Under 90c is fine and won't damage the card, u seem to have fairly decent case airflow, u could try removing the side panel from the case and testing just to make sure, other than that a re-tim of the card might help things a bit.

OH AND BY THE WAY - this card is running at like, 35-38c idle right now as I'm posting in this forum... a near 60 degree jump while gaming 

So your advice would be to not RMA because of the high temps, and instead re-TIM? I know what that is, but I've never done it before, and looks a lot more complex than re-setting a CPU heatsink lol.

Redoing the thermal paste on the 560ti SC is not a bad procedure. Fairly easy to follow and it's quite simple. From what I understand the existing thermal paste is a good quality from the factory. Not sure but I think someone can verify that they are using Shin-etsu from the factory.
 
You could always get a good ceramic thermal paste and redo the tim if you needed to. I have a 560Ti DS SC and I'm sitting idle at 35c and generally get to about 63c playing during the day when the weather is hot here. I've got excellent airflow in my HAF 912 so I'm not too worried about the board. 
 
I have some pic's of redoing the TIM itself for when I redid it for a friend's 560Ti DS.
 
-----------------------------
 
First you'll want to remove the screws holding onto the fan shroud, on the DS model there should be seven of them. Might be more for the non DS model. The ones to be removed are the ones with the red circle on them. The four in the middle are for the heat sink itself, the three smaller ones are for the shroud.
 

 
Just remember to carefully disconnect the fan cable once you have unscrewed the shroud.
 
From there you should see something similar to this with the heat sink still attached by the existing thermal paste.
 

 
Just a slight twist of the heat sink itself, just be careful not to bend the blades on the heat sink and it should pop off after breaking the seal of the thermal paste.
It should look something like this after removing the heat sink.
 

 
From there you'll want to clean up the existing thermal paste with some cleaning agent. Some people like to use the alcohol solution method (I think they use 91%) or something similar. I like to use the Arctic cleaning package which has the two step process. I just wipe using a paper towel to get the bulk of the previous thermal paste off and then use the solution to clean the remainder off using lint free cloths and some q-tips. Do that for both the GPU and the heat sink itself to make them as clean as possible. Once done the GPU should look something similar to the following.
 

 
And the heat sink like the following. Notice the heat sink has several rods that are placed together to form the connection point for the thermal paste to conduct heat from the GPU itself. I went over it several times to remove the existing paste from the cracks in between.
 

 
Make sure you use the solution to go over the heat sink and GPU in the second (blue capped) bottle to prepare the surfaces for the new TIM application if you are using the Arctic Silver preparation kit. It does wonders for filling in those gaps that might be there. If not just continue along.
 
From there I used a decent ceramic thermal paste to put several beads of the material on the GPU itself. Ceramic needs to be placed a bit thicker on the GPU so that you have a good connection with the heat sink when replaced. I put enough on there so that it would also seep into the cracks on the heat sink as well.
 

 
I also spread it using the credit card method. This seems to work best for me. I had to go back over the GPU to ensure there was enough on there before re-assembling the card back together. Notice I had a few uncovered parts of the GPU, I just put more of the paste on there and redid the credit card sweep across to make sure it was fully covered. Don't forget to wipe off any excess that comes around the sides if needed on the GPU. The ceramic thermal paste won't harm the circuitry but it always pays to be neat and clean.
 

 
Reassembly is fairly easy, just line up the screws to the heat sink after pressing it together initially on the GPU and use the criss cross method to reassemble the card from the middle and then replace the fan connection and bolt back the shroud.
 
That should pretty much cover it. I used The Artic Silver ACN-60ml to clean the GPU and heat sink connection points and then used the Arctic Silver CMQ2 tri-linear thermal paste to replace what was there.
 
http://www.newegg.com/Pro...x?Item=N82E16835100010
http://www.newegg.com/Pro...x?Item=N82E16835100017
 
And there you have it. Hope this helps you to understand how I did this.
 
 
 
 

I'd suggest not using silver based thermal paste on GPU's. I read somewhere that it's not good for them as it's not meant for that kind of application. I don't recall where I read it initially but I have been using ceramic based thermal paste only on my GPU's and other types for my CPU heat sinks.
 
And as far as the utilization goes, with one game I sit around 50-60% utilization on the card. My theory is that if you can set the graphics to 'ultra' or fully rendering and still have some amount of processing power to spare then it's fine. I don't see much in the way that having more utilization of the GPU processor and/or the CPU being fully utilized is going to matter much when you are killing things in the game. As long as there is no frame tearing, the game renders well, plays well and the graphics flow without stuttering then it's all good.

I can understand that - I just felt like I could be pushing this card more than it is. It's a powerful card and I feel I could be getting more out of it. If I knew how it was going to perform, I would have just purchased a 470, or SLI'd... 
 
By the looks of it, you almost have the same computer setup as me as for as capability goes, and you're getting completely different readings than me. That seems strange to me, and makes me think getting a new card would be worth the trouble.
 
Newegg offers the option of shipping a new card to me, and I have 30 days to return the old one. Seems like a good deal to me, and would really allow me to test them both out. If they both result in the same usage / temps / FPS, than I can exclude the GPU as the problem and I can start looking elsewhere for temperature relief.

