WoW Crashing Computer w/ Black Screen

**Ciderhelm** · #1 10-04-2008, 08:26 PM

I've been dealing with a problem that started about two weeks ago that coincided with a minor change in hardware (added secondary hard drive; WoW is not running off it) and an Nvidia driver update.

The Problem

About a third of the times I log in WoW, my computer screen will black out and sound will cut out shortly after within 5-10 minutes of play time. This has happened on my Warrior in Nagrand as well as my Druid in Stormwind, as well as various other places.

When my computer screen blacks out, my monitors give me a message saying they are not receiving a signal. However, my computer sounds otherwise fine.

The issue does not appear to affect the Wrath beta client; however, it doesn't always affect my live client, so it may be coincidence.

My issue is similar to these support tickets, which have not been resolved:
World of Warcraft (English) Forums -> The "Black Screen" Problem

World of Warcraft (English) Forums -> Black Screen

System Specs
Operating System: Windows Vista™ Ultimate (6.0, Build 6001) Service Pack 1 (6001.vistasp1_gdr.080425-1930)
Language: English (Regional Setting: English)
System Manufacturer: Gigabyte Technology Co., Ltd.
System Model: P35-DS3L
BIOS: Award Modular BIOS v6.00PG
Processor: Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz (2 CPUs), ~2.3GHz
Memory: 4094MB RAM
Page File: 2461MB used, 5954MB available
Windows Dir: C:\Windows
DirectX Version: DirectX 10
DX Setup Parameters: Not found
DxDiag Version: 6.00.6001.18000 32bit Unicode

Card name: NVIDIA GeForce 8800 GT
Manufacturer: NVIDIA
Chip type: GeForce 8800 GT
DAC type: Integrated RAMDAC
Device Key: Enum\PCI\VEN_10DE&DEV_0611&SUBSYS_34681458&REV_A2
Display Memory: 2287 MB
Dedicated Memory: 495 MB
Shared Memory: 1791 MB
Current Mode: 1920 x 1200 (32 bit) (60Hz)
Monitor: Generic PnP Monitor
Driver Name: nvd3dumx.dll,nvd3dum,nvwgf2umx.dll, nvwgf2um
Driver Version: 7.15.0011.7813 (English)
DDI Version: 10
Driver Attributes: Final Retail

**Satrina** · #2 10-04-2008, 08:37 PM

- How big is your power supply?
- Does the problem persist if you remove the second hard drive?

**Ciderhelm** · #3 10-04-2008, 08:44 PM

Will find the answer to the first and test the second. I know the guy I had talked with at PC Club had brought up that I could potentially need a new power supply.

Lizana · #4 10-04-2008, 09:34 PM

Sounds like one of two problems.

Problem one Heat issue
Solution - Take off the side of your case, blow a box fan into it. If it still continues to happen its not a heat issue.

Problem two - Power supply does have enough free volts. If its less than a 450 power supply methinks this would be your issue.

Now as a follow up question, Does this happen in other 3d games, like War, or things like 3dmark?

Oh and to see about your power supply, it should say on the sticker of PSU.

Evelaula · #5 10-04-2008, 10:01 PM

Just a quick note on one of my pet peeves. The wattage of the PSU doesnt matter so much as the quality of it. A High quality 500 watt psu will beat a rubbish chinese 750 watt one anyday.

My 350 watt shuttle psu runs a HD4850, 3 hard drives 2 gigs of ram and a Dual Core AMD just fine.

Could just be a crappy power supply and under high loads the 12v rail is dropping down under tolerances.

Lizana · #6 10-04-2008, 10:04 PM

yes, Quality does matter, but if hes running a 350 watt PSU and his components are drawing 600 watt, it doesnt matter how good of a quality PSU it is, it will still fail. Watts are not the end all be all, but thats like saying because your running high quailty RAM 512 is enough to run WoW in a populated area, Sometimes bigger is better, and PSU's are one of those times

BTW his vid card alone uses 165 watts. 2 hd's 50-100 watts 50 watts for the mobo 75 watts -100 watts for ram - ect ect

Now true these use different rails, but an underwatt psu is just as bad imo as a inefficient high watt unit. Always better to have more power to spare than to be power starved. ( And can still be a heat issue unrelated to power)

