Deprecated: Function get_magic_quotes_gpc() is deprecated in /disks/centurion/b/carolyn/b/home/boincadm/projects/beta/html/inc/util.inc on line 663
MultiBeam for NV GPUs under Linux - why we miss it?

MultiBeam for NV GPUs under Linux - why we miss it?

Message boards : SETI@home Enhanced : MultiBeam for NV GPUs under Linux - why we miss it?
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 51299 - Posted: 26 Jun 2014, 9:45:39 UTC
Last modified: 26 Jun 2014, 9:46:09 UTC

subj.

Looking at NV AP Linux share on main one can conclude there are quite many NV devices running under Linux to justify porting effort.
Nethertheless we have no MB NV under Linux even on beta. Why so?
ID: 51299 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 51300 - Posted: 26 Jun 2014, 10:12:03 UTC - in response to Message 51299.  

subj.

Looking at NV AP Linux share on main one can conclude there are quite many NV devices running under Linux to justify porting effort.
Nethertheless we have no MB NV under Linux even on beta. Why so?


Because the Cuda NV Linux app is stuck in alpha for lack of developers?
There's been some discussion on main from people who have built x41zc with varying success iirc.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 51300 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 51303 - Posted: 26 Jun 2014, 10:16:22 UTC - in response to Message 51300.  

subj.

Looking at NV AP Linux share on main one can conclude there are quite many NV devices running under Linux to justify porting effort.
Nethertheless we have no MB NV under Linux even on beta. Why so?


Because the Cuda NV Linux app is stuck in alpha for lack of developers?
There's been some discussion on main from people who have built x41zc with varying success iirc.


But AFAIK there is working binary. Maybe worth to pass it for beta testing at least on beta?
There is at least 2 threads on main regarding building under Linux, one is even in "manual-style". Binary was received in result, not?
ID: 51303 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 51304 - Posted: 26 Jun 2014, 10:20:14 UTC
Last modified: 26 Jun 2014, 10:29:46 UTC

Here: http://setiathome.berkeley.edu/forum_thread.php?id=74768&postid=1522788 even benchmark results listed, not bad Qs.

EDIT: and here is that "manual":
http://setiathome.berkeley.edu/forum_thread.php?id=74883&postid=1524105

EDIT2: I see some discussion of upcoming x42 CUDA build, but as I understand it's some improvement of currently released for other platforms i.e. work in progress.
But for Linux NV MB we have nothing at all as stock - even "older" variant - that's the issue.
ID: 51304 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 51305 - Posted: 26 Jun 2014, 10:51:20 UTC

It's not up to me.
But I can raise the issue.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 51305 · Report as offensive
jason_gee
Volunteer tester

Send message
Joined: 11 Dec 08
Posts: 198
Credit: 658,573
RAC: 0
Australia
Message 51306 - Posted: 26 Jun 2014, 11:06:42 UTC - in response to Message 51305.  

From my end the limitations have been resources, i.e. both time and equipment. Eric was made aware that working builds exist, but I never received a response to questions as to exactly what's needed for a Linux release. A mac version exists as well, but suffers similar issues and some extra due to some sortof library and driver issues I'm not familiar with.

So to now, Linux and mac builds remain in the domain of anyone who wants to build them ... I won't be chasing stock implementation of something I can't provide and test myself yet, so the field is open to anyone that wishes to try.
Chaos: When the present determines the future, but the approximate present does not approximately determine the future.
Edward Lorenz
ID: 51306 · Report as offensive
Profile Raistmer
Volunteer tester
Avatar

Send message
Joined: 18 Aug 05
Posts: 2423
Credit: 15,878,738
RAC: 0
Russia
Message 51307 - Posted: 26 Jun 2014, 11:40:29 UTC - in response to Message 51306.  

Thanks for summarizing current state.
Looking at relative app performance:
http://setiathome.berkeley.edu/apps.php
I would say promoting Linux NV MB to stock would be good step in computational power increase.
Pity I'm not too familiar with Linux environment to provide any binaries, will see what can be done on this front.
ID: 51307 · Report as offensive
William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 606
Credit: 588,843
RAC: 0
Message 51321 - Posted: 27 Jun 2014, 9:30:13 UTC - in response to Message 51307.  

Thanks for summarizing current state.
Looking at relative app performance:
http://setiathome.berkeley.edu/apps.php
I would say promoting Linux NV MB to stock would be good step in computational power increase.
Pity I'm not too familiar with Linux environment to provide any binaries, will see what can be done on this front.

I'll let Richard explain why that 'app performance' has nothing at all to do with how apps really perform.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 51321 · Report as offensive
Richard Haselgrove
Volunteer tester

Send message
Joined: 3 Jan 07
Posts: 1451
Credit: 3,272,268
RAC: 0
United Kingdom
Message 51324 - Posted: 27 Jun 2014, 10:00:51 UTC - in response to Message 51321.  

Thanks for summarizing current state.
Looking at relative app performance:
http://setiathome.berkeley.edu/apps.php
I would say promoting Linux NV MB to stock would be good step in computational power increase.
Pity I'm not too familiar with Linux environment to provide any binaries, will see what can be done on this front.

I'll let Richard explain why that 'app performance' has nothing at all to do with how apps really perform.

Later. Bit busy with more important things this morning.
ID: 51324 · Report as offensive

Message boards : SETI@home Enhanced : MultiBeam for NV GPUs under Linux - why we miss it?


 
©2022 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.