• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Security bulletin for nVidia drivers (Update em to 376.33)

Lain

Member
Installed the 376.33 drivers because of this security problem, started playing WoW and suddenly my 1070 fans went into overdrive while I was playing. What the fuck.
 

dr_rus

Member
Wtf is going on in this thread? How is fixing a bunch of reported security vulnerabilities "shitting themselves" and "making your PC blow up"? Or do you guys think that there are no security issues in any random software you're using only because there are no fixes for them? Jeez.

The joys of Nvidia update roulette. You never know what you're going to get.

I get a working driver 100% of the time. Must be something wrong with me.
 
According to this article, only Windows 7, 8 and 8.1 are affected.

http://hexus.net/tech/news/software/100423-nvidia-geforce-37633-whql-drivers-focus-security/

Not sure how they came to that conclusion though.

EDIT: Below is from the changelog.

Windows 10 Fixed Issues:

- Install the Display Container via co-installer. [200228206]
- [375.86, GP104] Corruption in YouTube video playback when two or more videos are playing at the same time in Chrome. [1843100]
- Clean up the escape calls that use physical addresses. [1626231]

Windows 8.1/Windows 8/Windows 7 Fixed Issues:

- Install the Display Container via co-installer. [200228206]
- NV_ESC_ID_NVAPI_REPORT_WFD_HOTPLUG and NV_ESC_ID_NVAPI_ENUM_DISPLAYIDS_EX don't validate the size of input, causing a KERNEL_SECURITY_CHECK_FAILURE. [1808374]
- Clean up the escape calls that use physical addresses. [1626231]
- NVL_ESC_ID_COMMON_REGISTRY_ACCESS escalates user privileges when the registry is accessed. [1359976]

Thanks for posting this. If it's just Windows 7/8 affected by the security issue then I think I'm going to give 376.33 a little while to see if even it manages to fuck something up too. Christ, it's total clownshoes over there.
 

Sky87

Member
I'm on Windows 10, drivers 376.09

Am i good?

A little late, but yes. According to the driver changelog itself, Win 10 users are safe.

Not sure how ''unsafe'' people with 7, 8 and 8.1 are though. Haven't seen any ramifications of this particular issue.
 
How far back does the vuln go? I might be running the base/stock Nvidia driver that Microsoft includes with Windows 10 for my card (GTX 980 Ti).
 

Mohasus

Member
Good to see nvidia releasing driver after driver like they care about the consumer and are always improving their products.

Just updated here, but I hate unchecking 3D drivers everytime.
 

Rodin

Member
A little late, but yes. According to the driver changelog itself, Win 10 users are safe.

Not sure how ''unsafe'' people with 7, 8 and 8.1 are though. Haven't seen any ramifications of this particular issue.

That's reassuring, thanks.

Nvidia really needs to improve QC though, their latest drivers have been a complete joke.
 

inky

Member
So I had the driver release one before this one and I got the old "Nvidia driver has stopped responding and successfully recovered" message twice in short succession, which was the bane of my existence with my old 560ti like 4 years ago.

I decided to update to 376.33 yesterday on this thread's recommendation, and since I did I had about 4 hard locks while gaming. I suspected it was the drivers because as I bought a new GPU a couple of weeks ago I've been testing it heavily and landed on a pretty stable OC already, which I abused for a whole week to really make sure it was rock solid (and it was). I still troubleshooted power management settings, rolled back my OC and a bunch of other stuff just to be safe, but the hard locks didn't go away and happened with 3 different games.

Early on today I switched back to the old November driver release and I had been pushing my GPU back with the stable OC all day (both with games and synthetic benchmarks) and I've found no more crashes or errors.

Sometimes it really feels like a lottery with these driver updates to the point I'm more scared when there is a new release, :lol.
 

dr_rus

Member
So I had the driver release one before this one and I got the old "Nvidia driver has stopped responding and successfully recovered" message twice in short succession, which was the bane of my existence with my old 560ti like 4 years ago.

