The key difference from V9 packs is :

The "team" mod now supports multi-GPU configs.

There will be number_of_GPUs file in SETI project directory. By default it contains number of 1.

If you have more GPUs just edit that file (enter number of GPUs installed in host instead of 1 ).
And don't forget to change your cc_config.xml
Now ncpus value required to be NUMBER_OF_LOGICAL_PROCESSORS+NUMBER_OF_GPUs for host.
For example, for Quad with 2 CUDA-enabled GPUs it should be 6
Here is example of minimal cc_config.xml file you need
<cc_config>
<options>
<ncpus>NUMBER_OF_LOGICAL_PROCESSORS+NUMBER_OF_GPUs</ncpus>
</options>
</cc_config>
File should be placed in boinc data directory (the one that contains projects subdirectory).

When you post to this thread seeking for help, please, don't forget to provide link on your host and description of your config (OS, number of GPU cards, what pack you use, video driver version).
It will wastly decrease number of unneeded questions and save time both for you and anyone who will help you.
Please, don't forget to check ALL STUFF (including these packs) you downloaded from Internet with updated antivirus.
http://lunatics.kwsn.net/gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg14734.html#msg14734

Thank you for all your incredibly fast work.

A question: What are the limitations regarding mixing different GPUs?

For instance, if I install a 9000 series and a 200 series GPU in the same machine, will the team mod work?

---

--
Classic 82353 WU / 400979 h

Seems to run fine on my x64 vista machine.

Question:
This host is a core i7 with a GTX295, so I set Number of CPUs to 10 and number of GPUs to 2, right?

Now the real question, if i specify in preferences for boinc to use 99% of the processors (thereby disabling one cpu core, I like to keep one core free for general computing...), will it still work both cores of the GPU?

I've been trying to keep an eye on the processes runing, and this seems to be the case. Can I have confirmation of that please.
/Anton

I keep getting these exit with zero status but without finished file since the switch over to V10. The message displays multiple times for the same workunit.

qutote from BOINC:
02/25/09 09:21:40|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:22:14|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:22:14|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:22:14|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:22:47|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:22:47|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:22:47|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:23:26|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:23:26|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:23:26|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:23:57|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:23:57|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:23:57|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:24:41|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:24:41|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:24:41|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:24:57|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:24:57|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:24:57|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608

Has this something to do with the (your) app, or is it just a coinsidence?
The computer in question is hostid=4242105

/Anton

Edit: Went back to V9 Multi_GPU and no more of these messages. Now that there are AP's out again, this will work just fine.

---

Thank you for all your incredibly fast work.

A question: What are the limitations regarding mixing different GPUs?

For instance, if I install a 9000 series and a 200 series GPU in the same machine, will the team mod work?

Yes, it should. All cude devices treated equally. PErformance will be different of course.

Seems to run fine on my x64 vista machine.

Question:
This host is a core i7 with a GTX295, so I set Number of CPUs to 10 and number of GPUs to 2, right?

Now the real question, if i specify in preferences for boinc to use 99% of the processors (thereby disabling one cpu core, I like to keep one core free for general computing...), will it still work both cores of the GPU?

I've been trying to keep an eye on the processes runing, and this seems to be the case. Can I have confirmation of that please.
/Anton

For team mod the ncpus value will define how many tasks BOINC runs simultaneously, nothing else. How these tasks will be distributed between CPU and GPU depends of BOINC scheduler, used projects and projects shares. That is, if BOINC enters into high priority mode for einstein (for example) it will run 10 einstein tasks (if it will allowed to do that) and NO CUDA MB at all.
V10 has this limitation, cause it as all prev "team" mod tries to do BOINC's job actually. Science app shouldn't replace BOINC scheduling mechanism entirely. It can only insert some corrections in its work.

But SETI fans (especially SETI MB ones) can do SETI MB work on all available cores and all available CUDA GPUs now.

