Solved BSOD when gaming

[essenbe]

If you still have the first log you sent, go to toward the bottom of the page at 16:08:15 You will notice tour GPU load jumped to 100%, fan speed dropped, dedicated memory went from 2105 to 0, Dynamic memory went from 117 to 0 and the VDDC in and out dropped drastically about 65-70% ( Memory voltage). The voltage to your card seemed to remain constant, so that is good.

But, that would be fairly normal if the game crashed. So, it's is it the game or the card?

 

[demongorgan]

If you still have the first log you sent, go to toward the bottom of the page at 16:08:15 You will notice tour GPU load jumped to 100%, fan speed dropped, dedicated memory went from 2105 to 0, Dynamic memory went from 117 to 0 and the VDDC in and out dropped drastically about 65-70% ( Memory voltage). The voltage to your card seemed to remain constant, so that is good.

But, that would be fairly normal if the game crashed. So, it's is it the game or the card?

sorry for the late reply. i dont have the original log anymore, so will take your word for it

it crashes on dayz standalone, dayz mod, elder scrolls online (once) titanfall (few times)

 

[demongorgan]

i would say the game, but its more than one game doing it, so the question is, is it the board or the card do you think?

*edit*
somtimes it says "display driver has stopped working and has recoverd" other times it says "dayz.exe has stopped working"

 

[essenbe]

Do you have another card you could try, or is there a friend that would help you out and switch cards with you? I think that would answer the question once and for all. If your card works in your friend's machine and his card crashes in yours, I would call it the board, which is what I suspect anyway. If your card crashes in your friend's machine but his works in yours, it would show it's the card. Just make sure your friend's machine has the PSU to run the card.


You could also try lowering your settings drastically to see if it still crashes. I'm sure you have everything on Ultra, try lowering the settings to medium and run the game.

 

[demongorgan]

hi mate

i tried running dayz on medium high, lasts longer but still crashes. i asked my work college/mate if i could bring my pc round his and try his gpu in my machine and his in mine. but he's a uber nivida intel fanboy and refused to let me use his gpu because i have an amd system... well there goes his future tech help as i wont be helping him with his machine no more (every other week) an i dont have any friends who are in to pc gaming/building

i think i have an old nvida 8800gs in my cupboard somewhere although i don't think that would help as i doubt it could run any of my games

 

[demongorgan]

strike that.. i just found a boxed HD6450.. that should do the trick

 

[essenbe]

It will work, unless it's the PSU. The 6450 won't pull near as much as yours does. But, I would give it a try. Just be sure to download the drivers for the 6450 and uninstall your drivers before you make the switch.

 

[demongorgan]

yeah hopefully that card will crash to then i know for sure its the mobo. the cx750m should be ok for running an r9 290 and its brand new, not sure of the odds of buying two busted psu's in a row from seperate stores lol..

cpu was ok, ram was ok, i think psu ok, so if the pc dont crash its gotta be the gpu right, and vice versa?

 

[essenbe]

I would think so.

 

[demongorgan]

ok i can replicate the problem when i run prime95 (small fft) and furmark (1366x768 extreme burnin) as i figured it would be just like when i play my games full out..

first time the pc reset itself complete with a dump file, the second time it lasted 3min then display driver stopped working.

coretemp said cpu never went over 38c

 

[essenbe]

Is this with the 6450? Are you running both Prime and Furmark at the same time. If you are, run just Furmark or Prime. I know it's the GPU crashing but Prime is for mostly the CPU and Furmark is for the GPU. However, if Prime is the cause, it can cause anything to happen. It would just help to separate them.


I'm not sure what to make of the GPUz Log. If you will look at 21:12:58 which is the first instance, Your core clock dropped from 99-100% down to 1, The power (12V) from the PSU is slightly low, but within tolerances, but remains pretty constant (11.75-11.63) and never drops. Note that below 11.4 is out of specs. Your VDDC In and out, rated in Watts (W), Amps (A) and Volts (V) drop as well. VDDC is basically the cards memory voltage. It was similar the time before. If I hadn't been working with you on this for a while, I would say it's the card. But, I can't see how it can be two cards in a row. The PSU is a little weak, but putting out enough voltage and it never drops, so the card seems to be getting the voltage it needs. All the temps seem to be fine.

 

[demongorgan]

Thanks for the reply and sticking at this with me appreciate it man.

Havnt installed the 6450 yet will do it when I get home in a bit and give the pc another round of testing. I'm running me86 again whilst I'm out just to tripplr check everything.

I ran both prime and furmark together as I figured it would simulate playing a game without having to wait for them to load up, watch intros then wait for server lists, join servers etc. As a few people said they ran them together to test stability within a system.

When I put the 6450 in I will just run furmark on its own. Or that one you linked me to, as its a little more interesting than watching a doughnut of fur

What I don't understand is that when I run just the furmark or the one you linked me the GPU will happily sit there and take it, poodling along happily until I end the program. And prime95, my CPU and ram will happily poodle along for as long as I let it. Yet when I run them together poof! Just like in games. That's the only time the PC misbehaves it seems is when components are working together yet on their own
Their fine. Never had such a trouble like I've has with this one

 

[essenbe]

