MSI Says My Overheating 7970 Lightning is Normal!!! - Page 12

Originally Posted by damric 

I disagree.
I don't think the target customer for factory OC'd cards is mainstream overclockers. Extreme overclockers, yes, but they use sub-zero cooling anyway. These cards are mostly for rich-boy gamers that want to one-up their neighbor. Mainstream overclockers are generally more than happy to take a cheaper reference card and push it on air/water. These pre-OC cards often run much hotter because they also have higher than stock voltages.
If your room temperature is hot, then you can expect your electronics to run hot, and that is a fact. If it concerns you, lower your thermostat.
Which brings up a good point. What is the room temperature of the OP? Looking at his idle CPU temperatures (well into the 40's) , I'm guessing it is hot, or air flow is restricted in his case somehow. Is OP in Africa or something?

Originally Posted by Catscratch 

Now here's the line. 70-80c while gaming is ok. 90c+ while furmark is also ok with stock coolers (because they suck). However 90c while gaming is NOT ok. So Factory OCed or not, no card should reach 90c while gaming, and op clearly stated very high fan profile to keep it cooled. The fact the second card also behaved the same means this 7970 line from MSI is flawed and isn't worth the money.

Originally Posted by MSIalex 

I would not make an assumption about an entire line of over 5000+ (and counting) cards based on 2 samples.

Originally Posted by Ganf 

People with problems are more noticeable because the people without problems aren't saying anything.
I'm not throwing my towel to one side or the other, I'm just going to say a few things to add my chip to the pile, even though I don't own a lightning.
My Sapphire 7970 OC ran at what I see people saying were typical temperatures for reference cards, and they supposedly have one of the best aftermarket coolers. Overclocked the thing was reaching 85+. Now that I've put a waterblock on it all I can say is that when my CPU and GPU are under full temperature load, the amount of heat coming from my 360mm radiator is pretty frickin' impressive. I could probably dry my hair with it. Bottom line: Some 7970's produce an incredible amount of heat due to hardware faults that have nothing to do with the reseller. They can only design their coolers to function well with the majority of the cards, and in all likelihood if your card is running that hot at stock it probably wouldn't have made it through their binning with a reference cooler.
Fact: No matter if a card is advertised as being optimal for overclocking or not, the only thing any company guarantees is that the product you buy performs within their quality standards as it is configured out of the box. Really AMD is the one that set the high standard for temperatures on the 7970, MSI just didn't try to one-up them like most resellers do.
I'm really not surprised that an RMA replacement card is running hot. That's probably why that card was RMA'd by the first owner, despite the reason they gave on their ticket. However, I for one feel that excessive heat is a legitimate reason for an RMA. I burned up the southbridge on my previous motherboard because my video card was exhausting too much heat onto the chipset heatsink. The MSI Lightning also exhausts heat in this area and if the card is posing a risk to other components in your computer because of less than adequate performance I don't see why an RMA for that reason would be an issue. Inadequate ventilation can't be blamed in this instance because the heatsink is literally millimeters away from the heatpipes. The amount of heat that radiates from those pipes is pretty significant. Crank up furmark and put your finger on one after five minutes, and even with a cool-running card you won't be keeping it there long. This is coming from a welder who is used to hearing his skin sizzle and pop like fatty pork in a frying pan, by the way. My definition of hot isn't the norm.
I guess my gist is that you can't crucify MSI for re-distributing a card that meets their minimum specs, but also that MSI should be careful in the future about how they choose their minimum specs, because there can be some serious unintended consequences.
By the way, the motherboard that I burned the southbridge up on? It was made by MSI. I do have a little experience with their products, even if it isn't their VGA's.

Originally Posted by nvidiaftw12 

On my old mobo, I couldn't run my graphics card over 70C because my pc would lock up, due to the northbridge being to hot because of the heat coming off the graphics card. Thankfully, due to my good stock cooler, i could keep the temps under 65C. Had I have had the OP's card, my computer would have been unusable, and I would have had to have bought a new mobo.

Originally Posted by nvidiaftw12 


He posted a picture of his room temp. 77F. I've had my room upwards of 95F and still couldn't get my 6970 over 65C with 100% fan speed.

Originally Posted by Redwoodz 


77F is hot for an electronics room. Generally for any component,a 1 degree increase in ambient temp will result in 2-5 degree increase in core temps. The 6 series is much cooler than the 7 series across the board.

Originally Posted by nvidiaftw12 

Psh. My room goes over 100F at times. My 6970 still doesn't go over 65C if I don't want it to.

Originally Posted by Thracks 

I'm interested in learning more about how you've defeated the laws of physics with a GPU that runs up to 50F colder than the environment it's sitting in. Are you a wizard?