Latest drivers (NVidia and ATI)

[Jason G]

Easiest would be to just try the test piece I made.  The original Boinc code works on my system since I have a display attached to it, but I'd expect your systems to show 'Unknown' for the Boinc method, and the correct version for the QuickFix2.

Attached nvApiTest, 7zip archived,with Example screenshot from mine (where the original Boinc method works).  Please extract the exe from the archive & run it on those systems where Boinc shows 'Unknown':

[benool]

There you go.

Seems to work! Congrats.

[Jason G]

Thanks for the verification.  Yeah it was a relatively trivial cosmetic, but oddly annoying, problem.  As mentioned the updated code was passed onto Berkeley, and I see no reason it shouldn't function the same way there when they get around to a build with the change. 

Jason

[Richard Haselgrove]

I've already sent a PM to Pepi, asking if he can pop back and perforn the same test (let's hope he has notifications going to the right email address).

If that works, I'll pass both reports back to Berkeley at the same time - no point going off half-cocked.

[Pepi]

Hi 
Works like charm

[Richard Haselgrove]

That's great, thanks.

The second line is just reproducing the way that BOINC does things, confirming the same error as you reported in this thread. Then the third line proves that Jason's correction of the bug works on your system too - which is what we wanted to check. I'll go and tell David the news.

[Pepi]

Ok, I am glad can help in any
Just now I am on Primegrid challenge, so  no GPU using only CPU

[Raistmer]

From  APP SDK 2.5:

OpenCL can now be used by a Windows service.

Anyone checked this alredy?

[_heinz]

Thanks for the verification.  Yeah it was a relatively trivial cosmetic, but oddly annoying, problem.  As mentioned the updated code was passed onto Berkeley, and I see no reason it shouldn't function the same way there when they get around to a build with the change. 

Jason

Hi Jason,
Any news from Berkeley ? Is the code already inserted into any boinc version ?
here a picture from nvApiTest6 on laptop GT540M

[Jason G]

I believe it's in the trunk code, but I don't keep track of when development releases are rolled out for testing, so couldn't tell you whether they made a build with it yet.  If you have a Boinc checkout, you could use the repository browser to see what revision number the latest tags are, then compare revision numbers & dates against the trunk logs. 

In any case, it's great that you all found the fix applies to other scenarios in addition to isle Optimus platforms.  Funnily enough it's the stupid cosmetic bugs like this I find the most annoying.  I guess because they are 'in your face' the most  .

Jason

[Richard Haselgrove]

The fix was applied to the trunk code at revision 24046.

The trunk code is currently being used to build BOINC v6.13.1 (will be BOINC 6.14.xx or BOINC v7 eventually - a long way from being ready yet)

For a working BOINC, you would need to build from the boinc_core_release_6_10 or boinc_core_release_6_12 branch: neither of those contain the fix yet. And there hasn't been a new build - even of the experimental v6.13 - since the fix was applied.

[Richard Haselgrove]

Could one of our resident ATI driver gurus have a look at http://boinc.berkeley.edu/dev/forum_thread.php?id=6846, please?

It's always been the case that NVidia drivers are inaccessible to BOINC installed in 'Protected Application Execution' (aka 'service') mode under Windows Vista and Windows 7.

The posters in that thread seem to be implying that ATI drivers up to and including 11.6 could be used in service mode, but 11.7 and onwards have moved towards the NVidia behaviour. Is that a change which the people who monitor such things can corroborate?

[Ghost0210]

Could one of our resident ATI driver gurus have a look at http://boinc.berkeley.edu/dev/forum_thread.php?id=6846, please?
........
The posters in that thread seem to be implying that ATI drivers up to and including 11.6 could be used in service mode, but 11.7 and onwards have moved towards the NVidia behaviour. Is that a change which the people who monitor such things can corroborate?

I'm running 11.5 drivers and Win 7, so when I get home I'll have a look at installing Boinc in service mode and see what happens

[Jason G]

NVidia behaviour

ROFL, you mean Microsoft WDDM + KHRonos Consortium behaviour....

We have a colourful expression in Australia, that is  'Pissing into the wind'.  Yes MS-DOS runs things faster....

[Mike]

I had no connection issues with 11.7 and 11.8.
But i dont use boinc as service.
Also i had SDK 2.5 dev installed in addition.