Freezing problems with the latest Nvidia 306.23 WHQL (Solved! Was not the Drivers!)

[The_Golden_Man]

Well, I upgraded from the 306.02 Beta yesterday, and have had about 4 -5 freezes. It just freezes on the desktop when surfing or just sitting there.

I have reverted back to the 306.02 Beta, which I've never experienced this behaviour with. So we'll see

Specs in signature. And running Windows 7 Ultimate 64 Bit with all the latest live updates.

Others having similar issues?

 

[chimaxi83]

Wait just a minute here... Nvidia driver problems? Does not compute /sarcasm

Well, I haven't tried them yet, so I can't say with certainty, but the last time an Nvidia driver upgrade (304.48 I think) caused freezing/instability, it was because my overclock wasn't as stable as I thought. Are you running those 670's at factory settings?

 

[cmdrdredd]

Haven't tried them. I'll report back over the weekend.

 

[The_Golden_Man]

Wait just a minute here... Nvidia driver problems? Does not compute /sarcasm

Well, I haven't tried them yet, so I can't say with certainty, but the last time an Nvidia driver upgrade (304.48 I think) caused freezing/instability, it was because my overclock wasn't as stable as I thought. Are you running those 670's at factory settings?

They are running stock. And it happened on the desktop. No problems so far, after reverting back to the 306.02 Beta again.

So it seems the latest WHQL was causing the freezing bug.

 

[Meghan54]

They are running stock. And it happened on the desktop. No problems so far, after reverting back to the 306.02 Beta again.

So it seems the latest WHQL was causing the freezing bug.

And this is why I always wait to upgrade drivers. Rather have others be the beta testers for new drivers....(Thinking back to the nvidia release that screwed up the fan profile so the cards overheated. Wonder if nvidia ever replaced the damaged cards from that faux pas. Probably not, given nvidia's history of denying responsibility for problems and trying to defer them to anyone else....the notebook fiasco is but one example.)

Good luck, OP. I'm searching for a sucker, errrr, nice person willing to trade a good 7970 for my Gigabyte Windforce GTX 670 OC.

 

[The_Golden_Man]

Turned out it was not the drivers causing the freezing!

 

[railven]

Turned out it was not the drivers causing the freezing!

Then what was it?

 

[nanaki333]

Then what was it?

probably 1D-10-T error. i hate when i get those

 

[railven]

probably 1D-10-T error. i hate when i get those

I guess I'm not as hip, had to Google that thinking it was a legit error.

I see what you did there. haha.

 

[The_Golden_Man]

Then what was it?

I've used 4x4GB Corsair Vengeance 1600MHz 9-9-9-24-2t RAM for the last couple of months. Running them at stock and everything has been fine.

Then the same day I installed the latest WHQL I switched out two of my Black Vengeance with Blue ones. Just to test that these blue ones worked (Have had them laying around). They are the excact same ones, except they are the blue edition.

The freezing bug started to occur, and I suspected the RAM, but Prime95 Blend worked fine, and they also passes memtest86, so I then turned my attention to the latest WHQL. Turns out it has to be those RAM sticks in combination with the two black ones that caused the bug. It froze even if I reverted back to the 306.02 Beta drivers.

I've installed the two black ones again, and installed the WHQL again. No freeze yet.

I suspect those two blue sticks have another revision or something, and did not work well with the black ones with current settings or something, even if they ran at stock XMP@1600 9-9-9-24-2t. Will have to test the blue ones by themselves.

 

[railven]

I've used 4x4GB Corsair Vengeance 1600MHz 9-9-9-24-2t RAM for the last couple of months. Running them at stock and everything has been fine.

Then the same day I installed the latest WHQL I switched out two of my Black Vengeance with Blue ones. Just to test that these blue ones worked (Have had them laying around). They are the excact same ones, except they are the blue edition.

The freezing bug started to occur, and I suspected the RAM, but Prime95 Blend worked fine, and they also passes memtest86, so I then turned my attention to the latest WHQL. Turns out it has to be those RAM sticks in combination with the two black ones that caused the bug. It froze even if I reverted back to the 306.02 Beta drivers.

I've installed the two black ones again, and installed the WHQL again. No freeze yet.

I suspect those two blue sticks have another revision or something, and did not work well with the black ones with current settings or something, even if they ran at stock XMP@1600 9-9-9-24-2t. Will have to test the blue ones by themselves.

Memory configuration issues piss me off SOOOO MUCH. They can pass Prime95/Memtest86 burn-ins with flying colors, and just for whatever reason still screw something up on the OS end.

Oddly enough, I had a similar issue with some Corsair XMS sticks back before I realized they sold them in different volt packages (under the same name, timings, speed, but of course different SKU) so when I tried to add 2 more sticks to a system the never ending BSODs drove me crazy since I couldn't figure it out.

 

[The_Golden_Man]

Memory configuration issues piss me off SOOOO MUCH. They can pass Prime95/Memtest86 burn-ins with flying colors, and just for whatever reason still screw something up on the OS end.

Oddly enough, I had a similar issue with some Corsair XMS sticks back before I realized they sold them in different volt packages (under the same name, timings, speed, but of course different SKU) so when I tried to add 2 more sticks to a system the never ending BSODs drove me crazy since I couldn't figure it out.