I keep getting these exit with zero status but without finished file since the switch over to V10. The message displays multiple times for the same workunit.

qutote from BOINC:
02/25/09 09:21:40|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:22:14|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:22:14|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:22:14|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:22:47|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:22:47|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:22:47|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:23:26|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:23:26|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:23:26|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:23:57|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:23:57|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:23:57|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608
02/25/09 09:24:41|SETI@home|Task 03ja09ad.7389.5389.6.8.26_1 exited with zero status but no 'finished' file
02/25/09 09:24:41|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:24:41|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.26_1 using setiathome_enhanced version 608
02/25/09 09:24:57|SETI@home|Task 03ja09ad.7389.5389.6.8.96_1 exited with zero status but no 'finished' file
02/25/09 09:24:57|SETI@home|If this happens repeatedly you may need to reset the project.
02/25/09 09:24:57|SETI@home|Restarting task 03ja09ad.7389.5389.6.8.96_1 using setiathome_enhanced version 608

Has this something to do with the (your) app, or is it just a coinsidence?
The computer in question is hostid=4242105

/Anton

Edit: Went back to V9 Multi_GPU and no more of these messages. Now that there are AP's out again, this will work just fine.

These messages probably mean that CPU part of team tries to reschedule task to GPU part (and that means GPU part was idle at that moment). Look into stderr output of that task.

EDIT:
There is a flaw in current re-schedule logic in CPU app that will lead to excessive exits for CPU apps. Thanks for noticing this behavior, bug-fixing in progress.

I presume this one requires CUDA2.1 too?
BR,

---

I presume this one requires CUDA2.1 too?
BR,

IT doesn't require CUDA 2.1 It contains CUDA 2.1 DLLs too. But if you experience problems with 2.1 DLLs version (app freezes) you can revert DLL to older version, the app should work with older DLLs too (at least there were reports it works).

Link to binaries removed for now, bug-fixing in progress.

V10a posted. Now should re-schedule from CPU to GPU correctly.

I presume this one requires CUDA2.1 too?
BR,

IT doesn't require CUDA 2.1 It contains CUDA 2.1 DLLs too. But if you experience problems with 2.1 DLLs version (app freezes) you can revert DLL to older version, the app should work with older DLLs too (at least there were reports it works).

Thanks, will try it out.
BR,

---

What version of video drivers does one need for this?

I've currently got driver 181.22.

---

Hi Raistmer - just a couple of questions: Is V10a a recommended upgrade to V9 for single GPU crunchers?

And if I want to temporarily suspend running MB on CPU (e.g. go back to AP on CPU and MB on CUDA), is there a simple mod to App_info.xml to do the trick (that will not borg the entire WU cache)?

Thanks!

---

Is the AP app in V10 or V10a 32bit or 64bit for Windows(XP x64 and Vista x64)? As I don't run 32bit code in Boinc...

---

Savoir-Faire is everywhere!
The T1 Trust, T1 Class 4-4-4-4 #5550, America's First HST

Sweet! Just stuck it on the Tesla. Thank you very much brudda for all your hard work. Know you are all aprreciated. Seems to be pluggin' right along. Say what ever happend to 6.6.10? Just went for try download em and it has some new wierd version instead 6.4.5 and only 6.6.9...
Anyway just like fo give some mahalos Raist!

---

"The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" -- Isaac Asimov

Hi Raistmer - just a couple of questions: Is V10a a recommended upgrade to V9 for single GPU crunchers?

And if I want to temporarily suspend running MB on CPU (e.g. go back to AP on CPU and MB on CUDA), is there a simple mod to App_info.xml to do the trick (that will not borg the entire WU cache)?

Thanks!

Better suspend all SETI MB tasks.
If you will remove app that processes MB tasks all these tasks will be purged at once.

And never break working system. There is no speed increase in V10 over V9.

What version of video drivers does one need for this?

I've currently got driver 181.22.

The same as for all "team" builds.

ok, im stuck with 9 ap's running on my 8-core with v8-10 with mbs sitting there waiting for some cuda action, not very satisfactory

i also want to return to mb on cuda only and optimised ap's on cpu only

is there an edit to the app info that will allow this ? is so could someone kindly post the edit please

boinc appears incapable of sharing the work in a logical manner, not suprising seeing as though we are trying to fool it :-)

this is in no way detracting from your great work Raistmer :-) i just want a simpler setup

regards

adrian

---

63. (1) (b) "music" includes sounds wholly or predominantly characterised by the emission of a succession of repetitive beats

SSSE3 x64 variant added.