so run the lunatic 64bit installer and then add the above to the app_info.xml file, leaving all the old 6.08 v12 stuff, what about the platform stuff?

I honestly don't know - I don't have a 64-bit OS here to test with.

My best suggestion - run the lunatics 64-bit installer as you suggest, including for the time being the Cuda23 application - but don't let it actually run on a Fermi, you won't get valid results.

Then, inspect the app_info.xml file the installer has created for you, and adjust it to match this sample - change the executable name, and the two cuda DLL names, wherever they occur. Leave the platform tag(s) as they were made by the installer. Edit: Remember to download the new files as well!

Jason is working on a new installer, to incorporate a genuinely optimised (though not yet optimal) Fermi application - hopefully the apps can go to beta testing in the next couple of days (if the project is up), and the installer will follow after that.

... leaving all the old 6.08 v12 stuff ...

No, don't do that. The 6.08 codebase won't run on a Fermi.

I tried, and got Beta task 7964564:

Cuda error 'cudaMemcpyToSymbol(cudaAcc_GaussFit_settings, (void*) &settings, sizeof(settings))' 
in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_gaussfit.cu' in line 454 : invalid device symbol.

Is this correct? it seems to be working... i took out the v6.08 stuff

<app_info> 
    <app>
        <name>setiathome_enhanced</name>
    </app>
    <file_info>
        <name>AK_v8b_win_x64_SSSE3x.exe</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>setiathome_enhanced</app_name>
        <version_num>603</version_num>
	<platform>windows_intelx86</platform>
        <file_ref>
           <file_name>AK_v8b_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>
	<platform>windows_x86_64</platform>
        <file_ref>
           <file_name>AK_v8b_win_x64_SSSE3x.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>
    <app>
        <name>astropulse_v505</name>
    </app>
    <file_info>
        <name>ap_5.05r409_SSE.exe</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>astropulse_v505</app_name>
        <version_num>505</version_num>
	<platform>windows_intelx86</platform>
        <file_ref>
            <file_name>ap_5.05r409_SSE.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>    
    <app_version>
        <app_name>astropulse_v505</app_name>
        <version_num>505</version_num>
	<platform>windows_x86_64</platform>
        <file_ref>
            <file_name>ap_5.05r409_SSE.exe</file_name>
            <main_program/>
        </file_ref>
    </app_version>
    <app>
        <name>setiathome_enhanced</name>
    </app>
    <file_info>
        <name>libfftw3f-3-1-1a_upx.dll</name>
        <executable/>
    </file_info>
    <file_info>
        <name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</name>
        <executable/>
    </file_info>
    <file_info>
        <name>cudart32_30_14.dll</name>
        <executable/>
    </file_info>
    <file_info>
        <name>cufft32_30_14.dll</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>setiathome_enhanced</app_name>
        <version_num>610</version_num>
        <platform>windows_intelx86</platform>
        <avg_ncpus>0.200000</avg_ncpus>
        <max_ncpus>0.200000</max_ncpus>
        <plan_class>cuda_fermi</plan_class>
        <file_ref>
            <file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name>
            <main_program/>
        </file_ref>
        <file_ref>
            <file_name>cudart32_30_14.dll</file_name>
        </file_ref>
        <file_ref>
            <file_name>cufft32_30_14.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_version>
        <app_name>setiathome_enhanced</app_name>
        <version_num>610</version_num>
        <platform>windows_x86_64</platform>
        <avg_ncpus>0.200000</avg_ncpus>
        <max_ncpus>0.200000</max_ncpus>
        <plan_class>cuda_fermi</plan_class>
        <file_ref>
            <file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name>
            <main_program/>
        </file_ref>
        <file_ref>
            <file_name>cudart32_30_14.dll</file_name>
        </file_ref>
        <file_ref>
            <file_name>cufft32_30_14.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> 

---

RottenMutt, Richard said this...

... leaving all the old 6.08 v12 stuff ...

No, don't do that. The 6.08 codebase won't run on a Fermi.

I tried, and got Beta task 7964564:

Cuda error 'cudaMemcpyToSymbol(cudaAcc_GaussFit_settings, (void*) &settings, sizeof(settings))' 
in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_gaussfit.cu' in line 454 : invalid device symbol.

