Deprecated: trim(): Passing null to parameter #1 ($string) of type string is deprecated in /disks/centurion/b/carolyn/b/home/boincadm/projects/beta/html/inc/boinc_db.inc on line 147
SETI@home v8 beta to begin on Tuesday

SETI@home v8 beta to begin on Tuesday

Message boards : News : SETI@home v8 beta to begin on Tuesday
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 14 · 15 · 16 · 17 · 18 · 19 · 20 . . . 99 · Next

AuthorMessage
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 7 Jun 09
Posts: 285
Credit: 2,822,466
RAC: 0
Germany
Message 55601 - Posted: 4 Jan 2016, 0:26:32 UTC - in response to Message 55600.  

Does this mean, I choose the apps I would like to test and use an app_info.xml file?

Or is this info just for Eric?

What's the difference between HD5 build and HD5 Signals on GPU build?
(Which I should use for R9 FuryX VGA cards?)

- - - - - - - - - -

I want to give 1 GPU app 1 CPU-Core.
On my PC it's '0.119 CPU' with stock SETI@home v8 v8.04 (opencl_ati5_nocal) app.

How should look the app_config.xml file?
I tested a few times, but nothing worked.

Thanks.
ID: 55601 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 55602 - Posted: 4 Jan 2016, 0:30:21 UTC - in response to Message 55601.  
Last modified: 4 Jan 2016, 0:34:51 UTC

These builds for open beta testing so you "should" use ALL of them that compatible with your hardware, find better configs, find not working configs and report back. What you asking is just for what they were posted here - to find answers!

EDIT: the difference between SoG and "usual" one:
SoG means Signals on GPU - new computation regime where GPU<=>CPU communications are greatly reduced.
Should result in better GPU utilization... in theory...
News about SETI opt app releases: https://twitter.com/Raistmer
ID: 55602 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 55603 - Posted: 4 Jan 2016, 0:47:29 UTC

smth wrong with this app on main:

Mac OS/X 10.3+ 8.00 30 Dec 2015, 21:14:57 UTC 0 GigaFLOPS

still no flops...
News about SETI opt app releases: https://twitter.com/Raistmer
ID: 55603 · Report as offensive
Zalster
Volunteer tester

Send message
Joined: 30 Dec 13
Posts: 258
Credit: 12,340,341
RAC: 0
United States
Message 55605 - Posted: 4 Jan 2016, 5:07:26 UTC - in response to Message 55603.  

Thanks Raistmer,

Finally got it the NV to run.

Now to see if they validate.
ID: 55605 · Report as offensive
Profile Gary Charpentier
Volunteer tester
Avatar

Send message
Joined: 9 Apr 07
Posts: 1702
Credit: 4,622,751
RAC: 0
United States
Message 55606 - Posted: 4 Jan 2016, 5:15:12 UTC - in response to Message 55603.  

Is that "SETI@home v8 8.00 x86_64-apple-darwin" on main?
They all seem to be generating -9 exits.
Same computer.
http://setiathome.berkeley.edu/results.php?hostid=6463983
http://setiweb.ssl.berkeley.edu/beta/results.php?hostid=57003
Beta never sent it a W/U.
ID: 55606 · Report as offensive
Zalster
Volunteer tester

Send message
Joined: 30 Dec 13
Posts: 258
Credit: 12,340,341
RAC: 0
United States
Message 55608 - Posted: 4 Jan 2016, 5:23:05 UTC - in response to Message 55606.  

Yes, My mac over there has been using it

http://setiathome.berkeley.edu/results.php?hostid=7604284

Have not had any problems with it.
ID: 55608 · Report as offensive
Profile Eric J Korpela
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 15 Mar 05
Posts: 1547
Credit: 27,183,456
RAC: 0
United States
Message 55609 - Posted: 4 Jan 2016, 5:26:03 UTC - in response to Message 55606.  

Hmmmmm... But the 10.3 (PPC) issue is separate from whatever problem the x86_64 version is having.

In beta the x86_64-darwin version was successful on 90% of hosts (although that seems to have dropped to 85% recently.) It seemed to have its worst problems running on darwin 13.4.0 (I don't know apple's word for that version).

http://setiweb.ssl.berkeley.edu/beta/setiathome_v8_x86_64-apple-darwin.html
ID: 55609 · Report as offensive
zoom314
Volunteer tester

Send message
Joined: 29 Nov 14
Posts: 12
Credit: 63,078
RAC: 0
United States
Message 55610 - Posted: 4 Jan 2016, 5:30:49 UTC - in response to Message 55609.  

