2 cards dead, need some answers..

D

dchrsf

Gawd
Joined
Oct 12, 2005
Messages
842
I just got back my 9600xt from RMA (they sent me a new one) after the first one mysteriously died by first black screening requiring a reboot, then a total non-post. I had it for 2-3+ yrs with the AC VGA silencer on it for few months. It would spit out random beeps over the speaker when inserted.

Now, got the new one last friday and have been playing on it all weekend. Everything checked out fine with stress testing it, and it was flawless.

Today I went to turn on my computer and it just has a long beep, then nothing displays. I tried my trusty S3 PCI vid card, and it boots fine again. I also had the artic cooling VGA silencer installed both times when it died, but dont know if that caused it.

I'm leaning twards a possible bad mobo that could be frying the AGP cards I put in, or maybe im over thinking it and its just the replacement AC VGA cooler that is somehow messing it up.

I also noticed on both dead cards that the area right next to the blue VGA out (next to DVI) are extremely hot (burning) when I attempt to boot up and leave running after the long beep and nothing on screen. Didnt notice weather the area was hot when the card was working normally in game or on win desktop.

When I touch in and around the area where the blue circle is, the faint squeling (from the black metal thing with wire wrapped around it maybe?) dissapears, then reappears when i lift my finger.
 
I'd check the voltage rails of your PSU. Sounds like something's getting too much voltage (or too little).

Ryan
 
I thought about that too, and while I dont have a multimeter, the first card did die using a 350w Antec. The new card was run only on my new seasonic s12, so I think that rules it out.

Would it be possible that the motherboard could fry the graphics somehow? Thats the main question because I need to know wether I need to get a new CPU/Mobo/Vid combo, or just a new graphics card. I plan on just a X800GTO AGP, so I wont be using the same cooler if that is causing the cards to die.
 
does the cooler plug into the card itself, or the PSU? if it plug straight nito the card then that could very well be the problem. Try to find someone who knows how to test the fan for excess current draw.

Ryan
 
also, it could very well be a bad motherboard, but I wouldn't know how to go about testing it to see without frying more cards. that could get expensive. maybe you could find a cheap $10 AGP 4x card and test it to see if it dies like the other. make sure not to put that cooler on it. But if the mother board is frying AGP cards, it would seem that it'd also fry other things like PCI cards or the CPU.
 
Yes, the cooler plugs directly into the same plug the stock HS/F plugged into on the board. I also thought about buying some cheap AGP card and testing it, but i havent had a chance to see if any local stores have a card for cheap. You know Best Buy and others..they overprice the geforce4 at like 50 bucks..its just not worth it to use as a test card. Maybe I'll find one, who knows.. I'll report back.

I hope the board isnt frying the AGP cards, but im gonna get another cheap agp to try out soon before I buy a X800 anyways to make sure.

I will continue to research this, and I am always open to anyone else who has an idea. Feel free to post your constructive or random thoughts regarding the prob.

Thanks for the ideas Ryan,

David
 
For reference, my P4P800 Intel motherboard fried cards on me until I replaced it a couple of months ago. The only card that survived for an extended perdiod of time was a GF4 Ti4200.

Cards that have died the last 18 months:

Sparkle 6800GT x 2
Leadtek 6800GT
Sapphire X850XT

The first Sparkle died after a few months, so I get it replaced. The replacement card was DOA (or so I thought), so I get a Leadtek card. This card dies after about ~6 months, so I ditch it and get a X850XT, which lived for about 6 months before dying. I've used three different PSUs during the same time, so it's not that. Most likely my motherboard is frying the cards.
 
Ok, I have an update:

I'm going to purchase the Asus V9400-X (nVidia geForce 4 MX4000 AGP 8X 64MB DDR) for 25 bucks from zipzoomfly (cheapest modernish card I found), and should have it here before the weekend to test it out for a week or so while I make up my mind on a x800 varient and have them ship it out. Hopfully it stays stable and isnt like vlad_Swe's GF4 and stay alive when it shouldnt! I need it to die if its gonna die!

Honestly, I really would like to think its not the board, because I purchased it along with the current cpu from a very reputal guy regularing on DFI-Street. I don't think he would rip me off like that. Yes, it may be the board, and vlad_Swe may be right..but spending excess of 300 bucks for a new mobo cpu and video card right now isnt the cheapest method out. I'd rather be able to spend a few hundred now and at least have the card from the rest of the summer. It's funny. Every time a break from school starts I end up killing somthing somehow (spring break last year very first day- my XP 2700+ fried making me buy the current cpu and board). Now start of this summer, my 9600xt dies leaving me with another dead RMAed card 2 weeks later. I'd just like a summer of gaming (and working) you know? I'm sick of all this bullshit.

I'm going towards the aftermarket cooler installation as the culpret because whenever I installed the cooler, after a few months it would die. Temps werent the problem, and my installation was perfect making good contact with the core etc. Frankly I dont know what exactly it is but I will just go from here, then reselling the RMAed 9600xt and X800 if they too die, then buying a A64 system.

It seems the people who know the most about technology are the ones always getting screwed over by it.

Thanks for reading,
David
 
I think a multimeter would still help, if only to put worries to rest. What are the amperage ratings on that PSU?
 
Straight from the SPCR review: http://www.silentpcreview.com/article596-page1.html



I'll buy a multimeter one when I get my paycheck on tuesday along with the X800 if I have enought money, and test some voltage lines before I install the card, but it should check out ok, I mean Seasonic is a reputable brand and anyways I had a different unit when my first card went south.

Since that area of the card I circled in pictures got hot, im wondering if that's the voltage regulation area of the card, and it is recieving too much juice? Does anyone know what those chips and caps do around that area? Maybe I can pinpoint the cause of the problem.

Larger pics:

Top left (Capacitor and Black dotted thing): http://www.oc.com.tw/article/0401/imgs/ATi-9600XT-BF.jpg

Top right (Chip and Ceramic Capicitors): http://www.ixbt.com/video2/images/ref/r9600xt-scan-back.jpg
 
even tho that psu has 2 12v rails ... 8 and 14a's are still pretty lite.

Plug your after market cooler into regular molex connector instead of the agp card power header .. over working the card to try and supply power to heftier fan could be what killen the cards and combination of using lighter duty psu's

just an unedjubicated guess tho
:)

..also you might want to look into getting a UPS of somekind to help alleviate inconsistancies in your power in your home going to the computer if you havent already
 
ThreeDee said:
even tho that psu has 2 12v rails ... 8 and 14a's are still pretty lite.
Click to expand...
Erm... yeah, the 3200+ draws 73w max IIRC, and since the NF7-s draws off the 12v, that means it needs a whopping 6a to support that chip. Since that 9600XT is drawing completly off the 5v (and only then, a max of about 30w, so 6a from the 5v) I would say the spec of his power supply is just fine. I would say the second rail of that supply is going completly unused, unless he has like 10 HDDs.
 
You must log in or register to reply here.
Back
Top