Make your SETI@home preferences to:
Run only the selected applications
SETI@home Enhanced: yes
Astropulse: no
Astropulse v5: no

If no work for selected applications is available, accept work from other applications? no
Use Graphics Processing Unit (GPU) if available yes
Use Central Processing Unit (CPU) no

With such setting and V10 pack he will not recive any work from SETI at all.

Run only the selected applications SETI@home Enhanced: yes
Astropulse: no
Astropulse v5: no

If no work for selected applications is available, accept work from other applications? no
Use Graphics Processing Unit (GPU) if available yes
Use Central Processing Unit (CPU) no


Just cut and pasted this from my preferences and I only get cuda like this.
I turn of the (gpu) and turn on the Astropulse v5 and I only get AP's

{edit}
V10 For cuda and Ak_v8

---

Run only the selected applications SETI@home Enhanced: yes
Astropulse: no
Astropulse v5: no

If no work for selected applications is available, accept work from other applications? no
Use Graphics Processing Unit (GPU) if available yes
Use Central Processing Unit (CPU) no


Just cut and pasted this from my preferences and I only get cuda like this.
I turn of the (gpu) and turn on the Astropulse v5 and I only get AP's.

Do you use V10 pack or just CUDA MB part ?

@ PaulDHarris

My recommendation was for usage with the team mod.

Like Raistmer said, he's the developer of this CUDA app (and whole team mod)..
With his team mod you must enable CPU usage also.

Or wait to the answer of FiveHamlet.. ;-)


Sorry, I didn't know that. I use/d only the stand alone CUDA app of Raistmer.
And made my test with BOINC V6.6.20 and selfmade app_info.xml and also stand alone CUDA app of Raistmer.

I didn't used his complete team mod.


The first (daily) graph show the last ~ 42 hours.
The second (weekly) graph show the last ~ 10 days.

If the green graph is at ~ 6 - 7 k, then the Berkeley server traffic is at 'normally'.


@ Raistmer

Hmm..?

So I need for stand alone now:
last CUDA.exe (app)
and
CUDA V2.2 (cudart.dll + cufft.dll)
and
driver 185.85 .

That's all? Or something more?

Isn't it a beta driver? It's a final?


But, where is the libfftw3f-3-1-1a_upx.dll ? *confused*
Or I could take also the stock Berkeley .dll ?



EDIT:
Or I must take the libfftw3f-3-1-1a_upx.dll out of a team package from the opening post of the thread?

---

Update to CUDA 2.2 runtime and video driver 185.85 recommended.
See this post:
http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg17425.html#msg17425

Thanks for mentioning the new update! :)

---

me@rescam.org

[color=blue]
So I need for stand alone now:
last CUDA.exe (app)
and
CUDA V2.2 (cudart.dll + cufft.dll)
and
driver 185.85 .

yes

Isn't it a beta driver? It's a final?

It's published for free download on nVidia site already.
CUDA runtime speed was anhanced slightly (for better results current CUDA MB code should use new CUDA features) and CUFFT, FFT library that runs on GPU, seems already use some of these new features cause its performance increased considerable. MultiBeam does many FFR transforms so CUDA MB gains advantage of any FFT library speedup.

But, where is the libfftw3f-3-1-1a_upx.dll ? *confused*
Or I could take also the stock Berkeley .dll ?

My last builds don't require this DLL at all. You can leave existed one for safety (stock app still requires it).

@ Raistmer

We from team seti.international like to support SETI@home. ;-D


@ PaulDHarris

If you use Raistmer's CUDA app, then errors are wanted for VLAR-WUs.

BTW.
With this app_info.xml thread.. you can use Raistmer's CUDA app also.
Only the .._CUDA_...exe and the three .dll's.

Have a look here:
http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg16715.html#msg16715

@Sutaru Tsureku
Thanks I made changes to my app_info.xml file see below and added the 3 dll files and MB_6.08_mod_CUDA_V10.exe file and removed the setiathome_6.08_windows_intelx86__cuda.exe file.

