Help BSODs!

Hi All,

I'm really hoping someone else has dealt with something similar. I am pretty experienced with building and fixing PCs so I am stumped here.

My system has been blue screening since I started playing again this last week. Almost every BSOD has a different fault. From IRQL to PAGEFAULT to MEMORY_MANAGER etc. Every one is different.

I'm running on new hardware since the last time I played XIV which is as follows.

Asus M4A785TD-V EVO Motherboard
AMD Phenom x6 1090T
EVGA gForce 560
Corsair 750w PSU
8GB Gskill Ripjaws
500GB WD Caviar SATA 3
64GB OCZ SSD
Liquid cooled

So here is what I have done.

Fresh copy of Win 7 x64 installed along with updated drivers for everything.
Fresh copy of XIV installed.

2x Passes with Memetest86+ with no memory failures
1 Hour Prime95 test 0 faults
3 runs of 3DMark 11 no issues

PC runs stable and cool around 55c within prime95 so I am not dealing with CPU overheat nor is anything else overheating.

Anyone seen similar issues only running XIV?

Thanks ~ Syn

are you running in windowed or FS? are you running at max settings?

in my experience AMD does not cooperate well with nvidia cards, but that's more biased personal preference i think. it looks like all of your pieces are working well on their own, but when put together something weird is happening.

have you checked voltages?

I prefer furmark to 3dmark for stress testing. a lot of times i have no problems in 3dmark but will crash on furmark.

Currently I have been running windowed at 1920x1080 with most settings on standard to high. I know shadows was turned down and no DoF or Ambient Occlusion. Game runs at 50-60 FPS most of the time and it doesn't tax a single core over 50% and the net CPU usage is about 25%.

Come to think of it, I believe I was having the same issues 6 months ago before I upgraded. Perhaps it is voltages as you mentioned. I just bumped my RAM voltage up and will look at NB/SB voltage suggestions for this board.

I'm downloading PC Mark for the stress testing.

For stress testing I've noticed FFXIV does it better than any synthetic benchmark like furmark. I just have to go somewhere with trees/grass/flowers and the GPU heats up to 90C with full use. GTX560ti Hawk seems to have something against nature. Furmark gets barely to 80C if even that high.

Considering it's always different error, I would suspect HDD, CPU and RAM, as some failure in those can result in completely random errors. Graphics card errors rarely result in bsod. Of course, then there's PSU and MB that can cause the problems too, like voltage spikes etc. You don't need too big of a spike / undervoltage to get funny bits.

Can also try Linx for CPU stress testing, at least for me it's been better at stressing my cpu than prime95.

WOW 90c is pretty damn hot..eventhough gpu's are capable of withstanding this much heat...it's not recommended to play with a temp like this though ! I myself own Asus GTX560ti 2-way sli and they never reach above 65 in this game.( just started again since a few days ago tbh,but still ).Is there something wrong with your airflow perhaps ( I see you have liquid cooling,is your gpu also liquid cooled ? ) And wich current nvidia drivers are you using ? The one's after 301.42 gives me Bsod's in several games btw,i had to switch back to 301.42 to fix this.

Also what i tried when i started playing FF again was,disable Ingame AA and use FXAA enabled through Nvidia control panel.After this i noticed a Huge FPs boost,and i say this because it Might get some stress of your gpu.

Quote:

---

Originally Posted by SunTzuuu

WOW 90c is pretty damn hot..eventhough gpu's are capable of withstanding this much heat...it's not recommended to play with a temp like this though ! I myself own Asus GTX560ti 2-way sli and they never reach above 65 in this game.( just started again since a few days ago tbh,but still ).Is there something wrong with your airflow perhaps ( I see you have liquid cooling,is your gpu also liquid cooled ? ) And wich current nvidia drivers are you using ? The one's after 301.42 gives me Bsod's in several games btw,i had to switch back to 301.42 to fix this.

---

OP's system is liquid cooled, not mine. :) and yes, 90C is a lot, but I managed to undervolt my 560ti to lower the temps quite a bit, and keeping fps cap to 30 makes sure it won't go past 80C, if even that much. I believe the high temps are because 560ti Hawk is overclocked to 950MHz, nothing wrong with my airflow, or at least opening the case etc doesn't affect the temperature to one way or another. Dual display setting keeps the temps a bit higher but I can't live without second display anymore. :) But again, 90C is my temps, not OPs.

Quote:

---

Originally Posted by Synfrag

*Snip*
2x Passes with Memetest86+ with no memory failures
1 Hour Prime95 test 0 faults
3 runs of 3DMark 11 no issues

PC runs stable and cool around 55c within prime95 so I am not dealing with CPU overheat nor is anything else overheating.
Thanks ~ Syn

---

Memory tests operate with probabilities. Most memory fauls have a high probability of being detected within one or two passes, while other memory faults are harder to detect. The more passes you run, the higher the probability of finding intermittent and rare faults becomes. Two passes isn't really an in-depth test, it's more like a pretty basic superficial test. I would suggest running memtest for maybe 16 hours. Put it on before going to bed, let it run till you get home from school/work/whatever. Technically, one rare memory error could evade the first 5000 passes of memtest, while the same fault could be detected within 10 passes on a different machine which had the exact same fault. However unlikely it is, it isn't impossible.

Furthermore, memory related issues can actually be caused by the cache memory on the CPUs themselves. Even if the CPU isn't overheating, faulty cache memory on the CPU could run into problems if the rest of the CPU is under a decent amount of load. Generally, the CPU is not under much load while running memtest.

When I test my OCs, I generally only test my system for a couple of hours before I consider it "stable enough", but in reality, this is very, very sloppy :p. I guess I'm just a bit lazy. If I ran into instabilities at a later point, I would most likely test it for maybe a full 24 hours, unless i was able to detect the problem earlier than that.

2 things to check that hasn't been mentioned... HDD bad sectors, and vram

I think I have the same exact SSD as you, 64gb OCZ Solid 3? I got it for christmas last year, and the performance blew me away. I had zero issues with it, until after like 2-3 weeks and I started getting BSOD like how you describe. The rig at the time I had installed it in, was somewhat freshly built (as I made the transition from 775 to 1155 in oct of last year) and I was almost certain this was a PSU issue or my video card, my ram, my bios.....I tested it all until I found a few online guides that helped me to setup the SSD. I figured I could just install it into my case, slap windows on it like a HDD....wrong. After doing some tweaking, downloading and installing, I finally got the BSOD's to stop and haven't had any issues since.

I suggest this short guide:

http://thessdreview.com/ssd-guides/o...ation-guide-2/

(note: i haven't done all that they reccomend, but certain things like disabling unecssecary services, can slow the number of writes to your SSD, thus increasing lifespan of the drive).

Wouldn't hurt to check to see if you have the latest drivers and BIOS for your motherboard, here's the link:

http://www.asus.com/Motherboards/AMD..._EVO/#download

EVEN more info on your SSD:

http://www.ocztechnologyforum.com/fo...BC-for-OCZ-SSD

^Straight from ocz's site. Very comprehensive guide.

Good luck and god speed. /gcsalute

I use a ssd for O/S but have installed games on an other hdd. Same?
Definately check out VoltaAsura has to say and perhaps check out concerns with "the Hawk" and high-temps.

btw, Is that 1 8gig stick or 2 4gig'rs? and your PSU has how many 12v rails, delivering how many amps-per, compared to "the Hawks" needs?