Video card crashing? Screen goes black with distorted pixels and color.

[bestcommenteverzzz]

I recently bought a 430 w psu for my pre built hp pavilion 500-056 and a gigabyte rx 460 4g oc. I've been playing games for about a week when this started happening. The first two times, the screen just went black and my monitor said no input signal found, please check the video cable/source so I assumed it was just a driver crash. After those 2 times, It happened again when I was playing h1z1. The game froze first, then the sound looped for about 3 seconds, before the screen turned black. The screen then turned on and off with distorted colors and pixels for about 5 seconds before going dark completely. I'm not sure what could be causing this but currently, I am looking for driver updates and monitoring the temps. I would appreciate any help with this. Thanks.

I also just installed GPU-Z. Maybe I could post my log and have you guys take a look? I'm don't exactly know quite what I'm looking at in the interface.

 

[FordGT90Concept]

Do the Application and System event logs have anything? If it was a driver crash, they should.

I would test with a different PSU.

 

[bestcommenteverzzz]

Usually that means not enough power through the 12v PCIE power connectors. I would definitely try a different PSU first.

Hi. My GPU doesn't require any PCIE power connectors and runs off of the motherboard slot. I also don't have any stronger PSU. Do you think that I should reseat the card?

 

[FordGT90Concept]

Doesn't hurt but probably won't fix it.

I edited in case you missed it (check event logs via Event Viewer).

You could also check the temps via "Radeon Settings."

 

[bestcommenteverzzz]

Doesn't hurt but probably won't fix it.

I edited in case you missed it (check event logs via Event Viewer).

You could also check the temps via "Radeon Settings."

Thanks for replying. I'll take a look at the event logs. I'm currently updating my driver as well. I appreciate your quick replies.

Edit: I'm at the event logs but I'm not quite sure where to look. Could you guide me to where the crash logs should be at? I'm currently looking under "Application" but there is only one error log and it isn't a crash. It seems really weird. I just uninstalled Razor Cortex.

 

[bestcommenteverzzz]

Doesn't hurt but probably won't fix it.

I edited in case you missed it (check event logs via Event Viewer).

You could also check the temps via "Radeon Settings."

I just updated my driver. Could you take a look at the event log screenshot I took and the edit in my response?

 

[eidairaman1]

What's the power supply?

 

[Komshija]

It's too weak PSU for my taste. Quality 430W PSU (minimum "80 Plus Gold" and minimum 400W actual power) is the minimum for such system, especially if you have a 100W CPU. There could be cable issues, so unplug the video cable, clean its connectors and plug it again. After that remove GPU, blow the dust out of its PCI-e slot, put back GPU and check its temperatures, especially considering VRAM. They shouldn't be above 94°C under maximum possible load - for the example with Fur Mark test for 20-25 minutes.

Test your GPU with Fur Mark for about 20-25 minutes and watch for the temps. If system crashes and the temps were OK, it's could be the PSU that's causing the problems. If you have no issues during low-intensity tasks such as watching 1080p or 1440p videos on youtube, working with MS office etc, watching movies etc, I'm 99% sure it's PSU. In such case, get another good PSU with minimum declared 500W output. I recommend Enermax Maxpro EMP500AGT or FSP Hyper HP500S.

 

[DeathtoGnomes]

dont use Furmark, too many people are having issue from using it.

 

[FordGT90Concept]

That event log you took a picture of is related to Razer, probably not causing the problem.

You're looking for warnings and errors, especially System -> Kernel-Boot which logs information about the previous shut down.

dont use Furmark, too many people are having issue from using it.

People with unstable systems you mean? >:D I can run Furmark all day and all night long without problems.

 

[Caring1]

Post a screenshot of GPU-z with the sensors tab open after the GPU has been used and warmed up.
If you have HWMonitor, post a shot of that too with all tabs open.

 

[DeathtoGnomes]

...

People with unstable systems you mean? >:D I can run Furmark all day and all night long without problems.

I'm not prejudiced against unstable users.

 

[Bill_Bright]

Why did you replace the PSU? If the problem started soon after putting in the new PSU, it sure points to the new PSU as a possible suspect. But if the problem was happening before the new PSU too, then probably not the problem.

What is the brand and model number of this new PSU?

Technically, a decent 400 - 430W PSU is plenty. I note Gigabyte recommends 350W minimum. But not all supplies are created equal. You don't buy a new Porsche then feed it generic fuel from the corner Tobacco and Bait shop.