Yeah... This was so strange... Passed every stability test with flying colours, but for some reason my system just froze on desktop once in a while. It had to be that blue Vengeance in mix with the two black ones causing it, because it did never happen when I used my 4 black ones. And now with the 4 black ones, I've not experienced the freeze again. So I cannot imagine it will happen again now.

 

[Final8ty]

I've used 4x4GB Corsair Vengeance 1600MHz 9-9-9-24-2t RAM for the last couple of months. Running them at stock and everything has been fine.

Then the same day I installed the latest WHQL I switched out two of my Black Vengeance with Blue ones. Just to test that these blue ones worked (Have had them laying around). They are the excact same ones, except they are the blue edition.

The freezing bug started to occur, and I suspected the RAM, but Prime95 Blend worked fine, and they also passes memtest86, so I then turned my attention to the latest WHQL. Turns out it has to be those RAM sticks in combination with the two black ones that caused the bug. It froze even if I reverted back to the 306.02 Beta drivers.

I've installed the two black ones again, and installed the WHQL again. No freeze yet.

I suspect those two blue sticks have another revision or something, and did not work well with the black ones with current settings or something, even if they ran at stock XMP@1600 9-9-9-24-2t. Will have to test the blue ones by themselves.

Yep had the mem issue before but most people just don't seem to understand that the tests don't prove everything.

Memory & motherboard compatibility issues that can lead to CTD, Blue screen, driver has stopped responding from 3 to 8 times a day while gaming

OCZ Gold 4GB (2x2GB) DDR3 PC3-10666C9 (OCZ3G1333LV4GK) (1333MHz)
Asus M4A79XTD Evo.
Passed memtest 12 hours, Passed Prime-95 8hours passed, Furmark passed, 3Dmark06 passed.
Changed the gfx card.
Drivers 10.4/ 10.5 hotfix/10.6.

Fired up WOW which is the only game the customer plays at the moment & instant multiple driver has stopped responding or sometimes hours or more would pass.

Put in some G.Skill RipJaw 4GB (2x2GB) DDR3 PC3-10666C8 & not had a single issue.

 

[BallaTheFeared]

Wait just a minute here... Nvidia driver problems? Does not compute /sarcasm

Ops?

On a serious note, glad to found the problem OP. Were you getting a blue screen after the freeze, or at least an error number on reboot?

 

[The_Golden_Man]

Ops?

On a serious note, glad to found the problem OP. Were you getting a blue screen after the freeze, or at least an error number on reboot?

No blue screen or error code. It just froze. Had to turn my computer off and on again. It probably happened about 4 or 5 times in 24 hours.

Now I'm 100% sure it was those blue Vengeance when mixed with two black ones, because I've been using the 4 black ones for about 24 hours without a freeze yet (And by all means, they have been stable for months).

As I've said, will have to test the two blue ones by themselves. Even if I currently don't need them, I must confirm they work.

 

[Termie]

I was going to guess this - in fact I had drafted a reply yesterday but figured I'd let the OP just answer the question.

Every major mystery bluescreen or failure to POST I've experienced has been due to memory settings. They are the bane of all system tweakers. They can pop up in an instant with the slightest change, even when stable with other hardware/software.

 

[The_Golden_Man]

Lol... Tried the two blue Vengeance sticks alone, and the same freezing problems occured... Then I tried to up the Vram Voltage to 1.6v, and it became even worse. Also tried upping VCCIO from 1.050v to 1.1v, did not help...

Then I tried to set the VRAM voltage to 1.45v. It actually fixed it. No more freezes. How strange! Stock on these is 1.5v. And my 4 Black Vengeance works perfectly at 1.5v!

Forgot to mention that even if my system froze once in a while at 1.5v VRAM it was stable in memtest86, memtest for Windows and Prime95 Blend. Very strange.

 

[Pheran]

Did you try HCI MemTest? I've had it find problems that are difficult to detect with memtest86.

 

[The_Golden_Man]

Did you try HCI MemTest? I've had it find problems that are difficult to detect with memtest86.

Yeah, I ran that too. All was fine. It froze anyway when in idle. I had to turn VRAM voltage down to 1.45v. Not had a single freeze since I did that. Pretty sick.

 

[Pheran]

Yeah, I ran that too. All was fine. It froze anyway when in idle. I had to turn VRAM voltage down to 1.45v. Not had a single freeze since I did that. Pretty sick.

Yeah that is pretty weird. Maybe the lower voltage actually cooled down the chips so they work better.

 

[The_Golden_Man]

Yeah that is pretty weird. Maybe the lower voltage actually cooled down the chips so they work better.

I highly doubt that is the answer. Because when they ran at 1.5v it never froze when stability testing (When they should get the most hot), only in idle/regular desktop use, or when just doing nothing. 1.45v fixed that.

 

[Pheran]

I highly doubt that is the answer. Because when they ran at 1.5v it never froze when stability testing (When they should get the most hot), only in idle/regular desktop use, or when just doing nothing. 1.45v fixed that.

Hmm, yeah that really makes no sense then. Proof that computers are black magic.

 

[The_Golden_Man]

Hmm, yeah that really makes no sense then. Proof that computers are black magic.

at 1.6v it froze when logging in to Windows. Lol.. 1.45v and all is fine.

I will test these two sticks a few more days, then put back my 4x4GB Black Vengeance RAM. They will operate just fine at 1.5v.

Only difference between the black and the blue ones, appart from the colour is the black ones are version 8.16, while my two blue ones are version 2.12