I upgraded to 190.38 a few days ago when advised it was best looking forward on future WUs. At the time it was fine, no problems for a couple of days - then it went bang big time. Prior to the switch PC had been pretty stable, couple of blips (one of those my fault), all had been fairly good for a few weeks.

I have a 9800GTX+ 512MB Vista32 Phenom2 4Gb 6.6.31

I then realised it all started going wrong when the latest Freehal WUs started up again - you get 25 WUs running simultaneously, its an NCIP. However, there were no previous issues attributed to Freehal, which in any case is cpu bound.

Sunday morning I had a thought re system resources, Freehal is a lot of WUs running, and would increase uptake of system resources. I had been a bit idle and there was a fair bit in the background, desktop cluttered etc, so I set to it and stripped it down to bare bones, stopped the GPUGRID WUs for the moment, and ran the rest with 190.38 loaded (one Aqua using all four cores, and 25xFreehal WUs). Ran fine, no issues, all cpu apps.

I then stepped up the eye candy back to normal, stopped a couple more nugatory services (indexing/terminal/RDC/Tablet etc), and still fine. I then turned on the GPUGRID WUs, and within 30 mins it was back to big boom time again. Turned off Freehal, all was well .... and Freehal I know has not caused me issues before.

I could only conclude that somehow 190.38 is using more system resources than 185.85 (previous driver). So I rolled back to 185.85, and been running full tilt with one Aqua using all four cores, a GPUGRID WU on the 9800GTX+, and 25 x Freehal WUs running concurrently in memory for a few hours now. Rock solid, problems appear to have gone.

Could still be something in the driver of course, I have seen the problems on the board re the 200's, for me it looks like system resources went over the edge after loading 190.38. I did get real wierd blue screens of multiple types during the saga, nothing consistent. Anyway, for what its worth there it is, might help someone track down the 190.38 issues.

See what happens in the morning - or later this morning ..... need to hit the pit ready for work in a few hours. Will post again if it goes bang in the morning, otherwise assume all is running fine.

My account is a bit of a war zone now, apologies, hopefully it will settle back down again.

Regards
Zy

I had a similar problem, but before installing the latest drivers. For the past month or two, Freehal (when running alongside GPUGrid), would severely slow down the computer. RAM wasn't bad or anything, but it just seems much less responsive. Now, if I suspended FreeHal, it went back to normal. I didn't test running Freehal without GPUGrid, but if I had to stop one or the other I chose to stop Freehal.

Now, this seemed to start over a month ago, and I didn't make an effort to resolve. I just don't run Freehal if I am actively using the computer. I never noticed any blue screens, etc., just an overall slow response from the computer in doing anything.

XP32 Q6600 4GB 6.5.0 w/ 2x 8800GT + 1x 9500GT

Anyone else having issue like this or Zydor's?

I've never had a problem with Freehal - always worked fine for me even with GPUGRID. Things changed when I went to 190.38, whether or not running Freehal. Should have stayed with the old maxim, if it aint broke dont fix it. Well now its broke, whether I go back to the old driver (185.85) or not. GPUGRID will no longer run on its own or with other apps .... looks like thats me on vacation for a while....

This was prior to adding the extra memory...... and other things :)

I was on Vista 32bit, changed to 190.38, and all hell broke lose. Initially I put it down to the 190.xx drama, rolled back to 185.85, and still the same, errors everywhere, and more BSODs than I have ever seen before in 20 years. Did all the usual things to try and recover re installs etc etc, still no joy. Other projects also got hammered as a consequence, whereby they in their own way hit whatever faults had crept into the system. Then to cap it all the RAID driver decided to go bye bye's.

Teddy got a cracked skull when he was thrown at the wall :)

In the end I decided to rebuild (again!) but this time go the whole hog. I have been toying with going 64bit prior to Win7.

So, I got a new C: drive, a 1Tb disc, took down what was left of the 4 disc RAID, took out one of the drives, leaving three of the ex-RAID drives in there (plus the new C: drive), formated the lot, loaded Vista 64bit, and at last ..... peace perfect peace. I then fitted an extra 4Gb Corsair Dominator PC2-8500 DDR2 RAM, turned up the wick to 1000Mhz, and she went bang again .... after 2 days of reloading applications.

..... Teddy got another cracked skull ..... (that was yesterday)

This morning I discovered that the motherboard will only support 1000mhz DRAM with one DIM per channel .....

...... Teddy got his third cracked skull

Sooo, this morning I turned it back to 800Mhz for the RAM, and all is now well. Give it a few more hours before I say case closed, but its looking good so far, starting to get WUs back to Projects again. Should be all settled by tonight.

Took a while as I had to do this whilst running my own home based business .... so it was "fun"

Long story, sorry .... but now you know :)

Anyway, now settled with a Phenom2 3.2Ghz 8Gb 800Mhz RAM, Vista 64Bit, 190.18 6.6.31 and enough storage/hard disks to store War & Peace a million times over :)

Aint PCs "Fun" :)

Regards
Zy

Case closed ....

Its been rock solid for 36 hours now, same with the other projects I run so the rebuild went fine. Two GPUGRID WUs sent back in succession since the rebuild, a third happily crunching away. Fortunately I had backups of Work and Personal data cooming out my ears available - I'm paranoid re backup - so lost "nothing" except time and three cracked skulls for Teddy :)

Looking back, what caused it is still not fully clear in terms of sequence apart from the fact that multiple issues appeared at the same time. The catalyst without a doubt was going to 190.38, that install screwed up my beast in a wierd way because as soon as a I started to crunch a GPUGRID WU after the change .... boooom.

Two other things didnt help - unknown to me at the time the RAID was on its way out, finally going "bang" halfway through the recovery rebuild. The red herring was the memory problem towards the end of the rebuild, that was my fault not realising the motherboard limitation of 1 Dim per channel at 1000Mhz RAM - as I had DDR2 RAM, I turned it back to 800Mhz as I previously had it set when on Vista32, and its fine.

Solid now on 64bit Vista / 190.18 / 6.6.31

Bottom line .... I reckon the RAID degrading caused all this, which at the time was not clear - it happened at the time of the doubts elsewhere on the board re the 190.xx drivers, and I drew the wrong conclusions. Meanwhile a degrading array was happily striping away garbage, with the inevitable result when a basic thing like a Video driver is being installed.

It was not a clash re 190.xx vis a vis GPUGRID to clear up any readers doubts, it was the degrading array as it turned out that was the root of my problem.

Such is life's sweet pattern......

Regards
zY

Hey, didn't I say in the other thread that brand memory which appears to be broken right after shipment can often be related to config issues? ;)
......

Hi!

My experience with 190.38 WHQL on XP 32 bit and Vista 64 bit are ok!

Everything works as expected with this drivers and 2x GTX260 SP216 on x38 chipset.

Today I install Windows 7 RTM 64 bit.... and.....

Fiasco!

Everything works ok, but it is 1 big error when suspending and resuming tasks.

WU witch is counted on first Geforce have no problems with suspending and resuming.

But WU witch is crunching on second GeForce - crash!!!

This hapends also when computer is normal restarted...

BOINC 6.6.38...

any tips?

On Vista 64 on the same config all works ok....
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!