At least they found the bug in millions of lines of coding... Not so easy as you think
Just revert back to the stable driver until the new one is released.
I can imagine it's not easy but I imagined it would be a priority at least.
And for the 3rd time dude the latest beta is the most stable driver so far for the 700 series.
At least they found the bug in millions of lines of coding... Not so easy as you think
Just revert back to the stable driver until the new one is released.
I can imagine it's not easy but I imagined it would be a priority at least.
And for the 3rd time dude the latest beta is the most stable driver so far for the 700 series.
You can have new functionality or bug fixes within a period of time but not both. Many people not involved with the actual software development hands on do not know this and often force both which leads to some very horrific code.
So today it finally crashed while I was at work. I see an interesting timeline of the event itself and before/after items in the event viewer:
Link
The bottom three tasks involving the Kernel and Service Control Manager tell that my computer was entering sleep mode due to the timeout settings I have on my PC. About 4 minutes later you see the nvlddmkm entry which is the error thrown by the TDR service (Timeout Detection and Recovery) when it detects that a driver has not responded to within a timely manner and forcefully restarts the driver to prevent a BSOD.
4 minutes later the Kernel-Power reports that an issue occurred with the MTRRs in the CPU while it was asleep. This looks to be the core reason why my computer appears unstable with browsing though I will have to do some fiddling with settings to see what settings I have that are bad.
The next two entries 8 minutes and 5 minutes later respectively are just some items being done while the system is asleep (I'm not sure if this is supposed to happen or not. If not it means that something isn't going to sleep mode when it should be).
The next two warnings are the display driver forcefully stopped by the TDR (both events have the same ID so only one crash occured). This triggered the system to exit sleep mode on its own.
I've seen this behavior on going for awhile now even on 300.14 drivers but never bothered to do anything about it. Not sure what is going on but it's definitely something more than nVidia's drivers. It pays to not jump the gun and actually analyze the problem and see what may be the issue. I'm pretty sure that MTRRs thing is relatively new and what's causing the general instability of 2D applications running on my machine.
Yup same errors to everyone with a card from the 700 series, you can rest assure that it's not the card and it's not related to sleep settings or any other thing but the driver.
I follow up closely the nvidia forum and like I said they finally found what the problem was and a fix is ready for 36 hour tdr's and the random desktop tdr's both.
Also after sending countless minidumps I got an email response from nvidia's driver tech support that confirms that.
The only bummer is we'll have to wait till the end of the month like we didnt wait enough already..
The system firmware has changed the processor's memory type range registers (MTRRs) across a sleep state transition (S4). This can result in reduced resume performance.
I think I have more problems than just that nvidia problem. The MTRRs message reads this:
The system firmware has changed the processor's memory type range registers (MTRRs) across a sleep state transition (S4). This can result in reduced resume performance.
Previously I think the crashes have been ongoing but with minimal impact on recovery. That's a CPU issue I need to investigate.
Hmmm interesting I searched a little about it and it seems that most people who suffer from it upgraded from win 7 to 8 and didnt have the problem before..Link
Btw I never use sleep mode, it's a risky situation when it's possible your computer wont wake up someday, I'm serious my brother's gtx275 droped dead because of sleep mode and we confirmed it because it still had warranty and the card was fixed but they never told us what they did exactly just that a weird bug related to sleep caused it...
[Edited by kingkob, 8/14/2013 8:47:14 PM]