..

---

News about SETI opt app releases: https://twitter.com/Raistmer

I will also check the box from my 630 to see if it tells me what flavor it is.

Very unusually for me, I don't seem to have the box. Oh well.

If you have GPU-Z, it should show you which one it is.

I have CPU-Z and it doesn't tell me. All it says is

Name: NVIDIA GeForce GT 630
Revision: A1
Memory: Size: 1024 MB

All other boxes are blank and grayed out.

---

David
signature sent back to alpha testing

The app_config that's quoted has an incomplete opening tag. It should be <app_config>. Perhaps that's your problem?

Yup, that was it!

Even I should have caught that. Thanks.

---

David
signature sent back to alpha testing

I will also check the box from my 630 to see if it tells me what flavor it is.

Very unusually for me, I don't seem to have the box. Oh well.

If you have GPU-Z, it should show you which one it is.

I have CPU-Z and it doesn't tell me. All it says is

Name: NVIDIA GeForce GT 630
Revision: A1
Memory: Size: 1024 MB

All other boxes are blank and grayed out.

GPU-Z shows much more GPU-specific detail than CPU-Z.

I will also check the box from my 630 to see if it tells me what flavor it is.

Very unusually for me, I don't seem to have the box. Oh well.

If you have GPU-Z, it should show you which one it is.

I have CPU-Z and it doesn't tell me. All it says is

Name: NVIDIA GeForce GT 630
Revision: A1
Memory: Size: 1024 MB

All other boxes are blank and grayed out.

GPU-Z shows much more GPU-specific detail than CPU-Z.

Where do I get it? I don't see it on cpuid.com and gpuid.com took me to some other placeholder site.

---

David
signature sent back to alpha testing

GPU-Z shows much more GPU-specific detail than CPU-Z.

Where do I get it? I don't see it on cpuid.com and gpuid.com took me to some other placeholder site.

https://www.techpowerup.com/gpuz/

The app_config that's quoted has an incomplete opening tag. It should be <app_config>. Perhaps that's your problem?

Yup, that was it!

Even I should have caught that. Thanks.

good to know ... maybe its time for me to do some more work ... sigh ...
Ed F

---

Could any NVidia users - especially those using older, pre-350 drivers - please check their caches and verify that they're starting to receive v8.01 cudaxx tasks?

If you have any of them, please could you run some through quickly, please?

I just attached my laptop to beta and got one cuda32 and a heap of cuda42 tasks.

This laptop uses driver 347.52. It has a Quadro K2000M board that normally takes about two hours on a cuda task.

The first cuda42 task has run for seven minutes and already claims to have done 25 % of the task. The cuda32 task would appear to be the next in line for execution. So we'll see...

Edit: Uh.. my two hour estimate was for AP work. MB V8 should run much faster.

It turns out this is currently not possible under BOINC. The scheduler first chooses the fastest app_version for a GPU, then discards any work that isn't feasible. So for a host that thinks the OpenCL app is fastest, it would choose the OpenCL app and then discard any no-VLAR results which would probably result in it getting no work much of the time. It would never use the CUDA app. There's no obvious way to put the apps on a level playing field without rewriting the scheduler. So we're stuck with CUDA alone or OpenCL alone on any given platform.

It seems we have a three-way problem, then.

1) Re-write the scheduler to allow project scientists to do science, the way they wish.
2) Re-write the CUDA app so it can process VLARs at the same speed as other ARs.
3) Re-write the OpenCL app so it doesn't demand so much CPU time.

I propose we club together and rent a desert island, and send all three developers there to work out a solution. First developer to solve their problem gets an air ticket home, second solution a boat ticket, third has to swim.

Cheating - i.e. collaboration - is allowed without restriction. If they declare a dead heat, they all win an air ticket home - but a return ticket straight back out if their solution doesn't work.

Do I have a seconder?

Does this island have beer ?

---

Chaos: When the present determines the future, but the approximate present does not approximately determine the future.
Edward Lorenz

Does this island have beer ?

Yes, I think we can stretch to that. Pizza, even.

Back to the serious business at hand, for a moment at least.

We have a release of the GPU apps at Main. Many thanks to those who helped with that quick final cuda check - you can let your tasks run in the normal order now.

Do I have a seconder?

---

David
signature sent back to alpha testing

Seconded ;-)

Definitely seconded.

Funding would come from revenue generated by the associated reality TV show.

---

I'm running the same Intel GPU app that is being tested here, under Anonymous Platform at the Main project.

Application details for host 7118033
(scroll to the very bottom for v8 on Intel GPU)

Currently showing: 220 consecutive valid tasks, and no errors.

From what I've been able to see, the application is fine, but it may have another of these unresolved driver dependencies. I'm running an HD 4600 with driver 10.18.10.3621 (Release Date: May 21, 2014): that has proved to be the gold standard driver since we first started running on these devices.

Richard, I've been running the Intel GPU apps here on Beta for some time, also on an HD 4600, without any problems, even though the driver is more recent - it's version 10.18.14.4264, dated 4 August 2015.

Currently, I have processed almost 350 tasks using the various Intel GPU app versions that have been released here for testing, without a single error or invalid task.

---

Thanks John.

Good to know.

---

With each crime and every kindness we birth our future.

@David,
To run two GPU tasks put this in app_config.xml in your setiweb.ssl.berkeley.edu_beta folder.

<app_config>

<app>
<name>setiathome_v8</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.4</cpu_usage>
</gpu_versions>
</app>

</app_config>

Just out of curiosity, what would happen if I took the cpu_usage line out of this? What I'm wondering is whether it really needs that much of a CPU to support the GPU. On the other hand, I suppose it might try to use more than .4 and bog down other CPU functions.

Any reason I shouldn't just try it and see?

---

David
signature sent back to alpha testing

@David,
To run two GPU tasks put this in app_config.xml in your setiweb.ssl.berkeley.edu_beta folder.

<app_config>

<app>
<name>setiathome_v8</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.4</cpu_usage>
</gpu_versions>
</app>

</app_config>

Just out of curiosity, what would happen if I took the cpu_usage line out of this? What I'm wondering is whether it really needs that much of a CPU to support the GPU. On the other hand, I suppose it might try to use more than .4 and bog down other CPU functions.

Any reason I shouldn't just try it and see?

BOINC uses that value only to reserve cores - two up, 0.8 - <1 no core - two cards two up 4 total 1.6 >1 - 1 core reserved. You can use it to tweak how many cores boinc reserves depending on what you run and how many. Besides that, iirc it doesn't do anything.

---

A person who won't read has no advantage over one who can't read. (Mark Twain)

I believe you are right there.

With 1 GPU (2 tasks running), these should be the same:

<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.4</cpu_usage>

<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.1</cpu_usage>

If you wanted to free a CPU core:

<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>