I decided to update to 376.33 yesterday on this thread's recommendation, and since I did I had about 4 hard locks while gaming. I suspected it was the drivers because as I bought a new GPU a couple of weeks ago I've been testing it heavily and landed on a pretty stable OC already, which I abused for a whole week to really make sure it was rock solid (and it was). I still troubleshooted power management settings, rolled back my OC and a bunch of other stuff just to be safe, but the hard locks didn't go away and happened with 3 different games.

Early on today I switched back to the old November driver release and I had been pushing my GPU back with the stable OC all day (both with games and synthetic benchmarks) and I've found no more crashes or errors.

Sometimes it really feels like a lottery with these driver updates to the point I'm more scared when there is a new release, :lol.

Been running this driver since release with no issues whatsoever. Considering that it's still the same 375/376 branch (started to roll out in late October) I find it very unlikely that your problems are driver related.

People tend to blame too much on drivers these days.
 

Timedog

good credit (by proxy)
My AMD cards I've been using for the last decade have not had any problems. Hopefully the AMD makes shitty drivers meme, propagated mostly by people who don't actually use AMD products, can end now? Pretty please?
 

Sky87

Member
So I had the driver release one before this one and I got the old "Nvidia driver has stopped responding and successfully recovered" message twice in short succession, which was the bane of my existence with my old 560ti like 4 years ago.

I decided to update to 376.33 yesterday on this thread's recommendation, and since I did I had about 4 hard locks while gaming. I suspected it was the drivers because as I bought a new GPU a couple of weeks ago I've been testing it heavily and landed on a pretty stable OC already, which I abused for a whole week to really make sure it was rock solid (and it was). I still troubleshooted power management settings, rolled back my OC and a bunch of other stuff just to be safe, but the hard locks didn't go away and happened with 3 different games.

Early on today I switched back to the old November driver release and I had been pushing my GPU back with the stable OC all day (both with games and synthetic benchmarks) and I've found no more crashes or errors.

Sometimes it really feels like a lottery with these driver updates to the point I'm more scared when there is a new release, :lol.

Nothing in EventViewer? Hard locks usually mean the driver couldn't recover and just hung. Usually happens if you get several driver crashes after eachother. Your EventViewer should tell you if that was the case.

Other than that, it definitely sounds like an unstable OC to me. Weird that it works fine on an older driver though. Keep in mind that even with an unstable OC some games might run just fine. I've had issues in the past with unstable factory OC's, and some games never crashed despite that, others crashed regularily.
 

Cday

Banned
It's cool that they now arbitrarily require an nvidia account to update through geforce experience too. Uninstalled that shit and just update them manually now. So there.
 

inky

Member
Been running this driver since release with no issues whatsoever. Considering that it's still the same 375/376 branch (started to roll out in late October) I find it very unlikely that your problems are driver related.

People tend to blame too much on drivers these days.

I appreciate that you have been running fine, but I don't see how that applies to me as I had the issue I mentioned.

I blame the drivers because with my limited knowledge and testing it was the only thing that did the trick, but you seem knowledgeable so if you can tell me with 100% certainty what was what caused my error I will absolve Nvidia of any and all faults.

Nothing in EventViewer? Hard locks usually mean the driver couldn't recover and just hung. Usually happens if you get several driver crashes after eachother. Your EventViewer should tell you if that was the case.

Other than that, it definitely sounds like an unstable OC to me. Weird that it works fine on an older driver though. Keep in mind that even with an unstable OC some games might run just fine. I've had issues in the past with unstable factory OC's, and some games never crashed despite that, others crashed regularily.

From what I remember it was the general critical error for unexpected rebooting or loss of power (41 I think?) and not a particular driver error.

And I know it's expected that an unstable OC causes this. It's not my first time doing it, so it was the first thing I troubleshooted. I'm not going through everything I did before rolling back to old drivers, but I checked power management settings, thermals, tested memory, hardware connections, OC on and off, AV, pagefile settings, reinstalled the new drivers, changed game settings and some other things. I've fiddle with my computers for about 15 years now, but I do admit I'm not a software or hardware expert or IT guy, or even a very knowledgeable enthusiast. I use my limited knowledge and google for these things and I definitely don't do it for fun, just when things break on me.