Evelaula · #7 10-05-2008, 12:01 AM

Source: Lizana
yes, Quality does matter, but if hes running a 350 watt PSU and his components are drawing 600 watt, it doesnt matter how good of a quality PSU it is, it will still fail. Watts are not the end all be all, but thats like saying because your running high quailty RAM 512 is enough to run WoW in a populated area, Sometimes bigger is better, and PSU's are one of those times

BTW his vid card alone uses 165 watts. 2 hd's 50-100 watts 50 watts for the mobo 75 watts -100 watts for ram - ect ect

Now true these use different rails, but an underwatt psu is just as bad imo as a inefficient high watt unit. Always better to have more power to spare than to be power starved. ( And can still be a heat issue unrelated to power)

Couple of your figures seem slightly off, hard drives for instance are suprisingly efficient and once spun up seeking rarly use more than 10-15 watts each. And plenty of intel motherboards have quite high power drain due to nb/sb (excluding cpu)

Reviews of the 8800gt measure power consumption report figures between 200 - 260 watts and thats for the total system so the card alone would be significantly.

Ram is also low power consumption or we would have horrible thermal issues and limitations. Tested figures are 1-3W range per 512MB stick.

Not really that relevent anyway. I've seen plenty of systems which have issues due flaky power supplies so cider may want to try and borrow another one just to test with.

Honestly though it seems unlikly that a hard drive would tip the balance of the power supply like that and start causing stability problems. Removal of the hard drive should quickly prove its lack of involvement in the crashes (or its cause).

Seems more likely to be a driver issue and as it occurred after new video drivers, and you say its not showing in beta. Your beta install is probably a much newer more recent installation and its pluasable that the driver update rooted your live isntall. Is it perhaps a reinstall of wow time?

**Satrina** · #8 10-05-2008, 06:23 AM

It is not uncommon for quality manufacturers to market a PSU at a lower wattage rating than it actually can sustain (via "tolerance margins") to guarantee performance in the labelled range. I'd wager that your 350 is actually quite happy to push 500 sustained.

Source: Evelaula
Honestly though it seems unlikly that a hard drive would tip the balance of the power supply like that and start causing stability problems. Removal of the hard drive should quickly prove its lack of involvement in the crashes (or its cause).

You'd be surprised. I have a 4x32A rail PSU in my machine, two of those rails are dedicated to PCI-E, and the others to everything else. When I upgraded to a 280GTX from an 8800GTS, it destabilised the system to the point where I had to change from DDR3 to DDR2 (ASUS P45 board with sockets for both) to keep WoW from crashing every 5-10 minutes. I couldn't stabilise the memory at 1.9V (it's rated for 1.7V) and the manufacturer warned me that trying 2.0V would be a good bet to fry it. It's sometimes like feng shui for computers. I've seen a few of those balance point problems over the years. Not common, but definitely not rare. The 15W drop from a hard drive is a better bet than memory to destabilise on a crappy or dying PSU.

I'm actually more worried that his PSU is dying than its output rating or quality, so I'm in agreement with Evelaula there. People will spend 500 bucks on a video card, then turn around and will not spend more than 40 bucks on the PSU. Good times to be had there, but even a great PSU will die at some point. I've had many that didn't make it past 6 months!

You don't really need to reinstall WoW for almost anything short of bad sectors. Delete the cache folder, copy the Interface and WTF folders off somewhere else, then run repair. Log in and see how it goes. You can then put you addons back, and selectively copy back macros and addon saved variables. Blizz TS just tells people to reinstall because most people haven't got a clue when you get to the selective parts.

**Cedix** · #9 10-06-2008, 02:13 AM

It would be good to know what PSU he is using

The important thing about PSU is not only the Watts it put out, but the efficiency it can deliver those Watts (80% is best) After that, the A of the rails (mostly the 12v one's) is important (even more for big video card)

PSU like Seasonic or Corsair are sure quality product.

For the problem,I would also look into the driver, as you said you changed them. I had many trouble with NVIDIA drivers in the past. On my last PC build, i had trouble with drivers on the SLI setup and only while running video playback (no crash under heavy game load like Crysis) reverted back to older driver fixed the problem.