<app_info>
<app>
<name>astropulse</name>
</app>
<file_info>
<name>ap_5.00r103_SSE3.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>astropulse</app_name>
<version_num>500</version_num>
<flops>6046682454.38398725</flops>
<file_ref>
<file_name>ap_5.00r103_SSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>astropulse_v5</name>
</app>
<file_info>
<name>ap_5.03r112_SSE3.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>astropulse_v5</app_name>
<version_num>503</version_num>
<flops>6073556598.62569386</flops>
<file_ref>
<file_name>ap_5.03r112_SSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>AK_v8_win_SSSE3x.exe</name>
<executable/>
</file_info>
<file_info>
<name>setiathome_6.08_windows_intelx86__cuda.exe</name> <- This Line Here
<executable/>
</file_info>
<file_info>
<name>cudart.dll</name>
<executable/>
</file_info>
<file_info>
<name>cufft.dll</name>
<executable/>
</file_info>
<file_info>
<name>libfftw3f-3-1-1a_upx.dll</name>
<executable/>
</file_info>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>603</version_num>
<platform>windows_intelx86</platform>
<flops>4702975242.29865675</flops>
<file_ref>
<file_name>AK_v8_win_SSSE3x.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>608</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>0.127970</avg_ncpus>
<max_ncpus>0.127970</max_ncpus>
<flops>3200000000</flops>
<plan_class>cuda</plan_class>
<file_ref>
<file_name>MB_6.08_mod_CUDA_V10.exe</file_name>
<main_program />
</file_ref>
<file_ref>
<file_name>cudart.dll</file_name>
</file_ref>
<file_ref>
<file_name>cufft.dll</file_name>
</file_ref>
<file_ref>
<file_name>libfftw3f-3-1-1a_upx.dll</file_name>
</file_ref>
<coproc>
<type>CUDA</type>
<count>1</count>
</coproc>
</app_version>
</app_info>


Your App_info.xml is pointing at two different .exe files and will not crunch Cuda w/u's properly.
Cuda Workunits will be handed to the CPU's if not corrected.

The red indicated line must be edited to <name>MB_6.08_mod_CUDA_V10.exe</name> in order to use the V-10 cuda mod.

After this is changed stop boinc completely and restart. Should crunch in Cuda there after, so long as the workunits are ver. 6.08.

Hang in there Paul! :)

Edit: And as long as the <file_name> line is correct, mine points at a different exe file than you have listed. This maybe an issue as well.
If all else fails I'll piece a App_info.xml together for you.

@Stefan Ver3

I used your app_info.xml file with the correction you suggested and it downloaded 2 CUDA WUs and then it computing error them. Well at least it downloaded some CUDA if I can just get it to stop the computing error it would be all right.

Thanks you Stefan Ver3


@Sutaru Tsureku

I did all the things you suggested and with 2 AP WUs all most at completion I did the BONIC manager update and I got a bunch of MB 608 WUs none of them CUDA and about 1/2 of them computing error out. So I deleted all of Raistmers files and reinstalled his Raistmer's_opt_pack_V10a_x86_SSSE3_MB_CPU_GPU_team.rar and made use of Stefan's app_inf0.xml file with his suggested correction and the BOINC manager aborted all the MB 608 WUs and I updated the BOINC manager and it started to download and run CUDA WUs which later computing error them.

Thanks for your help.

PS
I ste my seti@home preferences to

Run only the selected applications SETI@home Enhanced: yes
Astropulse: yes
Astropulse v5: yes

If no work for selected applications is available, accept work from other applications? yes
Use Graphics Processing Unit (GPU) if available yes
Use Central Processing Unit (CPU) yes

The old settings could have caused to CUDA WUs to computing erroe.

Greetings!

I have downloaded the AK_V8 files and am intending on installing it. Is this the correct ver for this 4720228 machine? It has ssse3 avail;able. I also have 2 cuda devices and need to know the best cuda app to use. The folling is the app_info.xml I will use but I suspect it will need changes to use cuda. Would someone be so kind as to show me the mods needed to it for the correct cuda app.

app_info>
<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>AK_v8_win_x64_SSSE3x.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>528</version_num>
<file_ref>
<file_name>AK_v8_win_x64_SSSE3x.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>603</version_num>
<file_ref>
<file_name>AK_v8_win_x64_SSSE3x.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>astropulse</name>
</app>
<file_info>
<name>ap_5.00r103_SSE3.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>astropulse</app_name>
<version_num>500</version_num>
<file_ref>
<file_name>ap_5.00r103_SSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>astropulse_v5</name>
</app>
<file_info>
<name>ap_5.03r112_SSE3.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>astropulse_v5</app_name>
<version_num>503</version_num>
<file_ref>
<file_name>ap_5.03r112_SSE3.exe</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>


Thanks in advance

---

And after the spanking comes...



Ni !

Greetings!

I have downloaded the AK_V8 files and am intending on installing it. Is this the correct ver for this 4720228 machine? It has ssse3 avail;able. I also have 2 cuda devices and need to know the best cuda app to use. The folling is the app_info.xml I will use but I suspect it will need changes to use cuda. Would someone be so kind as to show me the mods needed to it for the correct cuda app.

...

Thanks in advance

My app_info should work perfectly on your set-up - runs my Q9450 + GTX295 under 64bit Vista. The ony changes necessary should be to the <flops> values which should be adjusted to match your own hardware as described in this thread.

