[–]VijayPande-FAHF@h Director 2 points 35 minutes ago 
We have been working with NVIDIA on the issue and for them to resolve it in a new driver release soon. We do not have an ETA from them on the release, but they do see the issue and I believe have an idea on the fix.
 
Was posted on the Folding reddit.
 
 

This is great as long as the fix dose not break it for GKxxx Cards.

bcavnaugh
This is great as long as the fix dose not break it for GKxxx Cards.

Lets hope not

Kepler and I think AMD.

new driver installed today

A year ago promised that Nvidia was working with F@H on a driver fix for OpenCL to address the locked CPU core running 100%.  Still waiting on that annoyance.

-ZS-Carpenter

mdk7777777
yeah, game beta only, no change to open cl needed for FOLDING.
 
Oh course, there was a thread on FF discussing this.
 
I posted that the beta driver did not include the possible fix described by Dr. Pande....just a FYI, this is not the driver you are looking for...and of course I was censored by that sort of a geek.
 
I guess they don't even want informed information on the FF. (totally on topic without any negative implication)
 
All I can say is WOW.

there is nothing to be gained by posting at FF. It is really a worthless forum for almost everything

 That's a shame really it could be so much more. I hate hearing how people are treated over there. Why I never registered, been that way for a long time. On a brighter note hope the fix comes soon!

Just a friendly reminder everyone that swearing/cursing is not allowed. Such posts will be edited or deleted.

I got my first batch of Core 18 Projects on my GTX 680 Classified Rig and at 12% ZERO PPD. HFM My HFM is up to date.
http://weather.mfc-cs.com/haf/Foldit-3-EVGA-GTX%20680%20Class%20Slot%2000.html
http://weather.mfc-cs.com/haf/Foldit-3-EVGA-GTX%20680%20Class%20Slot%2001.html
http://weather.mfc-cs.com/haf/Foldit-3-EVGA-GTX%20680%20Class%20Slot%2002.html
6 Hour Run Time.
Running Current Driver 347.09
 
On the computer they show:
 
 Are they about right for Core 18?
 
 
Getting even lower at 20% and now 8 and a half Hours, WOW! Are Core 18 a re-release of Core 15?

 
 
Must be some kind of glitch.

 
Thanks

SwingOp3
A thread at Bit-Tech Forums explains it and provides the solution.
I'm not worthy of being able to add links to posts.
Not sure I can name other forums in posts - if not delete post.

You can post the link without the http://

How're Core 18 projects supposed to be performing on Fermi architecture? My trusty old GTX 480 used to crank along at ~40k PPD (stock clocks); now, with Core 18 projects, it's getting only ~20k PPD, even when over-volted (within reason, using EVGA PrecisionX) and OCed to 800 MHz clock (1600 MHz shaders). Is a decrease this significant to be expected?

! don't know if it has anything to do with the bombardment of Core18 projects
!'ve been getting in or if it's just user error, but boy oh boy did ! get screwed this morning!! :[
(the ratio of core18's to core17's is @ least 5 or 6/1)
 
Up most of the night folding among other things, from about 8/9pm to 6am & for
the entire time my GPU's were operating at a stable and consistent 67*celsius
(air cooled) without issue. Come 6am the sun is rising and ambient temps can go
nowhere but up so i decide to close all the windows & blockout curtains in the house
(to retain all the cool night air & maintain tolerable temps inside during the day)
pause all folding on my main rig, shutdown the system & get some sleep.
 
! get up @ 10:30am & say to myself:
"Self, you may as well finish those 2 gpu projects before you go buy some groceries. (since their so close to completion - 55mins to go on a core17 project & just under 2hrs on a core18 project)"
 
So ! fire up my main rig, wait for all startup processes to take place, open PrecisionX 16 set my cooling, open F@H control & start folding (just as i always do). Barely 20 minutes in:
 
Black screen......display recovers in no more than 3secs. Upon recovery ! get:
Message - Display driver kernel error 347.whatever....   something along those lines.
(y'all picking up what !'m putting out?)
 
And to my complete amazement & infuriation both nearly completed GPU projects are GONE!
GONE! GONE! GONE! ......and downloading in their place two core18 projects!
 
ww-w-what?? What the hell man!
! felt like ! had just been robbed and not of the 98,000+ points, but of my time and money.
And @ the time of typing this post it is literally to hot to do any folding on the GPU's.
(ambient temp 34*c 46% Humidity (feels like 40+) night time temps didn't come down enough to cool the house properly)
 
No new hardware, no change in procedure or operation since the day ! first started folding.
! don't know if it's user error or what (maybe ! shouldn't be folding till the rig is water cooled?),
but if anyone has any ideas.....
 
UPDATE: Powered through & completed the 2 core18 WU's that bullied their way onto my system w/out issue.
              Then got another 2 core18 WU's. Completed those w/out issue.
              Got another 2 core18 WU's. Completed w/out issue.
UPDATE: Got 1 core18 and 1 core17. Completed (possibly w/an issue, but no loss of wu's)
 
              7 in a row. Lucky 7.

D3V!rum3tM^ch!n^
Black screen......display recovers in no more than 3secs. Upon recovery ! get:
Message - Display driver kernel error 347.whatever....   something along those lines.
(y'all picking up what !'m putting out?)

I was having this same issue on my two 780ti Classifieds. When they would both pick up an 18, the driver would crash about 5% in and dump the units. So I deleted one of the GPU slots to have just one of them running and that took care of the driver crashing. Now I am noticing that if I am doing anything on my comp, even something as simple as web browsing, it will crash any 18 unit it picks up.