NVIDIA GeForce 375.86 WHQL Drivers Have issues

Yup this driver gave me a Stop Error/Blue Screen with Windows 10 64-bit Pro

 

No issues here.

 

BF1 plays perfect on my 980M with the 375.76 driver.
I will skip this driver and wait for the next driver fix from nvidia

 

Yet my card hasn't be downgraded from the drivers. Hmmm how does that work. Its almost like there is a bug that only happens to certain people and not a downgrade on all cards.

****ing idiot

 

:banana: Nvidia 2016.:banana:

 

More like another Gx x04 fiasco, as usual :grin: That's why I never bothered with those chip models.


I remember its been like that since Fermi and it hasn't changed yet.
Fermi 104/114 could TDR'd a lot, Kepler 104 had downclocking issues and stutter, Maxwell 3.5gb fiasco and now Pascal vram bios issues, driver vram issues, etc..

paranoid? perhaps, but I know these "performance" chips are almost 99% guarantied to have some kind of driver issue.

 

Got home from work, saw the downloaded driver kit on my desktop, said to check first if any beta drivers showed up before installing it.

Then i see this thread, i'll skip this one. Never skipped drivers before, first time for everything.

 

Works great with Rocksmith and BF1 here.

 

since they first introduced the "Optimal" power option in the drivers
iv seen more and more clock complains from different users. i had some unstable
clocks and some very strange artifacts in idle on my MSI 1070 Gaming X with
one of the first drivers that came with "Optimal" as default. ever since
i am constantly changing to "Adaptive" and i didn't encounter that
problem ever since. most of the time i am installing every new driver set,
but il skip this one.

im no expert, and i cant point if that power option fixes clock problems.
only encountered it a few months ago and switching it helped for me.

 

...

The issue effects the entire pascal lineup, so I'm not sure what your point is. A better argument would be "this is why i never bothered getting cards early on, because of potential issues that will be hammered out by the end of the life cycle"

That's not even to mention that this problem is like a randomly introduced driver issue. Comparing it to the memory problem on the 970, or the micron issue on 1070 doesn't make any sense. Like if I had made a post when the GIF problem arose, even on the 980Ti, a few drivers ago saying "This is why I avoid x02 models" how would you have responded? Because it's basically the same thing here.

I've had 0 problems on my 1080, aside from it not idling correctly - which happens on the 780Ti/980Ti/Titan X Pascal too.

 

Higher end chips had its moments too, but not so severe like "midget" performance chips and that's a fact. I've been actively following and using nv drivers since 6600GT so I've seen a number of bugs.:nerd:

Also please show me a TitanXP user reporting this memory anomaly, so far I haven't seen one.



From what I saw that idle clock is entirely different issue, and happens only to a specific user base - minority if you will.

 

It has nothing to do with it being "higher end" or "midget part" or "performance part" or whatever other label you want to use. x04 chips are the first of each architecture design out of the gate, thus they are obviously going to have more problems - those problems are fixed by the time the Ti variants ship.

And you're right, Titan XP doesn't have the issue - Nvidia is reporting that it only effects factory overclocked cards. There are no GP102 factory overclocked cards, thus none have the issue. But again, it just further proves the point that the card being x04 means nothing. Had they shipped factory overclocked GP102's they'd have the same issue.

 

No use bro. Tj is just stuck on that x04 hate train and will not get off.

 

You can shove those x04 chips up you know where, those are meant for phones

 

No issues with 375.86 here either, played PCars a few hours with G-Sync and TrackIR, steady fps and normal temperatures on GPU/CPU (someone over at geforce.com blames the driver to overheat the CPU, lol).
I'm running the card with the optimal power setting in NVCPL.

 

Titanfall 2 runs like aids with these drivers. Absolutely destroyed fps in games for me.

 

Looks like I'm lucky..No Problems here..

 

Seems to be working fine on my end. What's going on?

 

I really expected better from Nvidia, I cant for the life of me see how this made it past testing. Makes you question their driver testing department.

 

All these news-sites are incomplete. Let me fill-in the blanks. It is true that some 10x-tier cards are affected with a memory-issue that results in the clock-speed to be stuck in a low-mode. Which results in low-fps.

On the beginning it was already clear that not everyone with a 10x-tier card is having issues with the driver. I for example have a MSI Gaming X 1080GTX and have zero issues. And there are multiple other people reporting no issues either.

Yet it made us curious what was causing it. On the forum we did search. Eventually NVIDIA found the root-cause of it. It seems that all the 10x-tier cards with a stock OC-memory or a certain bios-version are having trouble. Some people did even flash the Gaming X-bios on the Z-variant and reported that the problem went away too.

To be fair. Nvidia testing department probably used some cards that were running stock-speed on the memory. Or had cards with a specific bios. Then this issue could be easily missed by the testing department.

So if you have low-fps. Roll back to 375.70 or wait on the hotfix. If you have no issues, then you can call yourself lucky and the driver works pretty well in games.