reference that version in my links. It works in all cases, whether or not you have remaining 4.09 work units to crunch. It's really nobody's fault though. There is merely a discrepancy between the original files that were provided, and the later one that AndyK devised. This optimization thing is still very much a work in progress. :) Dig |
©2025 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.