!!! WANTED !!! OPTIMISED CLIENT -> WINDOWS x86 32bit or 64bit

Author	Message
Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 199753 - Posted: 1 Dec 2005, 1:49:11 UTC - in response to Message 199731. Wich source did you use ? All of them. They worked all, but as you know yourself (you ran my client for couple of weeks too), some results finish prematurely with computation errors due to memory access violation. I'd like to know if it is something that comes from the sources, from the compiler flags, from the compiler or linker versions, or from some dll's or static libraries. So both, the source code, make files and information about your ICC/IPP/MSVC component build versions would help me to figure it out. You'll get the source @ weekend. Hopefully you can wait till then :-) Join BOINC United now! ID: 199753 ·

trux Volunteer tester Send message Joined: 6 Feb 01 Posts: 344 Credit: 1,127,051 RAC: 0	Message 199762 - Posted: 1 Dec 2005, 2:02:58 UTC - in response to Message 199753. You'll get the source @ weekend. Hopefully you can wait till then :-) Well, I do not understand why there is all this secrecy (finally you started with the source I sent to Harold anyway without any hassle), but will not force you to give it up. Could you at least tell me your version builds of MSVC/ICC/IPP and the OS build version you built it on? trux BOINC software Freediving Team Czech Republic ID: 199762 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 199773 - Posted: 1 Dec 2005, 2:20:10 UTC - in response to Message 199762. Last modified: 1 Dec 2005, 2:20:32 UTC You'll get the source @ weekend. Hopefully you can wait till then :-) Well, I do not understand why there is all this secrecy (finally you started with the source I sent to Harold anyway without any hassle), but will not force you to give it up. Could you at least tell me your version builds of MSVC/ICC/IPP and the OS build version you built it on? It's 3.15 am in the morning here you do'nt expect me to mail you the stuff right now do'nt you ?? (need to go to bed now) And there's no secrecy at all.. MSVC 2003 ICL Intel(R) C++ Compiler 9.0.024 IPP ipp_ia32 5.0.023 OS: Win64 Join BOINC United now! ID: 199773 ·

-= Vyper =- Volunteer tester Send message Joined: 5 Sep 99 Posts: 1652 Credit: 1,065,191,981 RAC: 2,537	Message 199942 - Posted: 1 Dec 2005, 6:47:50 UTC Last modified: 1 Dec 2005, 6:51:44 UTC No big words here fellows.. Remember i started this thread.. cough Joking!! :) I really hope you can sort it out because this is a huge leap forward in calculating workunits faster. But is it really 100% accurate or just nearly? I'm a bit suspicious to the cache thing though it works enough to let the validator validate the results with other calculated units.. What is the experts oppinions about this?? Crunch3r , Trux, Tmr... //Vyper P.S BTW isn't there any compiler out there that uses the X86-64 extra registers better in Win64 ?? I'm just so puzzled that other applications can get a 40% boost over a x86-64 Windows version compared to a 32bit version but Seti cannot?? That's a big question too why the use of extra registers in the processor won't even help there to the seti compiled linux versions which often are slower than their 32 bit counterpart.. D.S _________________________________________________________________________ Addicted to SETI crunching! Founder of GPU Users Group ID: 199942 ·

-= Vyper =- Volunteer tester Send message Joined: 5 Sep 99 Posts: 1652 Credit: 1,065,191,981 RAC: 2,537	Message 199948 - Posted: 1 Dec 2005, 6:54:06 UTC - in response to Message 199502. Sorry that i've missed my own thread.. "Sorry to disappoint you, but I guess you're looking at the output of the linux client. Harold's binary seems to print out a comment line that came in with some changes made by truXoft. Did you check the host's OS? You might post the host ID if you like." That person is running Windows XP SP2 Professional. HostID 541941 Hi Vyper, this person seems to run the windows version that has been pulled because it created access violations. Regards Hans Oh i see but it's nevertheless really annoying that a regular P43.2 squezzes past my X24400+ in top computer rankings.. Grr :) _________________________________________________________________________ Addicted to SETI crunching! Founder of GPU Users Group ID: 199948 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 200411 - Posted: 1 Dec 2005, 21:01:31 UTC - in response to Message 199948. Hello, i'm looking for Trux. Please contact me via email. (can be found on my page) Join BOINC United now! ID: 200411 ·

trux Volunteer tester Send message Joined: 6 Feb 01 Posts: 344 Credit: 1,127,051 RAC: 0	Message 200599 - Posted: 2 Dec 2005, 0:24:27 UTC thanks, Crunch3r, I've just sent you an email trux BOINC software Freediving Team Czech Republic ID: 200599 ·

Karsten Vinding Volunteer tester Send message Joined: 18 May 99 Posts: 239 Credit: 25,201,931 RAC: 11	Message 205862 - Posted: 7 Dec 2005, 17:44:47 UTC Is there any progress with this project? ID: 205862 ·