While another HD can the reason for PSU failure, it look more like to me a video driver issue

**Ciderhelm** · #10 10-06-2008, 11:58 AM

Thermaltake 430 (TR2-430NP)

Haven't had the black screen since turning off the secondary HD, but that's not a guarantee that it's gone.

What is a recommended wattage? 500? 600? I'm looking at Thermaltake and Raidmax power supplies on Newegg.

**Ciderhelm** · #11 10-06-2008, 12:09 PM

Newegg.com - Power never felt so GOOD

What I'm looking at.

**klor** · #12 10-06-2008, 12:12 PM

I just installed the new drivers for my 8500 recently and had some problems with it and ended up having to roll them back(this was three days ago, roughly). You may wanna check that too. Could have been a bad driver pack. Mine wasnt black screening and crashing my pc, but alot of weird things were happening like screen refreshing being waaaay off.

**Satrina** · #13 10-06-2008, 12:35 PM

How long have you run without the second drive? Also, a driver rollback may not be a bad idea for testing with the other drive put back in. Be thorough before spending money =)

That said...

Thermaltake, Antec, Zalman are all middle of the road, which doesn't mean they're bad by any means, but the tolerance margins on voltages and rails are a little wider, etc. If you want to take the next step, look at PC Power and Cooling, Enermax, Corsair. Personally, I am long done with Thermaltake and Antec, as they've been the worst culprits for up and dying after 6-8 months for me.

That PCP&C 750 at the top left would be a good choice, though some people shy away from single rail boxes.

A friend of mine picked up that exact Corsair box and likes it with 4 hard drives, a 260GTX and a brand new Core 2 Duo CPU. Another friend has a RaidMax 750 (maybe that exact one, hard to tell) running SLI and has no troubles.

If you really like multiple rails, that Zalman 850 is pretty nice, with 2 dedicated higher amperage rails for the PCI-E connections.

You'd probably be just fine on 500-600W. Going bigger is like insurance against upgrades later on.

**Cedix** · #14 10-07-2008, 02:31 AM

Before buying, you need to roll back your video drivers, make sure you use drivers cleaner and the ATI tool to remove video drivers.

Then plug your HD back in and test, if you still get crash, then thing about a new PSU

I would stay clear from non 80+ certified PSU and avoid name like themarlatake, antec and the like. You do not need anything as big a 750 for your gear.

As Satrina said, 500-600 of good PSU will do you good. But that Zalman is just overkill... unless your planning big SLI video upgrade

I use a Corsair HX620W (modular) for my SLI setup all running on water and nicely overclocked (E8500, 2x 8800GT, SIIE, 4g DDR3 1600,Raptor's, CPU, MB and video Water) and no issue since day one

**Ciderhelm** · #15 10-07-2008, 06:51 PM

With the second HD online (and being used), I have had zero problems running full burn with fraps in WotLK Beta. I'm tempted to believe it's a driver issue, and will attempt a driver rollback.

That said, I've known for a while that my power supply was below par, so I went ahead and picked up the raidmax. If that happens to be the solution, that'll be good.

**Warwench** · #16 10-07-2008, 09:45 PM

in doing a rollback of the driver, I'd suggest this --> Driver Cleaner Professional download from Guru3D.com

run it, reboot. install new driver.

**Dismember** · #17 10-13-2008, 08:00 AM

Sounds like a classic power supply issue. I had exactly the same thing happen to me when I went from my 8800GTX to my GTX 280. And, it didn't happen with the most graphically demanding games, either. WoW, Mass Effect, and NWN2 would all black screen my computer after a few minutes of playing, but AoC and Crysis never had any issues. Initially I thought it was a driver issue, but updates and rollbacks had no effect.

I finally replaced my 550W PSU with a 1000W and the problem disappeared.

**Ciderhelm** · #18 10-13-2008, 05:28 PM

Btw I can't report back on this issue for another few days. I've got the power supply next to me, and am ready to roll back drivers, but I'm just a bit too busy w/ this site to try either.

Skrymir · #19 10-27-2009, 03:54 PM

Any luck figuring this out? I have the same issue. It has followed me to new video cards, even between computers now.