The bottom line is, on the old drivers I have been playing without issues, and like I said above I tested the OC settings on games and benchmark programs running stable for, I'm guessing, around 100 hours of game-time/testing combined. Fair enough, I realize it could have been stable until it wasn't, but I was able to replicate the issue without OC in most of the games I had been playing. I reinstalled The Witcher 2 to see how Ubersample fares nowadays and ended up doing a full 30-35 hour run from start to finish because I love the game. 0 issues with it. I regularly play FIFA and Marvel Heroes which aren't really demanding games and for which I don't need or use the OC at all, and I played The Division with OC on for about 12 hours during the weekend (gorgeous game btw). The first game that hanged on me was TW2. Then Marvel Heroes, GTA V, and Skyrim which I haven't played that much but had taken for a spin with OC no problem.

After turning the OC off, all of the above hanged after about a minute (couldn't test The Division because the free period had run out). All things considered, it was a very predictable, very easy to replicate issue. It seemed that everything that could tax the card even a little would cause the system to hang, regardless of settings, OC or anything else. Then I did the other kinds of testing I pointed out above including a clean reinstall of the latest drivers. Finally I went to the old drivers and used most of my day to do another round of game and benchmark testing, OC on and Off, and it was the only thing that worked. So I'm partial to believing something in those drivers disagrees with my setup.

Although the "driver has crashed and recovered" error (which is a different error I got with a different driver release) is something that has happened to me on and off with completely different PCs over the years, and from what I can gather in that experience it was always or mostly related to driver releases.
 

AXE

Member
Not entirely on the point but being that I'm a peasant and can't make a thread, here goes

Anyone having problems with Shadowplay?

It feels like I'm taking crazy pills here. I'm sporting v. 2.x because the version 3.x is just a complete pos. Like, utterly. Now I have to install the v. 2.x again and again and again because the Geforce Experience decides to update itself to version 3 every now and then.

Of course, it fails even in that. There's a runtime error every time it tries that its been like that for couple of months now. Oh well, saves me the trouble to uninstall the version 3.

I might cope with the asstastic design of the UI in the version 3 if they had the only feature I actually use it for: save the last X minutes of gameplay. With version 2.x I could save the last X minutes with ALT + F10 if something nice happened. No point to record anything if nothing happens.

In version 3 I haven't ever gotten the record last X minutes to work. Ever. Never. Like... what the frack?

Is the feature complitely ruined or am I just doing it wrong?

I can get it record like non-stop, but I don't want that. Only if something fun happens. Don't feel like saving a 100 Gb video file when the funny thing takes place in the last 5 seconds.

Am I alone with this crap? How come the record last X minutes doesn't work? It feels that I can't record games properly nowadays either. The new overlay is crap. I have to apply the desktop recording option and I now have all of my recordings in a folder called "desktop" rather than arranged by the game I've recorded.

Nvidia... what the fuck is wrong with you? Why did you decide to go full retard on me?
 

dr_rus

Member
I appreciate that you have been running fine, but I don't see how that applies to me as I had the issue I mentioned.

I blame the drivers because with my limited knowledge and testing it was the only thing that did the trick, but you seem knowledgeable so if you can tell me with 100% certainty what was what caused my error I will absolve Nvidia of any and all faults.

What games did you have the issues in? What is "hard locks"? What is your GPU? Is it overclocked? Is your CPU and memory overclocked? What temperatures are they running under load? Did you check that the driver have actually installed properly? What exactly did you install from the driver package?

Your post gives precisely zero information on what's actually happening on your PC but you are quick to bash the drivers as it's clear to you somehow that it's their fault.

The bottom line is, on the old drivers I have been playing without issues, and like I said above I tested the OC settings on games and benchmark programs running stable for, I'm guessing, around 100 hours of game-time/testing combined.

