PDA

View Full Version : checkered screen = high temps? (pic included)


Pages : [1] 2

bluefire7
12-14-04, 02:51 PM
hi, this is my first time in this forum. so far as i know, this is an isolated problem. i figure this is probably the gpu heating up too much? but i was just trying to get a few more opinions here. when i encounter this problem, it's playing either doom 3 or half life 2 or even cs source. the screen basically freezes as you see below, but sometimes you can still see the outline of the last frame through the colored boxes. the sound starts to skip and eventually i have to restart my comp =\ any clues as to what may be causing this if not the heat and how to troubleshoot? my videocard oc= 420/1100, temps in windows right now are 70 gpu and 53 ambient, but that's running at 370/1100. after playing a game, i've seen the gpu temp at 76. your thoughts?

look:
http://www2.newpaltz.edu/~wang889/crappy.jpg
Temps were up to 86 when playing hl2 for 30 min.. but still way below the default threshold of 120 in coolbits

wEEt
12-14-04, 02:59 PM
70 in idle is really bad (53 ambient aswell). One problem is defenitely the high temperature. If this has something to do with strange pattern on the screen - I have no idea...

rflair
12-14-04, 03:12 PM
You card is probabbly overheating, 70c idleing in Windows is way to much, I idle at 45c. In games I would think your in the 100c or more range, considering you said you had a temperature of 79c after finishing a game, and you can expect a 10-15c temperature drop within a second of finishing anything 3D (and in your case i would say 15-20c+).

That being said I can replicate the exact same screen on my card by flashing a 1.4V BIOS and overclockcing to 400/1100, which is fine at 1.3V. And for me its not a heat issue as the idle and load temperature rise 2c and 4c, its probably the mosfets overheating and/or not being able to supply enough current at the same clock speeds.

But for you its most likely HEAT.

Get http://www.daionet.gr.jp/~masa/archives/rthdribl_1_2.zip and run it with the Nvidia temperature monitor (or whatever you use to monitor card temperature) showwing. Run it for at least 30 minutes, it should (will) max out your card temperature wise.

It could just be the heatsink on your card mounted badly.

rflair
________
CM185T (http://www.cyclechaos.com/wiki/Honda_CM185T)

Faster
12-14-04, 04:24 PM
This is exactly the screen that me and my friend get from the 1.4v bios.

It really isnt anything to do with the 1.4v bios, just that the temps of our cores get up to over 80c and then hardlocks with this screen.

Dont worry about, just downclock your overclocks. 400/1100 with original bios is the best stable overclock really for our two BFGs.

bluefire7
12-14-04, 04:47 PM
thanks for the input! i really appreciate it. i'll try it later on tonight and see. should i run this

http://www.daionet.gr.jp/~masa/arch...thdribl_1_2.zip

maximized or in a window? whats the max temp i should aim for in my overclock?

rflair
12-14-04, 05:02 PM
Run it in a window and run the temp monitoring software at the same time, so you can watch what is happening.

I wouldn't even try overclocking, your card is running to hot, until you resolve the heat/lockup issue why would you even bother.

rflair
________
Honda NSR500 (http://www.cyclechaos.com/wiki/Honda_NSR500)

noko
12-14-04, 05:29 PM
Overclocking your card to a point of failure is probably not wise. Also consider the FPS increase from 400 to 420mhz is virtually zero and probably no more then 1 FPS. So keeping your card cool, not artifacting, not BSODing and definitely not corrupting the screen will give you some use out of it before it fails.

ifusayso
12-14-04, 07:05 PM
When I first got my system my card temp climbed to 149c. How it did not fry it I will never know. My problem was the fan sensor was not work so I rewired it to its own powered molex.

Even when the temp climbed that high it did not lock my card I just caught it by accident when checking the temp so I turned off the computer real quick let it cool down. Restarted it then saw the temp climbing again felt to see the 6800's fan it was turing real slow. That was the problem.

Funny thing after giving the on board fan its on power i have never had a stutter or frezzing problem. But the 6800 now is running of 3 molex connections.

bluefire7
12-15-04, 12:35 PM
weird.. i try the stress program windowed for 30 min with AF at 16x and the highest it gets is 83.. whats the deal?? that's not that hot at all..

Morn
12-15-04, 12:55 PM
My 6800GT used to idle at 74C before I got the NV5 now it idles at 40C. I never used to get that pattern even though I have seen it go up to 120C. On hot days it would tend to pause and jerk a lot in games.

bluefire7
12-15-04, 03:16 PM
yea exactly.. thats what i don't get. it's not like it warns you before it freezes on that screen. with my experience, when i oc too much it will artifact and stutter, not just stop all of a sudden..

