ASUS GTX 960 STRIX 4GB Issues

[inyongthegr8]

Good day, everyone!

I'm currently using Windows 10 Pro and for some reason, the latest version of NVIDIA Game Ready Driver (359.06) seems to crash randomly. Is this normal? Before I changed my GPU from 550Ti to 960, Version 355.60 didn't crash at all. The moment I changed my GPU, I'm experiencing random graphics crashes, however, there is no black screen, freezes or any other kind of signs of faulty card symptoms in my card. I don't have any idea what's wrong with my card.

I am also using GPU Tweak provided by ASUS and I think it's only compatible with windows 7, Vista and XP.

Is this normal? According to what I have searched, it said it's normal since Windows 10 and Nvidia have driver conflicts, but didn't happen in 355.60.

Any thoughts for this one? Do I have a faulty card or something? Or should I not touch the overclock settings?

Thank you for your replies and have a nice day.

PS. I purchased it few days ago and I can get RMA for that if ever I need replacement

 

[xorbe]

Not normal.

TDRs (time-out detection and recovery) can be caused by many things, not just the gpu:

1) gpu problem (failed or aggressive oc)
2) cpu problem (aggressive oc / cooling issue)
3) ram problem (failed or aggressive timings)
4) psu problem (failed, noisy voltage ripples or voltage out of spec)
5) OS problem (malware)
6) nvidia driver bug, or other driver causing long delays
7) motherboard problem (failed caps, hairline crack traces)
8) software problem (msi ab/precision overlay, 3rd party screen cap)

Yeah, they are a pita to root cause.

 

[inyongthegr8]

Not normal.

TDRs (time-out detection and recovery) can be caused by many things, not just the gpu:

1) gpu problem (failed or aggressive oc)
2) cpu problem (aggressive oc / cooling issue)
3) ram problem (failed or aggressive timings)
4) psu problem (failed, noisy voltage ripples or voltage out of spec)
5) OS problem (malware)
6) nvidia driver bug, or other driver causing long delays
7) motherboard problem (failed caps, hairline crack traces)

Yeah, they are a pita to root cause.

I'm using a non 80+ power supply cooler master, so do you think I should change my power supply if that's the case?

I'm using 500w Power supply from coolermaster by the way.

Upgrading GTX 550Ti to 960 will only cost 4 watts in terms of power so it can still be handled from what I know.

I never OC my cpu, but I tried OCing the GPU but with Asus's OC Mode Preset only. Is it a bad idea to OC my GPU?

 

[stevorob]

I would first start with a complete GPU driver wipe. Use Display Driver Uninstaller (google it) and follow the directions there, then re-download the newest drivers straight from nVidia's site and reinstall.

I would also ensure that all your hardware is running at stock frequencies. Turn all overclocks down until you are able to isolate the cause.

 

[inyongthegr8]

I would first start with a complete GPU driver wipe. Use Display Driver Uninstaller (google it) and follow the directions there, then re-download the newest drivers straight from nVidia's site and reinstall.

I would also ensure that all your hardware is running at stock frequencies. Turn all overclocks down until you are able to isolate the cause.

Done it already. However, I couldn't rectify that kind of problem for some weird reason. Went to safe mode, used DDU and restarted my PC. Re-downloaded the newest drivers already straight from nvidia site (chose the right card and OS as well) and then the same thing happens.

 

[thebluebumblebee]

I'm using 500w Power supply from coolermaster

What's the actual P/N?
You're trying to use a modern mid range GPU (4 GB is a waste for this GPU) on a low end motherboard and (I think) low end PSU. Recipe for trouble.

 

[inyongthegr8]

What's the actual P/N?
You're trying to use a modern mid range GPU (4 GB is a waste for this GPU) on a low end motherboard and (I think) low end PSU. Recipe for trouble.

90YV07Q0-M0NA00

So that means my motherboard is a bottleneck?

 

[stevorob]

Done it already. However, I couldn't rectify that kind of problem for some weird reason. Went to safe mode, used DDU and restarted my PC. Re-downloaded the newest drivers already straight from nvidia site (chose the right card and OS as well) and then the same thing happens.

When does the driver crash? Does it crash while just sitting idle in Windows, or does it only crash when you start to load the GPU? If it crashes when you load up the GPU, sounds as if it could be a power related problem.

What's the actual P/N?
You're trying to use a modern mid range GPU (4 GB is a waste for this GPU) on a low end motherboard and (I think) low end PSU. Recipe for trouble.

I'd tend to think the PSU more of a culprit than the motherboard.

 

[inyongthegr8]

When does the driver crash? Does it crash while just sitting idle in Windows, or does it only crash when you start to load the GPU? If it crashes when you load up the GPU, sounds as if it could be a power related problem.

It crashes on background and windows notifies me that the display driver has crashed, but then, I do not experience any FREEZES, CRASHES or Black Screens or anything in that matter. It's as good as new, stevorob.

I'd tend to think the PSU more of a culprit than the motherboard.

I don't think my motherboard has a problem :/
Most likely. They say, better excess than lacking right?

 

[thebluebumblebee]

90YV07Q0-M0NA00

So that means my motherboard is a bottleneck?

