• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Stock 6600gt temp? 116???

mkerr

n00b
Joined
Oct 24, 2004
Messages
44
Hi people.
I made a thread a while back about a 6600gt of mine (new) that was having problems with green artifacts everywhere when running certain games (especially brothers in arms)

well, according to the store i bought it off, they sent it back to else and they apparently ran extensive tests on the card, and there was nothing wrong with it...

i've just gotten it back, thanks to their insanely slow posting speed. i'm still having the same fucking problems. i've since installed a new psu (thermaltake 470w) and now my brother's gainward 6600gt works perfectly.

i installed the elsa 6600gt this morning, and i'm still getting these damn problems. i ran rivaturner in the background to monitor temps, and this is what i'm getting. is 116degrees cel too high? i'm guessing so. if so, either elsa's technicians are monkeys or i'm being screwed around.

the screen shots don't look too bad, but they're far far worse than that, just had difficulty grabbing it.

thanks for any input :)

screen1.jpg

screen2.jpg

temp1_no_marker.jpg

temp1_marker.jpg

temp2.jpg
 
Yes, 116 is extremely high. Those artifacts are VERY likely the result of temperature issues.

Sounds like the fan isn't operating properly or your heatsink isn't seated. Or, you may just have your goodies turned up too high. Either way, 116 is about 30 degrees out of the comfort zone. You're probably doing some damage to your chip.
 
116 is way too high, avoid using it now as much as possible since it is actully damaging your videocard. Turn on the computer with the case open, see if the fan is operating, if it's not there must be a problem with the powe rconnection to the fan itself. If that is the case, contact the manufacturer and see what they can do.
 
Dude...wow. My 6600GT runs at anywhere between 40-60 degrees C under load (depending on what sort of load), but doesn't really get any higher. (The extreme temperature was playing BF2, with most of the eye candy turned to medium). Are you overclocking the card at all? I know that when I went crazy one night and let Coolbits auto-detect the frequencies, it was doing about 70 degress C when Milkdrop was playing in the background... so maybe check you clock setting as well.

But definately try to isolate it, the problem sounds like a fan problem or a heatsink not on properly, or not enough ventilation. Once you find the problem, RMA it quickly to them detailing the fault. The more you document it, the more they have to work with, and the more detail you have to say why they haven't fixed it (if it comes back unfixed).

EDIT: sorry, I forgot that its stock... but still, if you can, check the stock clocks. The core should be at ~500MHz, while the RAM will be about 900MHz (AGP) and 1000MHz(PCIe)
 
yep
everything is stock.
never messed around with any settings at all.
i've never even overclocked anything before, and only just started messing around witht he 4400ti i'm using whilst this piece of shit is trying to be replaced.

they apparently sent it to be tested, and apparently 3dmark03 was run for hours with no fault found. i'm finding this a tad hard to believe seeing as it's hitting 117 in my computer, and 108 with a desktop fan directly blowing on to it.

I installed it into my brother's computer (antec sonata with trupower 380w PSU) and it hit 118 degrees and artifacts galore.

All this happened after about 40 seconds of game 2 in 3dmark03... so i can't see how they didn't get artifacts. even if they didn't, i'd expect a tech / repair section of a manufacturer to at least measure the temperature of the card whilst it was running.

The card was doing this the day i received it. So it should come under the DOA terms. I've been fucked around for over a month now, mostly waiting for very slow shipping (which is apparently over night) and also waiting email replies that average at a week waiting...

I think i'm right in saying that i'm going to now kindly ask them for either a direct replacement of the card, or for a full refund.

for interest's sake, here's a link to my original posts, both on [h] and overclockers.com.au
http://forums.overclockers.com.au/showthread.php?t=392722
http://www.hardforum.com/showthread.php?t=929992

i've emailed ELSA regarding the problem, asking them about the unusual temperatures, why they weren't reported when it was tested, if in fact it ever was tested, and what they advise i do.

thanks guys...
 
GL with that, dont let them stick you with the card and the bill. I'd shoot them in the eye (verbally) for such a disgraceful card checking personally :D.

~Adam
 
hahaha
it's getting hard not to go nuts...
they take on average a week to reply to emails, i've needed to ring them to get anything done or replied to.

they emailed me to say there was a delay and that the reason was they were sourcing a new card for me. which in the end, looks like it was complete bullshit, seeing as they then turned around and said that there's nothing wrong with the card. i'd have though anyone with a shred of common sense would first check the card and then source a replacement. it was a nice sounding excuse though...

i'm also pissed off, as they won't accept drop offs, and they're a 25min drive from my house.

i was originally going to get a chaintech 6600gt from my usual store, and thought i might as well give these guys a try, seeing as they're an official overclockers.com.au sponsor... turns out that with all the fucking extra postage, i might as well have just got the chaintech. would have ended up the same damn price.

ahhhhhh.... at least far cry works somewhat nicely on the 4400ti...
 