Jobe
01-29-05, 08:18 PM
Hi all ,

Im sorry for digging up a lost topic, But this is the only topic that shows what is starting to happen to me latley.

I brought a BFG 6800 GT back in mid October and until last week it has been working fine, But now I am getting this Screen after say 30mins of game play in COD UO & UT2004 then my PC reboots.

I have not touched anything in the Clock setting as i dont want to waste 256 and void my lifetime support.

I have just this second emailed BFG about this problem, I should get a reply soon, But I was wondering what could be the cause of this ? My Card temps reach about 65 core when it crashes so I dont think it is heat problem.

I am at a loss here as to why all of sudden this starts to happen, I have not changed a thing in my PC since I got the card, Only GFX Drivers thats all.

Do you think my card is Defective ?

Thanks in advance

Jobe

SigmaOrionis
01-30-05, 12:38 AM
Hi Jobe,

That is also the same type of screen that my 6800GT is giving me. Like you I havent changed anything since I got the card. Weirder is that my friends 6800GT works fine in my comp. I've decided to RMA my card. I know this sounds crazy but have you tried to reseat your card. For some reason if I reseat my card it will work for a couple of days and then I'll get these type screens after about 20 minutes. Once again my friends card does not cause these problems in my PC.

If you figure anything out please post back.

Jobe
01-30-05, 07:42 AM
Hi SigmaOrionis, Thanks for your reply.

Thats just it , Once I started having this wierd screen followed by a reboot I Re-formated my System,Took the Card out to check the fans were ok ect and everything was fine(Or atleast I hoped) until it happend again last night while playing COD UO :( And after I posted here I decided to browse these forums a bit more, I hit the Back button and Bang! My PC reboots for no reason..

I have not had a reply from BFG yet, But as soon I do I will let ya know ;) The worst thing about all this is I have not got a spare Display Adapter to use if my Card has to go back, So Im without a PC until they send me a 100% Working card.

How long does it take for BFG to send an Replacment ?

Thanks

Jobe

Phalinx
01-30-05, 10:30 AM
I've seen this kind of thread on a couple of forums and there are some common denominators.

1. You own a BFG 6800 product.

2. Those cards come overclocked already. Question, did it dawn on anybody why their card crashes beyond BFG's oc? I have. Those cards were tested already for oc stability and guaranteed by BFG. Those gpu's might be seconds (or rather, they just made the 6800gt specs). You may not be able to go any higher w/o problems.

3. Those cards don't like anything beyond 1.3v. This goes along with stability. If those chips used by BFG can only handle 1.3v then don't fool with a bios mod.

4. Overheating is an issue with BFG. If it were me, I would try a different aftermarket hsf like eVGA's copper hs. I have an Ultra copper hs on my eVGA 6800gt. It keeps it nice and cool (gpu 48c, ambient 31c).

5. Open up your tower/case. Get good air flow going. I have a tornado w/adjustable speen mounted under my copper hs. Water cooling might be an option for the gpu...

Cheers

Phalinx

SigmaOrionis
01-30-05, 11:08 AM
Jobe,

Ya I know its frustrating. BFG took about 8 days to return my card from the time they received it.

The difference between what I'm getting and what you are getting is that my comp doesnt reboot when I get that screen. Your problem could very well be a Powersupply problem. I dont know how much of a hassle it would be for you to try another but it might be worth a shot. I've also had random reboots be caused by drivers when I was using an ATI card. But you said you tried a clean install of XP so I dont know about that. Memory going bad can also cause random reboots. Anyways if a different PSU doesnt fix it I would RMA the card. BFG's RMA process went very smooth. Just call there toll free number and they'll set you up. Good luck I hope you get it worked out.

Phalinx,

I'm not overclocking my card beyond what BFG has already out of the box. Also am using the stock bios.

Jobe
01-30-05, 12:23 PM
Hi agian SigmaOrionis.

Yer I debated if it was PSU related, All I know is I have a 400watt PSU and its been fine without a hitch for just over 3 months. If it was somthing to do with PSU would it not show right away ? Also somtimes my PC won't reboot at that screen, it would just hang there until I switched off manualy.

I was looking at the 'Antec True Power 480w PSU', Do you think this is good enough for the job ? Mind you I am not looking into overclocking or anything like that in the furture.

