Crashes on login to windows (7)

RobHu

RobHu

RobHu

Associate
Joined
29 Jan 2008
Posts
1,017
Hi,

For the last month or so my PC has been bluescreening a few times a day, usually when gaming. However in the last week this seems to have stopped and has been replaced by a worse issue: crashes on login to windows (the screen will either go black or be filled with a random repeating pattern).

At first this would happen once or twice before I could successfully login but now it happens every time. The only way to avoid it is to login to safe mode and then uninstall my graphics drivers, it then (sometimes) seems to work on reboot, for that reboot only.

I've tried:
Booting with no periphals etc plugged in
Reinstalling gfx drivers
A different graphics card
Format
Checking HD for bad sectors
Different RAM

Specs:
Q6600 @ 2.4 -was OC to 3.2 but reverted it to stock now
Abit ip35 mobo
GTX 280
4GB kingston HyperX 1066 DDR 2 RAM
Corsair 520w PSU
Seagate barracuda 250gb HD

Any help would be very appreciated :)
 
Does a linux boot disc do the same?

Id run memtest to highlight possible errors with mem controller. Apart from that you have covered most of the basis i would and I would check PSU and Mainboard next.
 
My first thought was a driver issue but it sounds more like hardware.

memtest is the logical course of action.

Might be worth just booting in to safe mode and logging in just to see what happens.
 
what were the bsod's associated with, nv4disp.dll or anything else nvidia sounding? Id say corrupt graphics driver/card/ram. If you completely removed the driver then reinstalled a different one, that rules that out, you tried another card so that rules that out, so it leave RAM.

I assume that the green light on the card is on, so as such its getting enough power and not overheating? CMOS reset it also a good idea, since its quicker than running 8 hours of memtest lol.
 
Memtest did'nt throw up any errors, however turning up the RAM voltage a bit seems to have fixed the problem. Why it stopped working at the voltage that it had been at for a year or so is beyond me though :confused:. I'm going to do some stress testing hope this is not just a temporary fix.
 
meh, doubt it will make a difference if you arent using lots of stuffs. Cant hurt though, the temps arent silly high on the NB or SB are they? everest can tell you.
 
Nah temps all round are ok - a bit high on the cpu maybe but that needs thermal paste reapplying and its not too high.

Just swapped out gfx cards again (to an 8800GT) and it booted ok. Not sure why it didnt the first time I changed them. Looks like a second 280 RMA in as many months - Any chance that it's the PSU killing them?
 
Last edited:
theres a chance, I had a cheapo (Q tec lol) kill 3 7950gt's before i remembered that my branded 580w had died and id popped that back in to tide me over, then upgraded the graphics as "i had some money lying around". Needless to say after it killed 3 i remembered the money was there to buy a PSU :)

But, yours is a decent PSU, and plenty enough to run what your running. Get some voltage monitoring with history, see if it spikes or anything while playing. It should only kill stuff if it is being strained and as suck sends too much voltage/amps through.

Could just be you got a dodgy one back on the RMA.
 
You must log in or register to reply here.
Back
Top Bottom