Hmmmmm... But the 10.3 (PPC) issue is separate from whatever problem the x86_64 version is having.

In beta the x86_64-darwin version was successful on 90% of hosts (although that seems to have dropped to 85% recently.) It seemed to have its worst problems running on darwin 13.4.0 (I don't know apple's word for that version).

http://setiweb.ssl.berkeley.edu/beta/setiathome_v8_x86_64-apple-darwin.html

Eric, the code name for 13.4.0 is "Mavericks" according to the Wiki.
ID: 55610 · Report as offensive
Profile Eric J Korpela
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 15 Mar 05
Posts: 1547
Credit: 27,183,456
RAC: 0
United States
Message 55611 - Posted: 4 Jan 2016, 5:43:54 UTC - in response to Message 55606.  

Beta never sent it a W/U.


I can't seen any reason for it not to have been sent v8 work besides...
2016-01-03 21:21:45.3861 [PID=21015]    [version] [AV#655] Don't need CPU jobs, skipping
2016-01-03 21:21:45.3861 [PID=21015]    [version] [AV#656] Don't need CPU jobs, skipping

ID: 55611 · Report as offensive
Zalster
Volunteer tester

Send message
Joined: 30 Dec 13
Posts: 258
Credit: 12,340,341
RAC: 0
United States
Message 55612 - Posted: 4 Jan 2016, 6:39:36 UTC
Last modified: 4 Jan 2016, 7:32:39 UTC

Not sure what happen.

About 10 minutes ago I had a MBv8 error out on my computer.

Following that, I started to receive the following notice in Boinc Manager

Message from server: Your app_info.xml file doesn't have a usable version of SETI@home v8

I looked at the app_info.xml and it's no different from what I put in more than an hour ago.

So I have to think it's related to the error

https://setiweb.ssl.berkeley.edu/beta//result.php?resultid=21602786

I'm also starting to see

This computer has finished a daily quota of xx tasks.

That number goes up as more results are returned but not getting any new work.

Any ideas? I tried closing BOINC and restarting it. Now I'm debating turning off the computer and rebooting.

Edit..

rebooting, will recheck in a few, reboot seems to have cleared that message
ID: 55612 · Report as offensive
TRuEQ & TuVaLu
Volunteer tester
Avatar

Send message
Joined: 28 Jan 11
Posts: 619
Credit: 2,580,051
RAC: 0
Sweden
Message 55615 - Posted: 4 Jan 2016, 10:59:19 UTC - in response to Message 55600.  
Last modified: 4 Jan 2016, 11:00:56 UTC

testing 123
ID: 55615 · Report as offensive
TRuEQ & TuVaLu
Volunteer tester
Avatar

Send message
Joined: 28 Jan 11
Posts: 619
Credit: 2,580,051
RAC: 0
Sweden
Message 55616 - Posted: 4 Jan 2016, 11:00:23 UTC - in response to Message 55600.  

Any chance a basic app_info.xml can follow with each package?




Most GPU apps updated now so use links from this post:

Here are few MultiBeam v8 apps for public beta testing in anonymous platform mode:

Windows x64 AVX

Windows x64 SSE3

Windows x86 SSE

Windows x86 SSE3

Windows OpenCL NV

Windows OpenCL iGPU

Windows OpenCL ATI HD5 build

Windows OpenCL ATi HD5 Signals on GPU build

Windows OpenCL ATi non-HD5 build

Windows OpenCL ATi APU build
ID: 55616 · Report as offensive
Richard Haselgrove
Volunteer tester

Send message
Joined: 3 Jan 07
Posts: 1451
Credit: 3,272,268
RAC: 0
United Kingdom
Message 55617 - Posted: 4 Jan 2016, 11:05:49 UTC - in response to Message 55616.  

Any chance a basic app_info.xml can follow with each package?

They're always - automatically - included in Raistmer's download packages. Look for a file called 'MB8_win_x64_AVX_VS2010.aistub' or similar.

These app_info segments are intended for Beta testing only, and are not guaranteed to be usable on the main project.
ID: 55617 · Report as offensive
TRuEQ & TuVaLu
Volunteer tester
Avatar

Send message
Joined: 28 Jan 11
Posts: 619
Credit: 2,580,051
RAC: 0
Sweden
Message 55618 - Posted: 4 Jan 2016, 11:09:23 UTC - in response to Message 55617.  

Wow, thanks.

I newer even looked into that file.




