[asyikarea51]

It's been bugging me for over 7 hours now to no end.

Yesterday I loaded up some game to try out a mod, but the mod didn't work and the game quit. About half a minute later I'm back at the desktop... then the whole thing just FREEZES.

Obviously the first measure I take is to force a restart... then Windows loads before promptly BSODding on me with the dreaded nv4_disp.dll error, so I have no choice but to force restart again. But alas... BLANK SCREEN on the monitor!!! The mobo and everything else gets past POST fine, but I can't see crap. From the way the computer behaves... it somehow just keeps restarting again and again at some certain point after POST and going into Windows...

Playing around with the PCIE connectors and several restarts later I manage to get the gfx card working, but while loading Windows it BSODs again, this time on a different file (also related to the gfx driver)!!!

After the second BSOD I played around with weird RAM combinations and shifting the card to two other PCIE slots on the board, but this time I'm not so lucky; I haven't seen anything proper on the monitor appearing for the past 7+ hours...

I should add that, on the second (or both??) BSOD(s), there were these not-so-visible green boxes dotting the blue screen... and Windows mysteriously boots up okay from time to time if I don't use the main PCIE slot... but not much point if I can't see anything. But now none of the PCIE slots work at all in getting the OS to boot, blank screen or not...

The card in question is a well-worn single 8600GT.


------

Sounds fairly sudden I know... but I'll be in trouble if I can't get my main comp up and running again, + I'm not able to access the few folders I have over the network either. At the moment I'm on alternative hardware but the files here are out of date much...

Anyways, any ideas or help? This is the first major failure I haven't been able to solve by myself btw.

[Liberator]

First thing to do is ensure that the card is seated firmly in the slot.  Second thing is try a different card and see if it's the video card or the motherboard.  If it's the motherboard you're screwed.

If you are able to get it to boot properly, download ATITool and see if the card is overheating or malfunctioning in some way.

[asyikarea51]

If "firmly" equals to no exposed golden fingers, then yes.

I doubt it's the mobo, and more to the videocard (esp with the Windows BSODs on driver files... but OS BSODs actually screwing up my BIOS?!?!? /shock horror)... but playing a guessing game with screenshots on the interwebz while in the BIOS menu didn't help either.

I had hoped maybe the mobo setting was just to load PCI slot graphics and I could just change the option to PCI Express, but no change on reboot.

Trying to boot the comp again atm... I have nothing to work on except for guessing what it's doing.

It seems to boot to Windows, but the monitor continues to show absolutely nothing from POST to end of OS load... The DVI cables are connected correctly because the monitor can detect the connection.

[Liberator]

Then most likely the card is dead, but you won't know till you get another card to test in that slot.  If you have an old one, great, if not you may have to purchase one.

A word of warning if you have to purchase a new card to test with, buy the absolute cheapest card you can get, performance is a non-issue at this point.  All we wanna know if if the card is bad and if it is, was it the motherboard that fried it.  This is because if the mobo is frying cards, you don't wanna blow $200 on a card and have it blow up in your face because the slot is bad.

While you are in there, check your PSU connections as well.  If they came loose or got jimmied up some how, they they could have shorted and BBQ the card.  Also, check for swollen or burst capacitors.

[captain-custard]

try starting your computer in safe mode , this will not load the drivers for the card that way you an eliminate that the drivers have turned to mush , if this works then try and do a system restore this will hopefuly take you back to a point before the mess became messy

if none of this works then , go with the first idea that your card is now officialy retired

[Rodo]

might sound stupid but I had a similar problem a couple of months back, I was not getting vid out... I unplugged the PCI-E and then when I put it again in the slot I completely forgot to plug the extra power connector, so it was natural to not get any video.

Just check that to be sure, if it doesn't work after that It might be your power supply fried and it's not working properly.

[asyikarea51]

forgive the fast reply since i don't have any breathing room atm for the next couple of days or even longer.

getting a new card just for the sake of testing, i personally don't mind it at all, but damn it will take forever for me to even save a small amount. it'll be the biggest bottleneck in troubleshooting.

safe mode doesn't work either... if i don't see anything on the monitor on POST, i dun expect to see anything at a higher os level...

ironically, during the black screen just after the first bsod, i booted to windows completely blind and shut it down properly and when i rebooted, the card was working again although i'd shifted it to another slot... but then when it was working and i booted to windows, it bsodded on a different file (ie not nv4_disp) and since then it's all black screen, even after removing battery, reset cmos, blindly changing bios settings, swapping between all available PCIE slots, etc...

the 8600gt i use has no power connector of its own, it runs purely on main power...

btw come to think of it... when i dc'd some of the parts before putting them in one at the time, i'd press the power button to start up the com, but the psu would just behave like a kickstarter... VERY ODD... it'd only start in 1 shot if i'd connected enough components to the mobo... else its kickstart frenzy... lol...