CM PSU P/N's start with RS, like "RS-525-PCAR-D3"

 

[inyongthegr8]

CM PSU P/N's start with RS, like "RS-525-PCAR-D3"

oh ok sorry here: RS500PCARD31114711201

 

[inyongthegr8]

So far I have uninstalled GPU Tweak II by Asus, as well as did stevorob's advice and I'm observing how things will go at the moment

 

[inyongthegr8]

Alright here's the event viewer... It crashed on background while I'm playing a game, and I don't know why

 

[stevorob]

This smells of a power issue. I'd try to find a spare PSU (something more modern) from a friend/family member and give that a shot.

 

[thebluebumblebee]

I can not find a review for that PSU, but a very similar one failed at [H]: http://www.hardocp.com/article/2009/02/16/entry_level_power_supply_roundup/2#.VmXa6NUrKUl

I might be too quick to point at the PSU, but it is one of the easier things to troubleshoot.

Most likely. They say, better excess than lacking right?

You do not need a larger PSU. Just a better one. I'd rather see you get a good 500 watt than a mediocre 750 watt.

Here's a database of PSU reviews to help you choose: http://www.realhardtechx.com/index_archivos/Page364.htm

You might check for a BIOS update for your motherboard. Some of these newer GPU's don't play well with non-UEFI motherboards.

 

[inyongthegr8]

This smells of a power issue. I'd try to find a spare PSU (something more modern) from a friend/family member and give that a shot.

I wish I have one friend that has a spare power supply... Because if not, then I might spend on a new power supply. The tech support of the shop where I bought my graphics card asked me if I use a true rated 600w and I told them I don't... I used 500w and I know it's stable because GTX 550ti uses 116W while GTX 960 uses 120W (according to the reference card). But since it's factory overclocked, it can be more. According to Asus, it may need up to 150 watts at peak usage.

I can not find a review for that PSU, but a very similar one failed at [H]: http://www.hardocp.com/article/2009/02/16/entry_level_power_supply_roundup/2#.VmXa6NUrKUl

Based on what I read in the conclusions section, the parts is built from AcBel... I can suspect that it's a bad PSU from two years up to now... I never knew this before. I think AcBel is also sold in our country in the past... which I'm also trying to avoid.

You might check for a BIOS update for your motherboard. Some of these newer GPU's don't play well with non-UEFI motherboards.

No updates since then. It's the latest motherboard bios update that it can offer since 2014. In addition to this, my Motherboard Bios is up to date.

 

[inyongthegr8]

Update 1: Replaced with a new power supply (650W). The driver crash still occurs, but only only in Dragon Nest. I'll try using compatibility mode at the moment. The rest of the games (Counter Strike: Global Offensive, Lucius, Dream, to name a few) do not crash and does not experience driver timeouts.

I'm suspecting that it's the fault of poor coding in the game itself...

 

[xorbe]

Update 1: Replaced with a new power supply (650W). The driver crash still occurs, but only only in Dragon Nest. I'll try using compatibility mode at the moment. The rest of the games (Counter Strike: Global Offensive, Lucius, Dream, to name a few) do not crash and does not experience driver timeouts.

I'm suspecting that it's the fault of poor coding in the game itself...

TDRs like to play hide and seek ... don't be so sure.

 

[stevorob]

I take it the other games that are now doing fine were crashing before, no?

What PSU did you end up getting as a replacement?

 

[thebluebumblebee]

You might check for a BIOS update for your motherboard. Some of these newer GPU's don't play well with non-UEFI motherboards.

That is a UEFI motherboard.
IMHO, that motherboard is meant for the workplace. You might want to double check all of the power saving features that it has, especially the "EPU" settings.

 

[stevorob]

That is a UEFI motherboard.
IMHO, that motherboard is meant for the workplace. You might want to double check all of the power saving features that it has, especially the "EPU" settings.

Speaking of power settings, should also make sure that in nVidia control panel, under Manage 3D Settings, set Power Management mode to "Prefer Maximum Performance"

 

[inyongthegr8]

I take it the other games that are now doing fine were crashing before, no?

What PSU did you end up getting as a replacement?

Other games do not experience TDRs, only in Dragon Nest and it happens in background for some reason. I've played Lucius, Portal 2, CS:GO and the likes for a very long time and never experience freezes, black screens and glitches.

I have a friend who has the same card as mine, using windows 7 but do not experience TDRs when playing Dragon Nest. I might suspect it's the application itself. Tried setting the compatibility mode to its lower versions of windows and still the same thing.

I replaced my PSU to FSP Raider 650w 80+ Silver. Still experiencing the same problem.

Now I'm downgrading the graphics driver to 355.60 (stable) to see if Dragon Nest experiences this problem again.

Edit 1:

Speaking of power settings, should also make sure that in nVidia control panel, under Manage 3D Settings, set Power Management mode to "Prefer Maximum Performance"

Does this apply in games only or in general? Because if the global setting prefers maximum performance, I will get 3d clocks during idle loads, am I correct?

 

[inyongthegr8]

Update 2: 355.60 is stable with the factory settings. Will not upgrade to 359.06 at the moment

 

[xorbe]

Update 2: 355.60 is stable with the factory settings. Will not upgrade to 359.06 at the moment

It could be the driver+game. It could also be the driver stressing the card in a different way that sensitizes the problem. At the end of the day, I realize you just want to play the game, whether you've found the actual problem or not.

 

[inyongthegr8]

Update 3: Tried upgrading to 359.06 again.

I can certainly say that Game DVR is the culprit... Turned off Background recording to see if it occurs again