Now that v7 has rolled out..........

Message boards : Number crunching : Now that v7 has rolled out..........
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 18 · Next

AuthorMessage
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34268
Credit: 79,922,639
RAC: 80
Germany
Message 1372814 - Posted: 28 May 2013, 20:57:02 UTC - in response to Message 1372807.  
Last modified: 28 May 2013, 20:57:16 UTC

I think William forgot the first step:

Wait until v7 is actually out in the field, then....


It would be a shame to run caches down, edit files, then find a problem has delayed the release by a day or so and you've nothing to chew on.


Maybe we are simply one step ahead.

I`m almost certain he didn`t forget that.


With each crime and every kindness we birth our future.
ID: 1372814 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65817
Credit: 55,293,173
RAC: 49
United States
Message 1372942 - Posted: 29 May 2013, 5:10:28 UTC

I can only wait, when it's available someone will post a link, I'll figure it out from there or at least I should, I hope.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 1372942 · Report as offensive
Profile Sutaru Tsureku
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1372946 - Posted: 29 May 2013, 5:23:37 UTC
Last modified: 29 May 2013, 5:26:02 UTC

Is then there an easy way to see somewhere how Eric named the new SAHv7 apps here?
Is there a file which I could D/L manually?

I mean not just this overview: http://setiathome.berkeley.edu/apps.php.

How will be the entries in client_state.XML file?

Example, like at SAH-beta:
<name>setiathome_v7</name>
?

Thanks.


[EDIT: Or, if someone is running stock, please could you post the entry then?]


* Best regards! :-) * Philip J. Fry, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *
ID: 1372946 · Report as offensive
Profile Dingo
Volunteer tester
Avatar

Send message
Joined: 28 Jun 99
Posts: 104
Credit: 16,364,896
RAC: 1
Australia
Message 1372952 - Posted: 29 May 2013, 5:41:55 UTC - in response to Message 1372755.  
Last modified: 29 May 2013, 5:43:36 UTC

This post http://setiathome.berkeley.edu/forum_thread.php?id=71728&postid=1372741#1372741 is based on Windows machines, I presume as there are no standard Linux GPU apps so you will need to download and change all the app_info.xml I believe ??

Proud Founder and member of



Have a look at my WebCam
ID: 1372952 · Report as offensive
BetelgeuseFive Project Donor
Volunteer tester

Send message
Joined: 6 Jul 99
Posts: 158
Credit: 17,117,787
RAC: 19
Netherlands
Message 1373002 - Posted: 29 May 2013, 8:17:36 UTC - in response to Message 1372755.  


If you run optimised seti only on GPU and run boinc 7.0.64 -> go stock: run down your v6 cache, when empty remove app_info.xml, tick v7 and untick v6, reset project to get stock apps. Use app_config.xml to run several tasks at once.
[you won't get v6 resends, but you would be processing them with 6.08/6.09/6.10 which isn't really desirable (if you have an NV that is)]


I am running both optimized CUDA multibeam and optimized CUDA astropulse. Is a stock CUDA astropulse application available or will I no longer be able to run astropulse on CUDA if I revert to stock ?

Tom

ID: 1373002 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14655
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1373017 - Posted: 29 May 2013, 9:15:43 UTC - in response to Message 1372946.  

Is then there an easy way to see somewhere how Eric named the new SAHv7 apps here?
Is there a file which I could D/L manually?

I mean not just this overview: http://setiathome.berkeley.edu/apps.php.

How will be the entries in client_state.XML file?

Thanks.

[EDIT: Or, if someone is running stock, please could you post the entry then?]

Eric hasn't named anything yet.

Until entries for "SETI@home v7" appear on the applications page you linked, nobody will be running stock, and nobody can predict exactly what entries will be placed in client_state.xml - please don't encourage premature speculation.
ID: 1373017 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1373019 - Posted: 29 May 2013, 9:25:42 UTC - in response to Message 1373002.  