Thanks for your help ;)



Phalinx,

I'm not overclocking my card beyond what BFG has already out of the box. Also am using the stock bios.

^^ What he said hehe .

Thanks

Jobe

SigmaOrionis
01-30-05, 08:32 PM
Hi Jobe,

I'm using just the standard Antec 400 and it has been fine for my system. The thing I hate about saying its the power supply is I wouldnt want you to spend money that you didnt have to. When PSU start to go bad what usually happens is your computer will just start acting flaky with random reboots, failed cold boot attempts, and just plain instability. At the same time it could just be that your Graphics card is starting to fail and the PSU might be perfectly fine.

You could use the hardware monitoring software that came with your motherboard and watch the rails to see if they have any fluctuations. The more accurate method would be to use a multimeter on one of the 12 volt leads and monitor the +12 while you play a game and see if its dipping when you get your reboots.

To make sure it isnt one of your memory chips going bad you could use memtest.

For the Video card the only way to really rule it out as the problem unfortunately is to put another identical card in your comp and see if the problem goes away. Luckily I had a friend locally that let me test his card in my computer. So I'm 99% sure in my case that it is my graphics card.

NemesisChild
01-30-05, 08:56 PM
I had this exact same problem with my BFG when I first got it.
Here's what I had to do:
1.) Upgrade my PSU.
2.) Added extra and upgraded case fans.
3.) Installed an NV5.
My load temp never exceeds 64c now. Now, I did get the same lock-up the other night ago, but it was because I was attempting some aggressive overclocking. I would suggest taking the card out and see if the heatsink screws have become loose. Maybe the HS isn't flush against the GPU. If that's not the problem, RMA it immediately! Good luck............

Jobe
01-31-05, 08:33 AM
Hi Sigma.

Well I've held off in investing in a PSU for now, And I'm going to wait to see what BFG has to say (STILL waiting for a Reply). I would phone them but being based in the UK and them in the US the phone Bill might get a bit large hehe .

So far I have taken out 1 Ram Module to test to see if its duff Ram, I did try Memtest once the other night left it on, Got up the next morning(9hrs later) and it was still runnnig hehe so I canceled it, I did not notice any errors before shutting it down though.

I played COD UO last night for a couple of hrs fine, But I need to test it a bit more, As I tryed your Idea of Reseating the Card so it could be that why it worked fine, Will see!

Thanks.

NemesisChild -

Maybe the HS isn't flush against the GPU.

Funny you should say that, As the HS does NOT look flush with the GPU, It looks like its sat on top a lump of Gel in the center of the Chip, I can see daylight around the lump except for the Center where the HS is touching, so I guess this is not Flush ? But my Temps seem to be normal 53 core while not playing and about 60/70 under load.

Hmm I wish BFG would hurry up and reply,I also sent them the Screen shot of what happens so they have some idea of what Im on about, So much for there 24/7 support.

Thanks.

Jobe

diklittleboy
02-01-05, 03:41 AM
That temperature is not high at all. Ive seen other 6800 such as sparkle, xfx idling as hot as 71C in hot summer days. When gaming it gets as hot as 80+.
I think comparing BFG temperature readings to other cards is not correct the thing is that BFG always gets lower temperature readings as oppose to say Leadtek, or MSI.
My 6800 gets as hot as 65C idle, and get as hot as 71C after hours of doom3 both in summer. So the bottom line is heat is not your problem.

jAkUp
02-01-05, 03:42 AM
Yea mine gets up to about 75 degrees C. I don't know how they are able to withstand such extremely high temps.

ViN86
02-01-05, 12:09 PM
quick question.

do your BFGs have the copper/aluminum hsf (with the two fans and two blue LED's) or the stock hsf (its green with one fan)?

i have the one with the copper/aluminum hsf and it works great. my card idles at 55 regardless of ambient temps for some reason, but my card doesnt go above 65 ;)

NemesisChild
02-01-05, 05:47 PM
quick question.

do your BFGs have the copper/aluminum hsf (with the two fans and two blue LED's) or the stock hsf (its green with one fan)?

i have the one with the copper/aluminum hsf and it works great. my card idles at 55 regardless of ambient temps for some reason, but my card doesnt go above 65 ;)

It all depends, the BFG's come with both.
I happened to purchase the BFG with dual cooling fans.
A unique design, but sadly, it doesn't do a good job of cooling.
Be happy you have the reference design, it's a much better cooling system.
With my NV5, I also idle at 55c and never go above 64c under load.

life of budha
02-01-05, 06:01 PM
moniter short in it might be ****in it up