 

[eidairaman1]

I'm not prejudiced against unstable users.

Newer cards auto throttle when that program is run anyway, I use Unigen for cpu and gpu stability testing.

 

[EarthDog]

Newer cards auto throttle when that program is run anyway,

Which is a problem because you aren't testing the 'throttled' clocks, but what you set it at.

 

[eidairaman1]

Which is a problem because you aren't testing the 'throttled' clocks, but what you set it at.

Ya, its funny because I use unigen for system stability testing.

 

[Prima.Vera]

oven bake trick will fix it.

 

[EarthDog]

Ya, its funny because I use unigen for system stability testing.

You = the "royal" you.

Some people swear by it. To me, it simply needs to go away. Nothing else will throttle your cards out of the gate like Furmark will... bone stock it has throttled several cards for immediately tripping off power/VRM limits (thermal isn't close).

 

[eidairaman1]

You = the "royal" you.

Some people swear by it. To me, it simply needs to go away. Nothing else will throttle your cards out of the gate like Furmark will... bone stock it has throttled several cards for immediately tripping off power/VRM limits (thermal isn't close).

Yuppers-I followed it since the inception, i dont waste time on it.

MSI KOMBUSTOR is based on it lol.

 

[EarthDog]

I don't touch Kombuster either. Last I checked, it did the same thing.

 

[INSTG8R]

oven bake trick will fix it.

It's a brand new card...We're not talking about some old 7970 or something it's a RX460....

 

[FordGT90Concept]

Newer cards auto throttle when that program is run anyway, I use Unigen for cpu and gpu stability testing.

The fans accelerate to meet the temp target (I think it was 60C). I don't think it goes higher than 40% fan speed and, no matter how long I run it, temp bobs between 60-61C with the fans staying constant.

 

[Vario]

I'd return the card and get another.

 

[bestcommenteverzzz]

Hi all. Sorry for taking so long to respond. I've been busy recently but the problem has gone away for a while. Originally I had the 500-056 with a shitty 280w psu but I wanted to upgrade to a new video card so I got the gigabyte rx 460 GB OC Edition but to account for the needed watts, I also bought a corsair 430w bronze certified. The problem no longer displayed any distorted pixels and became more of a screen freeze for 4-5 seconds and then it would turn black for another 2-3 seconds and it would return to working normally again. I downloaded GPU-Z and CoreTemp to check everything in my computer and running games like bl2 and h1z1, the temperatures never went above 60C. I tested Furmark and it went up to 65C. This morning I was playing h1z1 when it happened again. Although this time, it did not recover. It went black and the monitor stated that my HDMI was inactive(told me to check the video cable and/or video source). I opened my case to take a look at the card and it was literally dead. Fans weren't spinning at all. I just left it open there for a bit and after about a minute, the fans started back up. After another minute, it went dead again and again it recovered. I'm thinking it might be my motherboard? It seems like the card isn't getting enough power but the thing is that the card doesn't have any PSU connectors. It runs merely off of the PCI-E slot from the motherboard. Would really appreciate any help. Thanks guys.


EDIT: Just took a look at System Logs. Thanks @FordGT90Concept . I found that there were repeated 4101 Warnings that said the display driver crashed and recovered successfully. I'm pretty sure this is it. I highlighted them in red boxes.

EDIT 2: According to Gigabyte's Website for the Rx 460, it involves active 3d Fan which says, "The semi-passive fans will remain off when the GPU is under a set loading or temperature for low power gaming. It allows gamers to enjoy gameplay in complete silence when the system is running light or idle." Does that mean the fans should be stopping? I'd appreciate any and all help.

 

[Vario]

First of all I wouldn't use Furmark, its bad for graphics cards. You can test instead using something like Unigen Valley benchmark.

Secondly, might be to update your drivers.
https://technet.microsoft.com/en-us/library/cc756372(v=ws.10).aspx
Resolve
Check Windows Update for a New Display Driver


Might be the HP motherboard doesn't provide enough power along the PCI-Express slot. Now that you have a 430W corsair, you could try a videocard with an external 6+2 power plug. At a certain point you might be better off getting a different non HP motherboard and migrating the CPU to it. That HP Pavilion uses a micro atx board.

Of course since you changed power supplies, double check the power plugs are plugged in correctly, both the 4 pin power plug near the CPU and the 24 pin plug on the right side of the motherboard.