If you run optimised seti only on GPU and run boinc 7.0.64 -> go stock: run down your v6 cache, when empty remove app_info.xml, tick v7 and untick v6, reset project to get stock apps. Use app_config.xml to run several tasks at once.
[you won't get v6 resends, but you would be processing them with 6.08/6.09/6.10 which isn't really desirable (if you have an NV that is)]


I am running both optimized CUDA multibeam and optimized CUDA astropulse. Is a stock CUDA astropulse application available or will I no longer be able to run astropulse on CUDA if I revert to stock ?

Tom

Sorry I keep forgetting we have AP too...

There is a stock AP for NVidia app - it's not CUDA btw it's OpenCL.

I've no idea what revision is currently going out as stock and if there are plans to update it. The installer will have the latest revision.

IOW yes there is a stock app, but it may be an earlier version than what you are running now and installer will certainly have a newer version.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1373019 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1373024 - Posted: 29 May 2013, 10:03:13 UTC - in response to Message 1372952.  

This post http://setiathome.berkeley.edu/forum_thread.php?id=71728&postid=1372741#1372741 is based on Windows machines, I presume as there are no standard Linux GPU apps so you will need to download and change all the app_info.xml I believe ??

made link clickable - highlight url and press url button above edit window for quick BBcode tagging.

Urs is working on updated ATI apps and AFAIK is planning to release when we know what app_info.xml entries we need.

IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.

Yes, for Ati there should be new release packages incl. app_info, for NV/x41g you only need to update app_info.xml. We will try and provide guidance when V7 has been launched but our efforts are concentrating on the Windows installer.
Basically you only need to add a v7 section but it needs the correct <name> entries among other things, which we'll only know AFTER Eric has released.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1373024 · Report as offensive
Profile Sutaru Tsureku
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 1373037 - Posted: 29 May 2013, 12:21:36 UTC - in response to Message 1373017.  

Philip J. Fry wrote:
Is then there an easy way to see somewhere how Eric named the new SAHv7 apps here?
Is there a file which I could D/L manually?

I mean not just this overview: http://setiathome.berkeley.edu/apps.php.

How will be the entries in client_state.XML file?

Example, like at SAH-beta:
<name>setiathome_v7</name>
?

Thanks.


[EDIT: Or, if someone is running stock, please could you post the entry then?]

Richard Haselgrove wrote:
Eric hasn't named anything yet.

Until entries for "SETI@home v7" appear on the applications page you linked, nobody will be running stock, and nobody can predict exactly what entries will be placed in client_state.xml - please don't encourage premature speculation.

You don't know my poor english?

After installation of course, there is no other way to see online which app name is used, only to look in client_state.XML file?

I meant, if the SAHv7 apps are installed if someone can tell the name which is used in the client_state.XML file.


* Best regards! :-) * Philip J. Fry, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *
ID: 1373037 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1373044 - Posted: 29 May 2013, 12:54:10 UTC - in response to Message 1373024.  

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1373044 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51469
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1373045 - Posted: 29 May 2013, 13:03:12 UTC - in response to Message 1373044.  

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

And how does x41z fare?
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1373045 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1373046 - Posted: 29 May 2013, 13:07:35 UTC - in response to Message 1373045.  

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

And how does x41z fare?

There were various Bugfixes between x41z and x41zc to insure the apps found the correct signals.

Claggy
ID: 1373046 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1373048 - Posted: 29 May 2013, 13:11:57 UTC - in response to Message 1373045.  
Last modified: 29 May 2013, 13:14:17 UTC

And how does x41z fare?


z was first commit to Berkeley of (X-branch). Changes after that were relatively minor, though nVidia did make some driver changes at or after 301.42 that caused Cuda 3.2 to snuff out some code, making the pulsefinding unstable on Kepler GPUs (applying to Cuda 3.2). Those problems in that build could reappear with new hardware/drivers, so it's deprecated.

x41zb introduced a typo bug, so with that fixed in x41zc, x41zc will be the one to use all around (for the time being anyway).
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1373048 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1373050 - Posted: 29 May 2013, 13:19:53 UTC - in response to Message 1373044.  