i rapidly press the power button to kickstart the thing, and it'd either fail to start even after a "top DPS attempt" or... it'd somehow start when it gets enough juice...???!?!?? does that explain potential psu issues? but it still works perfectly fine in booting the os even when "blinded" and with whatever stuff i have plugged in to the mobo and psu... i just feel strongly that it does, with that little blinking light showing loads of activity and the fan noise pattern...

not half bad imho considering this is the first time the com screwed up this bad. close to two years trouble-free...

mobo capacitors are all solid, gfx card is still electrolytic or however it was called... power cables i wun mention it at this time, out of time atm

[Liberator]

The PSU sounds like a kick starter...something is shorted...seriously DO NOT use that power supply again until you have it tested.  Your best bet at this point is to get this to a local technician and let him or her take over.  This is sounding more serious than simple hardware misconnection or software failure.

[IceFire]

Its either the PSU or the GPU or possibly both from what you've described.

Its somewhat well known that the 8000 series of nVidia cards (except the 8800s) were made to a rather lower standard of manufacture.  Lots of the motherboard versions were dying within the warranty period and many of the others were dying just outside of it mostly due to the more extreme heat cycling that notebooks go through.

Of course the death of your video card could be precipitated by your PSU starting to die and is providing inconsistent levels of power.  My experiences with power supplies and video cards and the random things that can sometimes happen suggest that anything to do with power and sensitive electronics can be really "funky".

If you want to tweak it yourself (and have the cash to solve it if it goes wrong) then by all means....but if not then taking the system to a trusted technician sounds like the next best bet.

[asyikarea51]

i'm not sure if it matters but this mobo has some sort of power control... meh. this is really my first comp from the newer core 2 line, everything else i used to date is actually within the P4 generation, or even older.

initially when i got this board with a different PSU, the PSU wouldn't supply the power - refused to boot. brand new btw, heck they tried more than 1 example. the fellas at the shop, they got fed up and just swapped the psu i bought for a different brand of the same wattage, and that one worked. been in my case ever since.

kinda odd. the psu i'm using should be rated for far more power than all the components i connected (the shop had a 430W and mine was way higher).

i didn't finish the build but i already had the psu and a comp to work with at the time, so i just used it.

(i know there's more to PSUs than just wattage but i'm not able to bring anymore info to the table).


lol... less components connected = kickstart. all my mobo-connected devices seated in their proper places = one-shot power on. but just saying that im not sure if it's a psu problem or if it's the mobo playing defensive. might as well try and figure out something to do over the coming week.

[Liberator]

What we're saying is that most likely the PSU is defective or broken and by being in that condition you could have very easily fried the motherboard and the video card.  Even if the mobo is in working order, shorting the PSU like that isn't helping it.  It needs more help than can be given over a message board, you need to carry it to a professional, even if they off of work and you know they personally.

[asyikarea51]

Been too long since this thread, I know - the computer's date for some reason went all the way back to MARCH.

I turned on the computer today at random and the GFX card finally output something onto the monitor after so long...

Boot into safe mode, removed old driver, worked ok, shutdown. This was with the card in the bottom-most PCI Express slot on the board.
Shifted card back to main slot, booted again in safe mode - ok, rebooted back to Windows (there was this strange pattern of dots all over the screen just as the computer exited the boot screen and prepared to go into Windows itself).

So that looked... "okay"... I know it's not okay with all those dots, but what the heck. Went to install the 185 driver as present at the nvidia site, install...

EWWWW UGLY BLACK LINES ALL ACROSS THE SCREEN... and the install program just plain stopped responding so I killed the process with task manager. Not to mention everything on the display was running faster with "no driver" than with the driver installed, even if the install was only halfway through...

Successfully shut down the computer, tried to boot again but now the GFX card has decided to die again.

Any ideas? As it is now I'm thinking it's just the GFX card that's done, and not the main board (these are solid capacitors, come on ) or the power supply...

[Liberator]

Again, the only real way to test this is with a replacement video card or taking it to a shop to be tested professionally.

[asyikarea51]

True... that I know.

Well at least I got to bicker around with it, transfer a file out... freaky black lines.

[jr2]

Boot into safe mode, removed old driver, worked ok, shutdown. This was with the card in the bottom-most PCI Express slot on the board.
Shifted card back to main slot, booted again in safe mode - ok, rebooted back to Windows (there was this strange pattern of dots all over the screen just as the computer exited the boot screen and prepared to go into Windows itself).

* jr2 thinks you may want to leave that card in the bottom slot just to see if it's your PCI-E slot?  could be.  And if you have compressed air, clean all of your slots.. you could use a vacuum you just have to be careful it doesn't latch onto anything and suck it off or if its metal that it doesn't touch.  Dust is your computer's enemy.

[asyikarea51]

I shifted it back after the card died again in the main slot and nope, nothing appeared on the display... XD

Apart from Liberator's suggestions (which I can't do until I save enough for at least a new card) there's nothing much I can do right now really.