Thanks for the help, I really appreciate it!!
 
I'm going to go over my my local Micro-Center to purchase a larger case. I'm sure I would be okay with the temps I have, but I'm very picky and my computer is my baby; I want to make sure it has all the best, and takes in plenty of fresh air to cool it's sexyness 

Quick play on Crysis 2 with the system in my sig to show u gpu usage:

Just because the gpu usage shows 99% doesn't mean it only has 1% more to give :P
A low gpu usage usually means bad drivers which u see more often on older games that are no longer supported by driver upgrades, it can also been a bottleneck of some kind with the rest of ur system and this is normally a cpu bottleneck.

Yea, they would be awesome temps with the stock coolers :P

51c on the bottom card is the highest i have seen since the rebuild, and temps are pretty warm here atm.
Ran out of MX4 so used MX3 on the bottom card but i have no clue if that has anything to do with it, could be application or just a hotter running card, top card has a VID of 1.0v where the bottom one has 1.038v so could be part of it too.

So I went out and bought a new case.
I also invested in a Corsaid H60 CPU water cooler (That wasn't worth the money; going to return it and put my Hyper 212 back on )
 
BTW, there's a front intake fan; x2 smaller HD intake fans; a 200mm Side fan (that I've been experimenting with intake vs. exhaust); x2 top exhaust fans (I only have 1 installed, closer to the CPU / RAM); and a rear exhaust fan.
I've also been switching intake / exhaust on the rear fan because I installed the water cooler and I read that it's supposed to be intake from the rear to cool the radiator with cold air, but I haven't really found the best way to position the rear fan yet.

Anyways, I did some stress testing on my CPU / GPU: Intel Core i5 2500k @ 3.3ghz (OC'd to 4.0ghz using the bios turbo)
Prime95 - 40c idle / 60-65c full load (using the most stress / heat / power settings) - This actually surprised me, I thought I was going to get WAY better temps using a water cooler. I tried switching the radiator fan to intake and exhaust and it only changes the temps maybe 3-4c. With my CM Hyper 212 I was getting just about the same temps, sometimes even COOLER.
 
Stress Test on my EVGA GTX 560 Ti SC (SuperClocked edition) - 900 / 1800 / 2106
EVGA OC Scanner (Using the "GPU Stress Test" for 5 minutes): 38c idle / up to 85-88c @ 95-99% load.
I'm using a customer fan profile - fan speed matches temperatures ( 40c/40%; 60c/60%; 80c/80%; etc...)
Now this was really weird because when I first put the computer together in the new case, I tested my temps and was happy because it wasn't running about 75c full load. I have no idea what has changed between then and now, but I'm getting over 10 degrees more than before.
 
EVGA OC Scanner (Using the "Start Benchmark"):
Test 1 (30 seconds) - 36c idle / 68c peak level @ 99% load
Score: 3330 (111FPS)
Test 2 (60 seconds) - 36c idle / 78c peak level @ 99% load
Score: 6670 (111FPS)
Test 3 (150 seconds) - 38c idle / 81c peak level @ 98% load
Score: 16330 (110FPS)
 
These temps just seem aweful high to me, considering the new case, and switching over to a CPU water cooler for great air flow. Is this normal? I'm still thinking about having NewEgg send me a new card so I can compare results between the two. Maybe I'll get lucky with the new card =\

More

I don't think using the 200mm fan on the side for exhaust would be something to do considering it's position is over the GPU itself. Perhaps make that intake so that it blows cooler air into and on the card to help keep temperatures down. If you have top case fans up by the CPU use them for exhaust above and out the back for the cooler you have for the CPU. You might be sucking across hot air because of the 200mm fan pulling air out of the case instead of pushing cooler air into the case. My theory is to push the hot air out by bringing in cooler air from the outside to replace and push the hotter air up and out of the top.
 
By the way, that is a wonderful case and good job on keeping the wires and stuff out of the way.
 
[edit]
I looked at my fan profile and I have mine currently set for 40% at start and then 60% at 40c, going up 10% for each 10 degrees from there after. Fairly aggressive for my card.
[/edit]

I don't know much about water cooling for CPU's. I use a Swiftech Polaris 120 air cooler for mine but the CPU I have goes from around 39-40c up to about 50 or so when playing games. I would have thought water cooling, even a closed loop system like the corsair one, would have more of an immediate effect if you use the right thermal paste for CPU's. The CPU one seems a bit high considering I just finished an i7-2600k system for a friend and his idle speeds on air with a Noctua cooler are around the same as mine and his even with prime95 doesn't get much hotter than about 61c or so. I would think that water cooling would be better.
 
[edit]
I took a look a the corsair install video for the H60 (which I think is the one you got) and they included a fan that is supposed to pull air into and through the radiator (if I am not mistaken).  Perhaps having the 200mm fan exhausting is a good thing in this case. I had considered using an H60 or even an H80 if I could afford one over the cooler I have but if the response as far as temps go then I'll hold off.
[/edit]

From what I understand about silver based thermal pastes is that there is a burn in period that you need to wait for in order for the proper curing of the paste. Give it a day or two and decide then if it doesn't drop your temps.