As for not reporting or requesting tasks, do you have a bunch of uploads waiting? This could keep you from downloading and you've probably already reported everything that had uploaded prior to this mess we are in now.

---

PROUD MEMBER OF Team Starfire World BOINC

i rmvd the 6.08 stuff.

all my completion time estimates are way off, 11.5 hours for a fermi wu, 13.5 for a MB enhanced 6.03, 150 hours for a astropulse...

---

i rmvd the 6.08 stuff.

all my completion time estimates are way off, 11.5 hours for a fermi wu, 13.5 for a MB enhanced 6.03, 150 hours for a astropulse...

I assume you haven't put in the Flops calculations yet. Message 1002997, this thread.

I have seen a few people stating that they are running 2 tasks on their gtx470&480's - why? When I look at gpu usage, I typically see 69% usage peaking in the 90's for short bursts. Is there any benefit to doing this, would it actually work out any quicker as Im thinking that it might crunch 2 units but would do so at half the speed? Am I right?

1 WU on my GTX 470 took 12 minutes average to process.
2 WU's on my GTX 470 take 16 minutes average to process.
So my output is much higher with 2 tasks on this card.

On my 9800 GTX it was actually a bit slower to run 2 tasks.

I would have liked to know if all that were only for the gtx 4xx? Because I have only a gtx 260? thank you

---

Alliance Francophone

Yes, most of this was triggered by the new 4xx series Fermi's. Many people ran out and bought them as soon as they hit the market and set them to work on SETI. The problem arose when they tried to do regular GPU tasks with them which don't run on the Fermi cards. They immediately started giving -9 overflow results. When paired with a regular CPU or GPU giving good results they would cause a resend and often, if the third wingman had a 4xx card would validate the work as -9 overflow and discard the good result.

---

PROUD MEMBER OF Team Starfire World BOINC

Flops estimating for Fermi

In the middle of all the error messages last night, somebody asked about getting the running time estimates for the various applications into balance. This was my initial background explanation:

My figures ended up at:

CPU:	    3,826,762,478
GPU:	   57,462,450,464

or, in multiplier terms,

CPU: multiply <p_fpops> by 1.45
GPU: divide peak GFlops by 18.95

That's a rather narrower spread than we're used to with the earlier cards, which I think is perhaps accounted for by:

Ok, so I have had a go at this, can someone confirm my app_info is correct as Im not sure I have put the flops bit in the right place? The fermi section of my app_info is posted below...

<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</name>
<executable/>
<flops>57462450464</flops>
</file_info>
<file_info>
<name>cudart32_30_14.dll</name>
<executable/>
</file_info>
<file_info>
<name>cufft32_30_14.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>610</version_num>
<avg_ncpus>0.200000</avg_ncpus>
<max_ncpus>0.200000</max_ncpus>
<flops>57462450464</flops>
<plan_class>cuda_fermi</plan_class>
<file_ref>
<file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>cudart32_30_14.dll</file_name>
</file_ref>
<file_ref>
<file_name>cufft32_30_14.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>
<file_info>
<name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</name>
<executable/>
</file_info>
<file_info>
<name>cudart32_30_14.dll</name>
<executable/>
</file_info>
<file_info>
<name>cufft32_30_14.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>610</version_num>
<avg_ncpus>0.200000</avg_ncpus>
<max_ncpus>0.200000</max_ncpus>
<plan_class>cuda_fermi</plan_class>
<file_ref>
<file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>cudart32_30_14.dll</file_name>
</file_ref>
<file_ref>
<file_name>cufft32_30_14.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>

---

MadMac, try this Fermi section:

<app>
<name>setiathome_enhanced</name>
</app>
<file_info> 
<name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</name> 
<executable/> 
</file_info> 
<file_info> 
<name>cudart32_30_14.dll</name> 
<executable/> 
</file_info> 
<file_info> 
<name>cufft32_30_14.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>610</version_num> 
<avg_ncpus>0.200000</avg_ncpus> 
<max_ncpus>0.200000</max_ncpus>
<platform>windows_intelx86</platform>
<flops>57462450464</flops> 
<plan_class>cuda_fermi</plan_class> 
<file_ref> 
<file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name> 
<main_program/> 
</file_ref> 
<file_ref> 
<file_name>cudart32_30_14.dll</file_name> 
</file_ref> 
<file_ref> 
<file_name>cufft32_30_14.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_version> 
<app_name>setiathome_enhanced</app_name> 
<version_num>610</version_num> 
<avg_ncpus>0.200000</avg_ncpus> 
<max_ncpus>0.200000</max_ncpus>
<platform>windows_x86_64</platform>
<flops>57462450464</flops> 
<plan_class>cuda_fermi</plan_class> 
<file_ref> 
<file_name>setiathome_6.10_windows_intelx86__cuda_fermi.exe</file_name> 
<main_program/> 
</file_ref> 
<file_ref> 
<file_name>cudart32_30_14.dll</file_name> 
</file_ref> 
<file_ref> 
<file_name>cufft32_30_14.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>