Get right up them. If you are experiencing problems with the card, then they should be picking it up too. Tell them what you think, and yeah tell them to either give you a new card or a refund. And if it doesn't work in two computers, then you can really say tht its the card.

And you should get better service than that.
 
thanks for the support :)
i'm trying to be as nice about it as possible, i get irate customers every day in my work, so i know how it is to be on the receiving end. but when they're this slack on their customer support, it seems you've gotta get up them to get anything done...

here's the latest installment.
i uninstalled all my antivirus, spyware, monitoring, konfabulator, MBM, everything i could think of.

i took apart the whole computer. re-seated everything, reapplied paste to my cpu heatsink, double checked all fans and connections. i gave the vid card its own PSU rail.

after all this, it seems that the card has far far fewer artifacts. it took me about 4 runs of game 2 in 3dmark03 before i got any, and i was getting about 10-20 throughout each entire run of the game scene. while this still isn't fine for a stock and brand new card, it at least brought a smile to my face.

i had rivatuner monitoring temps on my other monitor whilst i was running 3dmark03. each time i noticed artifacts it coincided with these strange spikes in temperature (edit: on the piccie below, the spikes i'm referring to are where it almost hits 120degrees then drops to 110 for a short time). it was averaging around 113-116 degrees whilst under heavy load, as you can see from the piccie.

I'm still waiting on a reply from ELSA themselves regarding the temperatures, but it seems that the general consensus is that temps this high are dangerous, and will kill the card, and most likely resulting from a poorly attached heatsink. i can't take off the sink as it'd void my warranty.

So the latest is, the artifacts are somewhat better, but still present. the very high temperatures are still there, even with a desktop fan blowing directly at it.

latest_temps.jpg
 
Good to see you're being thorough, but I believe the consensus is that your card is running TOO HOT - lol. If you want to burn up the gpu to teach elsa a lesson, that's cool, but that's the road you're on at the moment.
 
unfortunately it's the only route i was able to take.
i've been getting messed around it seems...
i'll be calling tomorrow and if they don't have a definite answer i'll be asking nicely for either a replacement sent out immediately, or a full refund.

i can't believe the shit service. if my work ran like this it'd fold.
 
I hope that the other cats reading this thread have developed the same appreciation for good customer support that you have.
 
it's a serious lesson learnt
i'd always taken the approach that all computer bits (made by the same manufacture / brand of course) are the same. so i've gone through the cheapest route / store. thankfully this has usually been umart (brisbane, australia) which has good support and service.

decided that i'd go for the cheapest, so used staticice.com.au to find the cheapest 6600gt that wasn't an xpertvision.

turned out i could have bought a chaintech 6600gt from umart for the extra money i've now had to pay in postage...

i'm certainly going to stick to my trusted dealers from now on... :)
 
DNA Doc said:
I hope that the other cats reading this thread have developed the same appreciation for good customer support that you have.
Agreed, I work in a computer shop myself, and when our supplier screws us around, we have to take the blame and cop the abuse. Not to mention people just being plain rude in general.

Nowadays, even when I go into a fast food joint I am always as kind as I can be, cause I appreciate what they are doing after experiencing it myself.

However, there is a limit to bad customer service. If you are completely in the right (which, I and most others beleive you are), then you have every right to go and ask them for a replacement and a refund. I sure as hell would :)
 
Find the spot in BIA where the card gets really hot and just leave it on for a day or two till it gets really bad and hopefully dies, then send it back :)
 
mkerr said:
i had rivatuner monitoring temps on my other monitor whilst i was running 3dmark03. each time i noticed artifacts it coincided with these strange spikes in temperature (edit: on the piccie below, the spikes i'm referring to are where it almost hits 120degrees then drops to 110 for a short time). it was averaging around 113-116 degrees whilst under heavy load, as you can see from the piccie.

Someone correct me if im wrong, but isnt the Thermal shutdown for the 6X00 cards 120C? This would explain the problem you saw with the spies. As the cards hits 120C, it probabaly goes into thermal shutdown for a small amount of time to let the temps drop alittle so that the core doesnt burn up, during this time, the card will of course stop rendering correctly, if at all.

You can check the shutdown temp for your core in the nvidia control pannel (on the temp monitoring page), but im sure its either 120C, or very close to that.
 
See if SpeedFan will pick up the sensors on the NV board. Don't run RivaRuner at the same time because SpeedFan won't pick it up. It did on my rig. I usually monitor my temps with RivaTuner and SpeedFan at the same time but yesterday, I forgot to load RivaTuner and SpeedFan picked up two additional temperature reading. Doing a little research, I found out these were MAX6648 readings. It means these are temperatures from the Nvidia board. It has been reported that these reading are 10C lower on the core temperature because Nvidia puts an additional 10C as a safety factor.

What time interval are you using on those graphs? They seem awfully sharp for default intervals. My guess is that the GPU heatsink is not seated right.
 
Woah man thats freaking hot my gigabyte 6600GT which is clocked at 550/1219 reaches 60-70 on load, try replacing the fan.
 
Back
Top