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

Unfortunately the Linux community doesn't have much of a choice at present...
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1373050 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1373051 - Posted: 29 May 2013, 13:26:20 UTC - in response to Message 1373050.  

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

Unfortunately the Linux community doesn't have much of a choice at present...

And the MAC community does have an app to test, but they have neglected to do any running on Seti Beta with v7 Wu's at all. (When I last checked the top 1000 odd hosts at Seti Beta)

Claggy
ID: 1373051 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1373052 - Posted: 29 May 2013, 13:27:11 UTC - in response to Message 1373050.  
Last modified: 29 May 2013, 13:31:29 UTC

Unfortunately the Linux community doesn't have much of a choice at present...


All someone really needs to do there, as happened with Mac already [...well started anyway...] , is build it & report problems to me. Sadly I'm not setup for Linux development at this time (or Mac either), and have some personal time pressures.

Typically, these guys come out into the sunlight as soon as things become urgent. Usually a day or so after public rollout.
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1373052 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1373053 - Posted: 29 May 2013, 13:32:55 UTC - in response to Message 1373052.  

Unfortunately the Linux community doesn't have much of a choice at present...


All someone really needs to do there, as happened with Mac already, is build it & report problems to me. Sadly I'm not setup for Linux development at this time (or Mac either), and have some personal time pressures.

Typically, these guys come out into the sunlight as soon as things become urgent. Usually a day or so after public rollout.

Vapourware?

Until there is something a bad app might be better than no app at all.
There is a differnce between something being 'viable' and 'recommended'.

Ok, so you don't recommend it, but is it _viable_ ?
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1373053 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1373055 - Posted: 29 May 2013, 13:36:36 UTC - in response to Message 1373053.  

Ok, so you don't recommend it, but is it _viable_ ?


You'd have to ask someone that runs Linux that.

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1373055 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1373059 - Posted: 29 May 2013, 13:47:04 UTC - in response to Message 1373055.  

Ok, so you don't recommend it, but is it _viable_ ?


You'd have to ask someone that runs Linux that.

D'oh.

Wonderful. Do I hear anybody going 'Me, me!' When I ask whether anybody has tried running Linux x41g on beta?
Does anybody feel like checking beta for any Linux users who are doing that?

@Dingo I'm afraid you are going to have the unique opportunity to Alpha x41g for v7 here on main. Do start a separate thread.
[I'm pretty sure there are more Linux people running x41g right now, you just happen to be the one who asked the question]
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1373059 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1373115 - Posted: 29 May 2013, 16:48:14 UTC - in response to Message 1373051.  
Last modified: 29 May 2013, 17:10:41 UTC

...IIRC x41g is v7 ready. Oh shoot. What it hasn't got is 2/3 of an alphabet worth of improvements and some of them are accuracy and such. I'm looking at a (partial) version list to try and gauge the v7 implication. From my POV x41g will have a higher rate of inconclusives than x41zc but should be viable. I hope Jason can confirm.


Sortof. The V7 part was a working 'proof of concept'. It has rather inefficient implementation of the Autocorrelations, compared to x41zc, and overall zc is so much further along the refinement path I would recommend avoiding x41g for v7 use altogether.

Unfortunately the Linux community doesn't have much of a choice at present...

And the MAC community does have an app to test, but they have neglected to do any running on Seti Beta with v7 Wu's at all. (When I last checked the top 1000 odd hosts at Seti Beta)

Claggy

I tried putting both my old CUDA cards in my Lion Mac to see how that would work for the new Mac App, and the thing wouldn't even boot/post. It boots fine with the PC ATI 6850 and the Mac/PC NV 8800, boots fine with the PC 6850 and the PC GTS 250. But it won't boot/post at all with the Mac/PC 8800 and PC 250. Whut? Anyone have a clue about that? Seems awful strange to me. An easy, for me, solution would be an ATI Mac App to work with the 6850 :-)
ID: 1373115 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 18 · Next

Message boards : Number crunching : Now that v7 has rolled out..........


 
©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.