PDA

View Full Version : Long term stability @ 1.4v


Pages : [1] 2 3

anzak
10-10-04, 09:33 PM
As many of you know I have been running my GT with Ultra BIOS for some time (I was one of the first here at nVnews to try it). Just moments ago my monitor went black and I got the check signal message. I was just surfing the nvnews forums. I quickly shut off my computer and started it up again... no signal. Nothing, the computer starts but the graphics card was not working. So, scared ****less, I remove the card to look for physical damage. It looked alright so I put it back in and bam it worked. Could this be the long term effects of running 1.4v? My temps are well under control (48c idle and 61c load). I plan to flash it back to the original BIOS after posting this.

anzak
10-10-04, 09:59 PM
Why did I post this in the Forceware forum? Could a mod please move it to the nvidia graphics forum. Thanks.

I flashed the card back to 1.3v. Im still worried though. What could cause something like that to happen?

Raptorman
10-10-04, 10:44 PM
I got that once and it was because a connector had worked loose on the card. Make sure they're in nice and firm.

Rytr
10-10-04, 10:53 PM
Why did I post this in the Forceware forum? Could a mod please move it to the nvidia graphics forum. Thanks.

I flashed the card back to 1.3v. Im still worried though. What could cause something like that to happen?

Something like what Raptorman experienced, a loss of power. I think you made the right move for now. It may have nothing to do with it but if it did others will start to experience the same problem. If not, maybe it was just a loose connection...you can always flash it again.

anzak
10-10-04, 11:50 PM
Something like what Raptorman experienced, a loss of power. I think you made the right move for now. It may have nothing to do with it but if it did others will start to experience the same problem. If not, maybe it was just a loose connection...you can always flash it again.

Thanks for moving the thread!

Great now im experiencing blue flashing lines. Could this really happen to me again! My first BFG GT died but it was still under warranty! I am now starting to believe this is more of an issue with my specific card. If it dies (which I know it is with the BLUE FLASHING LINES) im done with After market coolers, volt modding, BIOS flashes, and even overclocking! Chris Ray and I may both be SOL.

hypertek
10-11-04, 05:05 AM
hey what ultra bios did you use? my bfg doesnt take too well to ultra bios nor gt bios modded to 1.4v..... so right now i put stock bios .

Warden13
10-11-04, 07:44 AM
If you are using an Ultra Bios I would suggest stopping. There are a few differences in the Ultra and GT cards themselves that could cause issues. Ultra's have 2 speed fans and likely have different memory timings. I also think they have dual DVI whereas the GT does not. I had issues with my card when using an Ultra Bios but have had ZERO issues when using a modded GT bios to run at 1.4v and 460/1200.

Warden

Marcos
10-11-04, 07:49 AM
6800u has 2 power connectors for a reason. now i dont think ill be getting that nv5 cooler thingy and I wont be messin w/ my ****.

wickfut
10-11-04, 09:27 AM
doubt it has anything to do with the voltage - my leadtek GT runs stock at 1.4 volts for 3d

anzak
10-11-04, 10:47 AM
doubt it has anything to do with the voltage - my leadtek GT runs stock at 1.4 volts for 3d

How long have you been running your GT with 1.4v BIOS? The title of thread is long term stability. I have had mine flashed for months now without any issues. It's not just Ultra BIOS, I have tried modded GT BIOS during that time too. I dunno if there is anything even wrong with the card but the fact that it quit working for no reason and refused to boot is very scary.

The blue lines appear to have been a driver issue. After flashing the card I installed the 66.51 drivers. After going back to the 65.73 drivers the blue lines disappeared. Anyone else have this issues with the 66.51?

wickfut
10-11-04, 10:55 AM
the bios came with 1.4 volts already set up - so its been about 6 weeks now

i cant remember the program that i checked the bios with tho , i think i have a reviewers sample or something because it came OEM with a standard A400 single slot HSF on instead of the large double HSF that the usual A400GT comes with.

also it has a strange bios vendor ID - GT is 2996 - mine is a 2997 ? so winfast will not let me load any 6800 (standard/gt or ultra) bios into it , go figure?

http://www.futuretrance.pwp.blueyonder.co.uk/images/a400gt2997.rom
http://www.futuretrance.pwp.blueyonder.co.uk/images/a400gt.jpg

anzak
10-11-04, 02:45 PM
Another problem. Tons of artifacts filled the screen and the speaker on my card went off. The card is booting up just fine but now I know it's borked. To top it off my warranty was voided...

boro
10-11-04, 03:08 PM
could also be your powersupply starting to act up perhaps ?

Burnt_Ram
10-11-04, 03:18 PM
Another problem. Tons of artifacts filled the screen and the speaker on my card went off. The card is booting up just fine but now I know it's borked. To top it off my warranty was voided...
that totally sucks man ! i feel for ya

anzak
10-11-04, 03:20 PM
could also be your powersupply starting to act up perhaps ?

I doubt it. Not with the checkerboard that filled my screen just a few minutes ago. Since I don't have a backup card I purchased a 64mb Radeon 9000 from newegg for $40. I should be able to play Half Life 2 at low detail. It will get me by until I can afford another high end card (this time from ATI).

aAv7
10-11-04, 03:22 PM
As many of you know I have been running my GT with Ultra BIOS for some time (I was one of the first here at nVnews to try it). Just moments ago my monitor went black and I got the check signal message. I was just surfing the nvnews forums. I quickly shut off my computer and started it up again... no signal. Nothing, the computer starts but the graphics card was not working. So, scared ****less, I remove the card to look for physical damage. It looked alright so I put it back in and bam it worked. Could this be the long term effects of running 1.4v? My temps are well under control (48c idle and 61c load). I plan to flash it back to the original BIOS after posting this.