Well, first of all with both Furmark or Prime, your computer will never see that much stress in real life. No game will put your card under as much stress. Not even close. You may find this interesting even though it is an older article, I don't believe anything has changed. It is about the GTX 580, but note in the beginning it says AMD has done the same thing http://www.geeks3d.com/20101109/nvidia-geforce-gtx-580-the-anti-furmark-dx11-card/. The whole idea of stress testing is to push your hardware well beyond anything it will see in real life. prime will push your computer and CPU to it's limits and run your CPU as hard as it is possible for it to run. I tend to be 'old school' about most things, but not this. The old school philosophy says if you can't run Prime for 24 hours straight with no WHEA errors (which many don't even know about) you cannot say your computer is stable. I'm not a believer in that, although I do run it, but not for 24 hours, for one simple reason. I have run it for 24 hours before without any errors and crashed as soon as I started some game or stressful application. I don't believe I have Furmark on my computer, unless it is for these forums to show people who do use it, how it should look. But, I never run it myself. I run Heaven. I am an overclocker and run a lot of benchmarks, at times. So, I am concerned with stability, but more concerned with real world stability. As you have seen, you can run Furmark with no problems, but crash in games. So, what does that show about stability? Not very much really. Having said that, I do run a lot of different stress tests, especially when overclocking, but don't rely on them as much as others. It is more important to me that I can run BF4 with no problems than if I can pass Prime or a GPU stress test.
If you are overclocking anything, set everything back to defaults. As stated above, I am not opposed to overclocking if done the right way. But, I have done it long enough to know that if you are not stable at stock, you will never be stable with any overclock.


Don't be concerned with me sticking with you. I'll be here as long as you need me. I just ask that you stick with me and do as I ask. If I get into an area I'm not familiar with, there are some real experts on this forum in many areas that I will ask to come in and help.

 

[demongorgan]

Appreciate the in depth response and I will have a look at that article when I get in shortly. I was tempted to pick up another mobo whilst I was out. Buy maplins are so damnes expensive..

This new build has never been oc through the mono or through aftermarket GPU software. I was waiting until I get the sabertooth and my H100i cl cooler. So everything has been left at stock until upgrades.

I have an old PC running vista with ddr2 3gb and and old x64 CPU. (Unsure of the mono until I'm home but it does have a pci-e port so might try it in that if its compatable

 

[essenbe]

My guess is the older rig won't have the PSU to handle the 290. Personally, I think you should get the new Mobo as soon as you can. I believe you have several places in the UK which have decent prices. I have a friend in the UK I can ask. I believe Aria is one of them though. your bad PSU could very well have damaged the mobo and the card as well. It most likely damaged something, and the mobo would be my first suspect.

 

[essenbe]

Try some of these places

http://www.aria.co.uk/

http://www.overclockers.co.uk/

http://www.scan.co.uk/

http://www.ebuyer.com/

 

[demongorgan]

hi mate, yesh i normally use aria as i can get trade disont when i buy in bulk and they always price match against all the companies you listed. so i shall stick with them.

i installed the 6450 and played the games that normally crash me.. and they haven't crashed yet, also in gpuz it now shows pci-ex16 @2.0 rather than 1.1. do u think the weight of the r9 290 could be pulling it out of the pcie-e slot?? will try installing it again and see what happens.

even tried playing the games at 1080p on ultra settings, just to see if i could get the card to crash and it wouldnt. the cx750m is a compatible psu for the r9 290 so i cant see it being inadequate power requirements right?

*edit*
just reinstalled the r9 290, and in gpu-z its showing [email protected] again until i start render then it goes to 2.0. have power set to high performance mode so i dunno.

will try my games again to see what happens

 

[essenbe]

It showing 1.1 in GPUZ is not uncommon when idling on the desktop. That's why the render button is next to it. I seriously doubt the 290's weight is the problem, or shouldn't be. I have an EVGD GTX 780 Classified, which is a huge card and it does not bother mine.

I'm a believer in Corsair power supplies and have about 4 of them myself right now. 750W is plenty to run your card. That being said, if the PSU is not malfunctioning it is enough power. It does happen occasionally. When diagnosing computer problems, you have to take everything into consideration.

 

[demongorgan]

happened again. wasnt gaming just sitting on decktop:

computer name: M3TAL-PC
windows version: Windows 7 Service Pack 1, 6.1, build: 7601
windows dir: C:\Windows
Hardware: GA-78LMT-USB3, Gigabyte Technology Co., Ltd.
CPU: AuthenticAMD AMD FX(tm)-4350 Quad-Core Processor AMD586, level: 21
4 logical processors, active mask: 15
RAM: 4292386816 total
VM: 2147352576, free: 1909829632




--------------------------------------------------------------------------------
Crash Dump Analysis
--------------------------------------------------------------------------------

Crash dump directory: C:\Windows\Minidump

Crash dumps are enabled on your computer.

On Thu 22/05/2014 14:17:41 GMT your computer crashed
crash dump file: C:\Windows\Minidump\052214-45583-01.dmp
This was probably caused by the following module: ntoskrnl.exe (nt+0x4ADB3C)
Bugcheck code: 0x124 (0x0, 0xFFFFFA8004CC98F8, 0x0, 0x0)
Error: WHEA_UNCORRECTABLE_ERROR
file path: C:\Windows\system32\ntoskrnl.exe
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: NT Kernel & System
Bug check description: This bug check indicates that a fatal hardware error has occurred. This bug check uses the error data that is provided by the Windows Hardware Error Architecture (WHEA).
This is likely to be caused by a hardware problem problem. This problem might be caused by a thermal issue.
The crash took place in the Windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time.

 

[essenbe]

It that with the 290 back in the computer? If so, I would under any circumstances change the Mobo. But, it sounds more like the card or power Supply. I would tend to lean more towards the card. What makes it difficult to determine is it ran fine with the 6450 installed. But, the 6450 is an older card and probably pulls less than half the power of the 290.