Any chance a basic app_info.xml can follow with each package?

They're always - automatically - included in Raistmer's download packages. Look for a file called 'MB8_win_x64_AVX_VS2010.aistub' or similar.

These app_info segments are intended for Beta testing only, and are not guaranteed to be usable on the main project.
ID: 55618 · Report as offensive
Juha
Volunteer tester

Send message
Joined: 18 Jun 08
Posts: 76
Credit: 113,089
RAC: 0
Finland
Message 55622 - Posted: 4 Jan 2016, 16:52:44 UTC - in response to Message 55599.  

Eric, small cosmetic bug discovered and fixerd in OpenCL GPU builds (printing TUNE option values and few others). Nothing serious but please postpone release a little while I rebuild...


I think you replaced one bug with another.

- fprintf(stderr, "TUNE: kernel %d now has workgroup size of (%zd,%zd,%zd)\n",
+ fprintf(stderr, "TUNE: kernel %d now has workgroup size of (%d,%d,zd)\n",
ID: 55622 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 55624 - Posted: 4 Jan 2016, 17:20:00 UTC - in response to Message 55622.  

Eric, small cosmetic bug discovered and fixerd in OpenCL GPU builds (printing TUNE option values and few others). Nothing serious but please postpone release a little while I rebuild...


I think you replaced one bug with another.

- fprintf(stderr, "TUNE: kernel %d now has workgroup size of (%zd,%zd,%zd)\n",
+ fprintf(stderr, "TUNE: kernel %d now has workgroup size of (%d,%d,zd)\n",

yep, unfortunately. New rebuild in progress....
News about SETI opt app releases: https://twitter.com/Raistmer
ID: 55624 · Report as offensive
Zalster
Volunteer tester

Send message
Joined: 30 Dec 13
Posts: 258
Credit: 12,340,341
RAC: 0
United States
Message 55625 - Posted: 4 Jan 2016, 17:42:21 UTC - in response to Message 55624.  
Last modified: 4 Jan 2016, 17:42:46 UTC

Was going to ask if we could start with multiple instances on a card since I had 300 validated returns, but going to wait now till you publish the new builds so I can retest singularly under those.
ID: 55625 · Report as offensive
Profile Jeff Buck
Volunteer tester

Send message
Joined: 11 Dec 14
Posts: 96
Credit: 1,240,941
RAC: 0
United States
Message 55626 - Posted: 4 Jan 2016, 19:04:55 UTC - in response to Message 55600.  

Will you have a Windows x86 SSE2 flavor available?

My old ThinkPad with a Pentium M did much better with AKv8c_r2549_winx86_SSE2xjf.exe than it did with stock on v7, and v8 stock looks even worse. (About 19 hours for a "normal" AR on v8 stock vs. about 12 hours for SSE2 on v7.)
ID: 55626 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 55628 - Posted: 4 Jan 2016, 21:36:47 UTC - in response to Message 55627.  

Windows x86 SSE2
News about SETI opt app releases: https://twitter.com/Raistmer
ID: 55628 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 7 Jun 09
Posts: 285
Credit: 2,822,466
RAC: 0
Germany
Message 55629 - Posted: 4 Jan 2016, 21:43:33 UTC - in response to Message 55627.  

The cmdline settings which I used for the SAHv7 ATI r2929 app, I can use them also for the HD5 build and HD5 Signals on GPU build apps?


BTW.
If I start BOINC is see:
OpenCL: AMD/ATI GPU 0: Fiji (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 5376 GFLOPS peak)	
OpenCL: AMD/ATI GPU 1: Fiji (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 5376 GFLOPS peak)	
OpenCL: AMD/ATI GPU 2: Fiji (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 5376 GFLOPS peak)	
OpenCL: AMD/ATI GPU 3: Fiji (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 5376 GFLOPS peak)	
OpenCL CPU: Intel(R) Xeon(R) CPU E5-2630 v2 @ 2.60GHz (OpenCL driver vendor: Advanced Micro Devices, Inc., driver version 1912.5 (sse2,avx), device version OpenCL 1.2 AMD-APP (1912.5))	

...does this mean that on my system it's possible to use an ATI app which support AVX?
The currently available ATI apps use 'just' SSE2 (on/off the CPU).

Thanks.
ID: 55629 · Report as offensive
Previous · 1 . . . 14 · 15 · 16 · 17 · 18 · 19 · 20 . . . 99 · Next

Message boards : News : SETI@home v8 beta to begin on Tuesday


 
©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.