Darth Vader Send message Joined: 10 May 02 Posts: 72 Credit: 110,105 RAC: 0	Message 205900 - Posted: 7 Dec 2005, 18:28:35 UTC - in response to Message 205862. Is there any progress with this project? I'm convinced Crunch3r will let us know right away when the work is done. ID: 205900 ·

kevint Volunteer tester Send message Joined: 17 May 99 Posts: 414 Credit: 11,680,240 RAC: 0	Message 205947 - Posted: 7 Dec 2005, 21:59:07 UTC - in response to Message 199309. To be honest i think the windows version will be faster. But you're talking about your new client, which you won't release,right? Actually I don't want to buy ipp for $199 and icc for $399. Maybe you could take up a donation - I would be willing to send you something for your efforts! ID: 205947 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 207005 - Posted: 8 Dec 2005, 19:36:17 UTC - in response to Message 205947. To be honest i think the windows version will be faster. But you're talking about your new client, which you won't release,right? Actually I don't want to buy ipp for $199 and icc for $399. Maybe you could take up a donation - I would be willing to send you something for your efforts! Hi, the new client is ready for download (got the licenses today). You can get it here. Join BOINC United now! ID: 207005 ·

spacemeat Send message Joined: 4 Oct 99 Posts: 239 Credit: 8,425,288 RAC: 0	Message 207011 - Posted: 8 Dec 2005, 19:43:08 UTC - in response to Message 207005. Hi, the new client is ready for download (got the licenses today). You can get it here. oh no, here we go ID: 207011 ·

Arnulf Volunteer tester Send message Joined: 23 Oct 00 Posts: 9 Credit: 14,854,293 RAC: 16	Message 207061 - Posted: 8 Dec 2005, 20:26:27 UTC - in response to Message 207011. Hi, the new client is ready for download (got the licenses today). You can get it here. oh no, here we go Yes - here we go! :D I just hope I can use the Athlon64 SSE2 client on my Opteron rig?! ID: 207061 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 207075 - Posted: 8 Dec 2005, 20:30:31 UTC - in response to Message 207061. Hi, the new client is ready for download (got the licenses today). You can get it here. oh no, here we go Yes - here we go! :D I just hope I can use the Athlon64 SSE2 client on my Opteron rig?! Yes you can. Join BOINC United now! ID: 207075 ·

spacemeat Send message Joined: 4 Oct 99 Posts: 239 Credit: 8,425,288 RAC: 0	Message 207088 - Posted: 8 Dec 2005, 20:35:39 UTC Crunch3r, have you benchmarked the improvements yet? i have not completed a unit yet, but on a 2.8 Xeon (SSE2) Boinc manager is projecting somewhere not far from a speed improvement of 100% over the last YAOSCW but who knows how accurate that guess is. ID: 207088 ·

Jim Send message Joined: 28 Jan 00 Posts: 614 Credit: 2,031,206 RAC: 0	Message 207092 - Posted: 8 Dec 2005, 20:40:44 UTC - in response to Message 207005. the new client is ready for download (got the licenses today). You can get it here. [/b] Hey there I rushed over to your site to give them a try and noticed your strong caution against OC'd cpu's and mem (no cheap memory here). What are the possible repercussions? I suspect wu's with computation errors or am I wrong? By the way, waiting for your post to say they were ready gave me the same feeling I get when I'm expecting a cool new computer part to be delivered any day. Thank you Santa! Without love, breath is just a clock ... ticking. Equilibrium ID: 207092 ·

Sergey Broudkov Send message Joined: 24 May 04 Posts: 221 Credit: 561,897 RAC: 0	Message 207133 - Posted: 8 Dec 2005, 20:59:39 UTC - in response to Message 207005. Last modified: 8 Dec 2005, 21:02:40 UTC the new client is ready for download (got the licenses today). You can get it here. From your site: News: December 08, 2005 Release of the SETI@Home clients with catching functions for Windows. Do you mean caching? Kitty@SETI team (Russia). Our cats also want to know if there is ETI out there ID: 207133 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 207139 - Posted: 8 Dec 2005, 21:02:57 UTC - in response to Message 207092. What are the possible repercussions? I suspect wu's with computation errors or am I wrong? Your're right. The client stores data in a cache for further use so it doesn't have to compute it all over again and again (thanks to hans dorn for his briliant idea). Heavy overclocking or faulty ram can cause a coruption of the data stored in the cache. That's what the warning is all about. Join BOINC United now! ID: 207139 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 207142 - Posted: 8 Dec 2005, 21:04:09 UTC - in response to Message 207133. the new client is ready for download (got the licenses today). You can get it here. From your site: News: December 08, 2005 Release of the SETI@Home clients with catching functions for Windows. Do you mean caching? Yes. hmm typo :D Join BOINC United now! ID: 207142 ·

Mr.Pernod Volunteer tester Send message Joined: 8 Feb 04 Posts: 350 Credit: 1,015,988 RAC: 0	Message 207150 - Posted: 8 Dec 2005, 21:13:04 UTC Crunch3r the app_info.xml in the zipfile for P4_SSE2 seems to be missing the entry to redirect 4.18 to the new optimized app ??? ID: 207150 ·

©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.