Message boards :
Number crunching :
AP V7
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 . . . 20 · Next
Author | Message |
---|---|
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
Well that went smooothly, can report a "percent blanked: 74.54" done in under 28 mins on a 670. The Validator reads the amount of blanking which the application writes into the uploaded result file. That's in fraction blanked form. 0.1 or over is still considered a runtime outlier, because CPU applications now run considerably faster on tasks with significant blanking than on tasks with no blanking. Joe |
Wiggo Send message Joined: 24 Jan 00 Posts: 35208 Credit: 261,360,520 RAC: 489 |
It's been very strange that no official announcement has been made. *ponders?* Cheers. |
Lionel Send message Joined: 25 Mar 00 Posts: 680 Credit: 563,640,304 RAC: 597 |
For those that might be interested (my first Nvidia v7 AP WU): Task http://setiathome.berkeley.edu/result.php?resultid=3776536957 @Wiggo, not really. They have a tendency not to communicate very well with the volunteers. cheers[/url] |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Lionel, are you using commandlines? Just curious.... |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
I actually had one validate on my DDR3 7750, Workunit 1612617361 You gotta love that 'score'. I saw a few of those in testing. It should 'normalize' after 11 completions... |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
Yes, there seem to be a few more being sent out at the moment. r2713/2721 seem to be running very well, no issues so far. Thank you to everyone involved in the development and testing. Does anyone has available the links for the stock Linux APv7 applications, please? I understand Windows builds are the focus for the moment as the majority use that one, but I've never been able to successfully build my own Linux AstroPulse binaries. Soli Deo Gloria |
Lionel Send message Joined: 25 Mar 00 Posts: 680 Credit: 563,640,304 RAC: 597 |
Lionel, are you using commandlines? Just curious.... yes, I am using an ap_cmdline_win..... file and I have now added v7 AP to my app_info files for 3 of my puters (1 left to go) (only doing GPU application at the moment) cheers |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I have a question to add to all of this talk of new apps.. As it turned out credit wasn't ever "balanced" so no really deep sence to continue that practice. Number of active stock hosts is high enough to not force few on anonymous (few regarding number of stock hosts) platform to force go to stock... |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
I've received a couple of AstroPulse v7 v7.03 (sse), and AstroPulse v7 v7.05 (opencl_nvidia_100) tasks, But no AstroPulse v7 7.04 (opencl_ati_100) tasks (yet) which is strange. Edit: a little later one is sent. Glad to see you involved again! |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
The time estimate says 11:37:30 which is rather long ... Unfortunately, smooth estimations transition wasn't developed on beta and from beta to main. Each version update we will face same ridiculous estimations even if minor version number updates... |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Looks like MACs with ATI HD4000 are having an issue with AP v7. Thanks for observation. There are some non-fitments in what Apple's OpenCL runtime reports for HD4xxx device capabilities and how AMD treats such devices under other OSes. Here I tried to explain what could be wrong: http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=2182&postid=52760 |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
-- |
Darth Beaver Send message Joined: 20 Aug 99 Posts: 6728 Credit: 21,443,075 RAC: 3 |
mike I have just been to your site to see what you have put up and i'm not shore which one to download , ! I need the AP v7 but for AMD which ones or one do i download . I have Luna version 0.42 all ready installed I also noticed that the cuda app is 42 and not 50 is there one for cuda 50 as i do have a Kepler chip . Thank's |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
Astropulse V7 packages are now available on my website. Any CPU SSE or SSE2 available? If I'm not mistaken, only SSE3 is up there for CPU work right now? My apologies if I missed it. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
@Zeus, No, I just switched to ap7_win_x86_sse2_opencl_nv_r2721 Just a heads-up that in your example, the second app_version plan_class is the same as the first one. Should be open_cl_nvidia_100 rather than cuda_open_cl_100, if I'm reading Mike's aistub correctly? As it reads right now, I think it's just a dupe ... Regards, Jim .. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Astropulse V7 packages are now available on my website. Here: https://www.dropbox.com/s/eb39ijd2czm32ev/AP7_win_x86_SSE_CPU_r2703.7z?dl=0 should be little faster than rev distributed as SSE x86 stock for Windows. Regarding SSE2 - current build not faster than "stock" SSE2 x86 windows one so better to use it. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
@Arkayn
Sorry for the duplicate question, as I don't know who's owning this. Any chance we'll see an SSE2 version of R2692 for those of us who lack SSE3 or AVX? Not sure I'll go back to doing any CPU work, but would be nice to have the option. Regards, Jim ... |
W-K 666 Send message Joined: 18 May 99 Posts: 19158 Credit: 40,757,560 RAC: 67 |
Well that went smooothly, can report a "percent blanked: 74.54" done in under 28 mins on a 670. Thanks Joe, But have to report that I am not seeing that. The high percentage blanking tasks are taking approx the same time as tasks with only a few percent blanking. http://setiathome.berkeley.edu/results.php?hostid=7006214&offset=0&show_names=0&state=0&appid=20 resultid=3776428437, percent blanked: 7.88, Run time 29 min 40 sec resultid=3776428962, percent blanked: 83.09, Run time 28 min 45 sec |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Well that went smooothly, can report a "percent blanked: 74.54" done in under 28 mins on a 670. Joe said about CPU app. You cite result for GPU app. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
Wow, was that ever quick. Many thanks!!! @Mike, @Arkayn, disregard. |
©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.