New drivers may trigger faulty parts of h/w which aren't triggered by older drivers, even just increasing the load on some h/w unit from new drivers may result in issues if that unit is running on the edge already. Fact is that these drivers are causing "hard locks" for you only - this is the first such report I've seen on the internet on this driver version - which lead me to believe that it's not the driver which is a problem here. It's also rather telling that you have these locks in different games - which is almost certainly points to a h/w issue or a general OS issue.
 

inky

Member
What games did you have the issues in? What is "hard locks"? What is your GPU? Is it overclocked? Is your CPU and memory overclocked? What temperatures are they running under load? Did you check that the driver have actually installed properly? What exactly did you install from the driver package?

Your post gives precisely zero information on what's actually happening on your PC but you are quick to bash the drivers as it's clear to you somehow that it's their fault.

I didn't give 0 information. I gave the amount of information I considered relevant to the point I was making because I already went through hours of troubleshooting, and I explained the point above. I already named the games, I already explained how the PC locked, image froze and needed to reboot, my GPU is a 1060 6gb amp, I already explained the overclock situation, no other OCs on my PC, temperature was around 66 C under load with I think a max of 72 in one instance, never gone above that under any different loads, it's controlled by an automatic fan curve. I already said I checked the driver install integrity, the previous numbered driver install as well, and reinstalled the latest one 2 times, including a clean reinstall, restarting my machine every time among other numerous checks I made. I never install the 3D vision package because I don't use it, I install the Driver, Audio and Physx controllers. Probably the only thing I didn't go was underclocking my card under the factory settings, but it didn't occur to me that would be relevant.

I was "quick to bash the drivers" because through my troubleshooting process it was what caused and eventually solved my issue. And I wasn't quick at all, considering I went through days of testing before I determined switching back to old drivers provided the stability I was experiencing before. If that is not enough for you then I'm sorry I can't convince you otherwise. From what I can tell, and I've been running the same hardware for a time now, nothing else has triggered an error of this kind under any kind of heavy load or stress testing or regular use of gaming, so unless something else triggers the same error under exactly the same circumstances then I am not going to assume everything or anything else is the issue when I can't determine such things.

I am not closed to the notion that something else might have been or be the problem, but until such time a problem shows up that isn't directly related to the latest drivers I had installed, I can't either say for sure that it is in fact my OS, or a piece of hardware or whatever else. To you it might sound like I'm out to bash Nvidia or whatever for any problem I had, but it was not the case because believe it or not, I'm more interested in my PC behaving properly than bashing whatever company in a videogame forum. I'm just sharing my experience as honestly as I can.
 
I didn't give 0 information. I gave the amount of information I considered relevant to the point I was making because I already went through hours of troubleshooting, and I explained the point above. I already named the games, I already explained how the PC locked, image froze and needed to reboot, my GPU is a 1060 6gb amp, I already explained the overclock situation, no other OCs on my PC, temperature was around 66 C under load with I think a max of 72 in one instance, never gone above that under any different loads, it's controlled by an automatic fan curve. I already said I checked the driver install integrity, the previous numbered driver install as well, and reinstalled the latest one 2 times, including a clean reinstall, restarting my machine every time among other numerous checks I made. I never install the 3D vision package because I don't use it, I install the Driver, Audio and Physx controllers.

I was "quick to bash the drivers" because through my troubleshooting process it was what caused and eventually solved my issue. And I wasn't quick at all, considering I went through days of testing before I determined switching back to old drivers provided the stability I was experiencing before. If that is not enough for you then I'm sorry I can't convince you otherwise. From what I can tell, and I've been running the same hardware for a time now, nothing else has triggered an error of this kind under any kind of heavy load or stress testing or regular use of gaming, so unless something else triggers the same error under exactly the same circumstances then I am not going to assume everything or anything else is the issue when I can't determine such things.

I am not closed to the notion that something else might have been or be the problem, but until such time a problem shows up that isn't directly related to the latest drivers I had installed, I can't either say for sure that it is in fact my OS, or a piece of hardware or whatever else. To you it might sound like I'm out to bash Nvidia or whatever for any problem I had, but it was not the case because believe it or not, I'm more interested in my PC behaving properly than bashing whatever company in a videogame forum. I'm just sharing my experience as honestly as I can.

You shouldn't let it get to you, I've observed dr_rus to be a tirelessly valiant protector of the Nvidia brand and surely doesn't mean anything personal by it.
 
Top Bottom