Claggy

Thank you Claggy - appreciated..

---

I got work for my CPU but still nothing for the Fermi...
Still the same message...

14-6-2010 22:56:25 SETI@home Message from server: No work sent
14-6-2010 22:56:25 SETI@home Message from server: Your app_info.xml file doesn't have a usable version of SETI@home Enhanced.
14-6-2010 22:56:25 SETI@home Message from server: (reached daily quota of 100 tasks)

I even made a backup, detached, attached and still got nothing because I reached the "daily quota of 100 tasks". I only got about 20 today so where are the other 80 WU's?...

Is there any hope I'll get work for my Fermi (soon)?
It worked perfect for 1 day (with the same app_info.xml)... And now it's just sitting there... Waiting...

I installed Boinc 2 weeks ago to give it a try (again) but for some reason Seti always crashes when I try to join after a period of being inactive...

Okay, I'm a bit frustrated because I can't test my new toy...
Hope you forgive me for that :P

I think I have returned my 1st good Gpu 470 result Validate state is Initial as I write.

---

I think I have returned my 1st good Gpu 470 result Validate state is Initial as I write.

setiathome_CUDA: CUDA Device 1 specified, checking...
Device 1: GeForce GTX 470 is okay
SETI@home using CUDA accelerated device GeForce GTX 470
V12 modification by Raistmer
Priority of worker thread rised successfully
Priority of process adjusted successfully
Total GPU memory 1309081600 free GPU memory 1238081536
setiathome_enhanced 6.02 Visual Studio/Microsoft C++

No, that version doesn't work on Fermi - see the eponymous thread. You need the stock application, as described in post #2 of this thread.

I think I have returned my 1st good Gpu 470 result Validate state is Initial as I write.

Speedy, you can't run the V12 app with Fermi's, it doesn't produce valid results,

Claggy

Which is 6.10 currently, On Fermi I think 6.08 and 6.09 are not Fermi compatible either.

Correct. 6.08 produces computation errors, 6.09 produces the -9 overflow on most tasks.

No, that version doesn't work on Fermi - see the eponymous thread. You need the stock application, as described in post #2 of this thread.

My understanding of a stock application is that you do not need to change anything. I thought post #2 applied if you are using a non stock app. I'm using stock cpu app.

---

No, that version doesn't work on Fermi - see the eponymous thread. You need the stock application, as described in post #2 of this thread.

My understanding of a stock application is that you do not need to change anything. I thought post #2 applied if you are using a non stock app. I'm using stock cpu app.

No you're Not running the stock CPU app, you're running an app_info with the AK_V8 SSSE3x CPU app and the V12 Cuda app,

Stock means you're running the bog standard apps from the project, with or without an app_info,

Claggy

No, that version doesn't work on Fermi - see the eponymous thread. You need the stock application, as described in post #2 of this thread.

My understanding of a stock application is that you do not need to change anything. I thought post #2 applied if you are using a non stock app. I'm using stock cpu app.

At the time it was written, there was no stock Fermi app here on the main SETI project. In order to run a true Fermi app, you had to use an app_info.xml file, and you had to add that section.

Subsequently, it became apparent that people were installing Fermi cards regardless, and the project code (as then written) was wrongly sending them the cuda23 app - which doesn't work on a Fermi. That's what led to all the kerfuffle of the last ten days.

But neither of those paragraphs applies to you. The task you linked, saying that you thought it was your first good result, was crunched using Raistmer's v12 application. That isn't, and never was, a stock application.

There now is a stock application, and you're free to use that, either using BOINC's automatic distribution system, or by manually installing it into app_info. But please don't use v12.