I had a similar thing take place....I havent modded my card in any way..not even oc'd.....but one day I opened my case with the pc still on, just to move a cord that was bothering me. ( I have the plexi-glass see through lan boy case, and one wire was hanging down) so I move the wire, which was from the dvd drive...I close the case...move my mouse and find out the screen is frozen...I reboot and my monitor gives me the signal loss screen and then turns black...reboot again, same thing...thinking I fried something, I check the psu, changed it with my old 350watt...still no signal, take out the 6800, place it back in...no signal, put my 430 watt back in, re-plug in all the wires going to the mobo, re-adjust the nv40 again in the agp slot, boom, everything loads up....so weird....this was about a month and a half ago...haven't had a problem since.

wickfut
10-11-04, 03:25 PM
what about your nv5 - have you checked to see if its still on perfectly ? may just be something come a bit loose and making it overheat - it is a rather large weight hanging on by 4 tiny screws after all

before i would give up i would do a full "back to factory" hsf , clocks , bios etc etc - even a fresh instal of windows before i blamed the card and tried to rma/sell it on ebay (joke)

anzak
10-11-04, 03:51 PM
what about your nv5 - have you checked to see if its still on perfectly ? may just be something come a bit loose and making it overheat - it is a rather large weight hanging on by 4 tiny screws after all

before i would give up i would do a full "back to factory" hsf , clocks , bios etc etc - even a fresh instal of windows before i blamed the card and tried to rma/sell it on ebay (joke)

My temps are fine, around 45 degrees idle.

Well im still getting blue flashes every once and a while. I believe it's on it's last leg. At least when / if it dies I'll have something to fall back onto.

Chippy
10-11-04, 04:11 PM
OK so we all know that we ditch the warranty when we mess with our cards - like fitting VGA Silencers etc.

However, I had been kindof thinking that the only risk was me doing something that messes up my card. And so long as I was careful, and I didn't mess anything up, then everything would be OK.

But what this thread reminds me is that computer stuff just dies sometimes. Not really for any good reason, but it just dies. I don't think Anzak's problems are because of his bios flash. I think his card is just dying.

Personally I reckon his card would have died even if he hadn't messed with it. Then it would have been a warranty claim. And ironically although his mods (imho) haven't caused the problem, they have nevertheless let the manufacturer off the hook.

Hope it pans out OK for you Anzak.

Chip

anzak
10-11-04, 04:31 PM
OK so we all know that we ditch the warranty when we mess with our cards - like fitting VGA Silencers etc.

However, I had been kindof thinking that the only risk was me doing something that messes up my card. And so long as I was careful, and I didn't mess anything up, then everything would be OK.

But what this thread reminds me is that computer stuff just dies sometimes. Not really for any good reason, but it just dies. I don't think Anzak's problems are because of his bios flash. I think his card is just dying.

Personally I reckon his card would have died even if he hadn't messed with it. Then it would have been a warranty claim. And ironically although his mods (imho) haven't caused the problem, they have nevertheless let the manufacturer off the hook.

Hope it pans out OK for you Anzak.

Chip

Thanks. I hope your right about this being exclusive to my card. This is the second BFG GT to die on me, thankfully the first one was covered by the warranty. Right now the card is running stable (no graphic abnormalities yet) @ 350/1.0 with 1.1 volts. I really need this to last me, I don't want a $400 paper weight and 800x600 in Half Life 2.

anzak
10-11-04, 08:00 PM
I posted this in another thread, but it may be of interest here as well. Since I figured my card was dying, I tried some crazy settings on it, and found something that was rock solid.

I think I may be on to something involving voltages.
By increasing my 2D voltage to 1.4 volts, the same as my 2D/3D voltage, and setting both to the same clock speed (in my case 445/1250) I've obtained a stable overclock with, get this, a LOWER temperature than I had when I was at 1.1/1.4 volts (49C at idle compared to 55C), and no more funky checkerboard crashes. Dunno if it is a temp fix or not, but I'll keep you guys in the know.

Thanks for the heads up. I'll give it a try.

anzak
10-11-04, 09:05 PM
Woah! Thanks! I believe you may have fixed the problem! No checkerboard crashing or blues lines so far! I will keep you updated.

Greasy
10-11-04, 09:52 PM
But what this thread reminds me is that computer stuff just dies sometimes. Not really for any good reason, but it just dies.

So true. I had my first 9800Pro die on me while playing Halo. It was a week old.

jefmor
10-13-04, 12:42 AM
Butter Bandit, I also set my 2D voltage at 1.4 and my temps never go over 67c when playing unreal2004 or doom3. Thank you for the info!!!

Chippy
10-13-04, 03:40 AM
No problem. I just played UT2004, max settings etc, for a few hours, and no issues at all since I changed the 2D voltage. I'm assuming that the stability issues may have been caused by the frequent switching of the GPU from 1.1 volts to 1.4 volts, by keeping both the same, I think it actually puts less stress on the GPU. I'm no hardware engineer though, we'll see how it holds up, and if I see anything crazy happen, I'll let you know.

More likely its a bug in the bios code, I reckon. I suspect that the voltage being applied in 3d is not the same, depending on what the 2d voltage is set at. Could be down to some quirk of the voltage regulation circuitry too. Who knows. Its very strange all the same. The thing that's *really* odd is the temps being a bit lower. Thats really spooky.

Chip