So, it seems that there is a bug in the compiler/hardware which appears only on pre G200 cards.
We found a way to avoid it for now, but it limits what we can do, so it is not a solution.

gdf

So, it seems that there is a bug in the compiler/hardware which appears only on pre G200 cards.
We found a way to avoid it for now, but it limits what we can do, so it is not a solution.

gdf

Well at least it isn't a mystery any more. When you are on the bleeding edge, one should expect some cuts.

Hope it gets resolved in the not too distant future.

Rob

So, it seems that there is a bug in the compiler/hardware which appears only on pre G200 cards.
We found a way to avoid it for now, but it limits what we can do, so it is not a solution.

gdf

How come the G200 based cards also get failures?

Will there be an updated app for the non-G200 machines, or perhaps all machines? Will this be a cuda 2.2 app or stick with the old version for the time being?

Can we use the 185.85 drivers now or with the new app (assuming there will be one)?

---

BOINC blog

I am running an 8800GT and 3 9800Gtx+ cards. I have had zero complete Kashif Wu's .
Could you make a way for me to "opt out" of those type of units?

I still get occasional errors with other units but the Kashif are 100% failure rates for me. (3 different machines)

I am really unable to babysit my machines as I am away from home for days at at time.....

we are running this set of workunits called

x-GIANNI_newFB-...

If they go on ok, then we have isolated the problem with G90 chips. It is not solved yet but still at least we would know where to look.

gdf

These hang on my Pent D 830 with a 9600GSO. See here for a hung result that was aborted after more than 24 hours of no progress (hung at about 21%).

Just as a data point of reference, I have had 100% success on all work units using a GTX 260 Core 216 card, running CUDA 2.2 and 185.85 driver, even on work units that have had failures previously.

There is another way .....

Tap your well heeled benefactor you have tucked away for a mere $400,000 worth of vouchers to upgrade Crunchers pre-200's to 300GTXs - a snip at the price....

And ..... added Value!! ..... You'd also solve chruncher recruitment for while, they'd queue round to the next street, let alone next Block, for that one .....

*sigh* ........ aways nice to dream once in a while :)

Regards
Zy

There is another way .....

Tap your well heeled benefactor....

Regards
Zy

Believe me, I am tapping my heels that my card continues to function as well as it has! I just bought it, so it would be a big disappointment if there were issues so soon....but the issue fix would appear to be possible without a card upgrade, hopefully....(although NVIDIA I would guess is tapping its heels that many will upgrade...ouch! ;)

How come the G200 based cards also get failures?

Let's see what the fix can do and who still gets failures afterwards. Mind you, there's also the "regular failure rate", some kind of "noise floor" which affects all cards.

Will there be an updated app for the non-G200 machines, or perhaps all machines? Will this be a cuda 2.2 app or stick with the old version for the time being?
Can we use the 185.85 drivers now or with the new app (assuming there will be one)?

Not speaking officially, but I wouldn't rush to introduce another variable in the current situation. Wait until the dust settles and we're confident that the problems have been solved. 185.66 has been running fine for me with non-troublemaker WUs, so I'll keep using it until I see problems.

I do have a WU issued today and it appears to use client 6.64, so it may look like no new app for now. But this could be tied to an old type of WU as well.

MrS

---

Scanning for our furry friends since Jan 2002

Just had a Kashif go bang

ERROR: c:\cygwin\home\speechserver\gpumd2\src\pme\CPME_cufft.cu, line 104: cufftExecC2R (gridcalc3)

http://www.gpugrid.net/result.php?resultid=699822

I have had a problem on that PC re Office, and got it back online 5 hours ago. However I dont think it was that issue, I think it looks like the old WU problem surfacing - maybe in one of the older WUs still in the system??

Regards
Zy

Looks like the old problem and the WU was created past 20 May 16:44 CEST, when the fix was applied. I think it would be better to post such observations in the new thread, so they don't get lost.

MrS

---

Scanning for our furry friends since Jan 2002