<app_info>
    <app>
        <name>astropulse</name>
    </app>
    <file_info>
        <name>ap_5.00r103_SSE3.exe</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>astropulse</app_name>
        <version_num>500</version_num>
        <platform>windows_x86_64</platform>
        <avg_ncpus>1.000000</avg_ncpus>
        <max_ncpus>1.000000</max_ncpus>
        <flops>8165936127</flops>
        <file_ref>
            <file_name>ap_5.00r103_SSE3.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>
     <app>
        <name>astropulse_v5</name>
    </app>
    <file_info>
        <name>ap_5.03r112_SSE3.exe</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>astropulse_v5</app_name>
        <version_num>503</version_num>
        <platform>windows_x86_64</platform>
        <avg_ncpus>1.000000</avg_ncpus>
        <max_ncpus>1.000000</max_ncpus>
        <flops>9436192858</flops>
        <file_ref>
            <file_name>ap_5.03r112_SSE3.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>
     <app>
        <name>setiathome_enhanced</name>
    </app>
        <file_info>
            <name>AK_v8_win_x64_SSSE3x.exe</name>
            <executable/>
        </file_info>
        <file_info>
            <name>setiathome_6.08_windows_intelx86__cuda.exe</name>
            <executable/>
        </file_info>
        <file_info>
            <name>cudart.dll</name>
            <executable/>
        </file_info>
        <file_info>
            <name>cufft.dll</name>
            <executable/>
        </file_info>
        <file_info>
            <name>libfftw3f-3-1-1a_upx.dll</name>
            <executable/>
        </file_info>
     <app_version>
        <app_name>setiathome_enhanced</app_name>
        <version_num>603</version_num>
        <platform>windows_x86_64</platform>
        <avg_ncpus>1.000000</avg_ncpus>
        <max_ncpus>1.000000</max_ncpus>
        <flops>6351283655</flops>
        <file_ref>
            <file_name>AK_v8_win_x64_SSSE3x.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>  
    <app_version>
        <app_name>setiathome_enhanced</app_name>
        <version_num>608</version_num>
        <platform>windows_x86_64</platform>
        <avg_ncpus>0.150000</avg_ncpus>
        <max_ncpus>0.150000</max_ncpus>
        <flops>21200000000</flops>
        <plan_class>cuda</plan_class>
        <file_ref>
            <file_name>setiathome_6.08_windows_intelx86__cuda.exe</file_name>
            <main_program/>
        </file_ref>
        <file_ref>
            <file_name>cudart.dll</file_name>
        </file_ref>
        <file_ref>
            <file_name>cufft.dll</file_name>
        </file_ref>
        <file_ref>
            <file_name>libfftw3f-3-1-1a_upx.dll</file_name>
        </file_ref>
        <coproc>
            <type>CUDA</type>
            <count>1</count>
        </coproc>
    </app_version>
</app_info>

F.

---

Worked perfectly. Couple quaestions though. Firstly, isn't there a optimized cuda app that's better to use and secondly, is the flops entry really neccesary ?

---

And after the spanking comes...



Ni !

Worked perfectly. Couple quaestions though. Firstly, isn't there a optimized cuda app that's better to use and secondly, is the flops entry really neccesary ?

At the present time there is no Optimised CUDA App. Raistmer has done some tweaks (e.g. cutting the time the CPU takes to load the GPU at the start of a WU, killing VLAR's with an error condition) but the basic crunch algorithm is the same I believe.
The flops entry is not essential; it helps smooth the wilder variations in Estimate to Complete between the different types of WU and can smooth the work-fetch behaviour (i.e. machine doesn't go into EDF quite so often).
Glad it worked OK.

F.

---

There is "slightly" optimized version in testing on Lunatics dev boards and further work going on, but it's subject of another thread.

"team" packs (V10 included) became obsolete now. To simplify life for users who can't read whole this thread to stay tuned with latest changes (not too latest already though, BOINC 6.6.20 exists as recommended version more than month already AFAIK) I ask moderators to lock this thread.
All my own support of "teamed" packs now ceased (until I get some example why "teamed" pack should be used instead of BOINC 6.6.20 native GPU scheduling at least)

"teamed" packs were temporary solution and well served their aim during lifespan. But now this period is over.

What should be used instead:
1) BOINC 6.6.20
2) latest CUDA MB build (can be found in this thread and will be posted in new thread devoted optimized CUDA MultiBeam app).
3) appropriate app_info.xml
some examples can be found in this thread: http://setiathome.berkeley.edu/forum_thread.php?id=52589

2) and 3) could be replaced by using Jason's Lunatics installer (currently in beta stage, can be found in beta area of this forum: http://lunatics.kwsn.net/installer-testing/index.0.html )

What should not be used
1) ncpus field in cc_config.xml
2) app_info.xml supplied with obsolete "team" packs.
3) "teamed" modification of AK_v8 (was part of "team" packs).