KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
Today we are testing the card in a Intel Core i7 990x system, which is CPU watercooled and running at 4.8ghz.
Test System:
Processor: Intel Core i7 990x @ 4.8ghz.
Cooler: Danger Den Watercooling.
Motherboard: Asus Rampage III Black Edition.
Memory: 12GB GSKILL 2133mhz Memory.
Power Supply: Corsair AX1200W.
Hard Drives: Patriot Pyro 120GB & 2TB Samsung hard drive.
Optical Drive: Asus BluRay drive.
Graphics: Sapphire HD6990 with reference cooler and Arctic Cooling Accelero Twin Turbo 6990.
We are testing today at the higher bios setting of 880mhz. Remember that the cards won't be covered under warranty at this setting. We wanted to stress both cores as much as possible.
Room ambient was maintained at 24c throughout testing to ensure accurate results.
At idle, the reference cooler holds a temperature of around 40c. The Accelero Twin Turbo 6990 cooler drops this by 6c to 34c.
Under load, the temperatures are reduced dramatically, by 20c-23c , which is a huge difference and will help ensure maximum PCB longevity. We decided to measure ‘return to idle' temperatures, highlighting cooler performance.
The video above was taken from our launch review of the HD6990 in March, it shows how ineffective the AMD reference cooler is at dissipating heat.
A return to idle figure of 26 seconds is a class leading result for the Accelero Twin Turbo 6990. The reference cooler takes around 2 minutes to achieve the same results, running close to 20c hotter throughout the curve.
We have built a system inside a Lian Li chassis with no case fans and have used a fanless cooler on our CPU. We are using a heatpipe based passive power supply and an Intel SSD to keep noise levels to a minimum. The motherboard is also passively cooled. This gives us a build with completely passive cooling and it means we can measure noise of just the graphics card inside the system when we run looped 3dMark tests. Ambient noise in the room is around 20-25dBa. We measure from a distance of around 1 meter from the chassis and 4 foot from the ground to mirror a real world situation.
Why do this? Well this means we can eliminate secondary noise pollution in the test room and concentrate on only the video card. It also brings us slightly closer to industry standards, such as DIN 45635.
We load the card with FurMark.
KitGuru noise guide
10dBA – Normal Breathing/Rustling Leaves
20-25dBA – Whisper
30dBA – High Quality Computer fan
40dBA – A Bubbling Brook, or a Refridgerator
50dBA – Normal Conversation
60dBA – Laughter
70dBA – Vacuum Cleaner or Hairdryer
80dBA – City Traffic or a Garbage Disposal
90dBA – Motorcycle or Lawnmower
100dBA – MP3 player at maximum output
110dBA – Orchestra
120dBA – Front row rock concert/Jet Engine
130dBA – Threshold of Pain
140dBA – Military Jet takeoff/Gunshot (close range)
The reference card can generate a lot of noise under load, especially when running Furmark, with the small fan spinning at an incredible rate. The Arctic Accelero Twin Turbo 6990 on the other hand has a dual 120mm fan system and therefore they can spin much slower, while generating significantly better air flow. At 1,500 rpm the noise levels peak at just over 33 dBa, barely audible in a performance gaming system. Fantastic results.
very impressive beast for sure
excellent cooler, shame about the ridiculous pricing. thats more than some low end video cards. decent video cards too !
Looks brilliant, I suppose if you could afford a HD6990 in the first place that something like this would be worth it, I would rather aim a bit lower down the food chain myself. a lot of cash for a GFX.
Awesome, if a little expensive IMO.
Very glad to see they did away with the seperate little heatsinks for memory etc, those were a nightmare. I had 2 of mine drop off over time in the c ase.
I ordered a 6990 last week and i should get it this coming week if the stock is still holding ok.
Only problem I have with this is the price. I think AMD should have done a better job out of the gate, rather than allowing these to be made in the first place. I dont know why companies like AMD dont work with partners to release a proper version of the cvard in the first place.
Bit disappointed with the reference cooler on this, thats shoddy.
OMG Zardon, you are the man! Not a week after owning a 6990 and already a wicked review. I am definitely buying one though I do have a question which affected my 5970 Accelero Xtreme (equivalent of this)….does this cool the VRM’s? In Gpu-z the would get really hot while the gpus themselves were cool and that was the only downside to the cooling.
Yeah there is voltage regulator cooling with this one. they also seem to have done a better job by making the central platform cooler section part of the main piece, although it ‘floats’. Also all the thermal padding looks in the right spots for component cooling. I hated those individual heatsinks with paste. what a cock up that was.
Very impressed myself, might pick one up as I have a 6990 and it drives me friggin crazy when im gaming.
AMD need shot for their coolers, they are pants.
Excellent product, although it doesn’t solve something that upset me with the majority of after market video cards coolers – you can’t replace the fans! With this one though it seems pretty easy to take those off and zip-tie 2 Gentle Typhoons AP-15….any change of testing that Zardon?
Hi John, If I get some time I can try. What fans are you interested in specifically?
@Zardon
I’m thinking established fans like Gentle Typhoon AP15s, some Yate Loons at higher fps and just for funzies some Deltas……you take your pick 😉
Thanks for the review, I think this is the first one out there. I installed this today in my Dell 435t (I’ve also added an SSD drive and an HX850). Took some wiggling to get the side of the case back on, not much space to spare at all, but it got back on after removing a metal cross bar and a useless plastic piece.
This cooler’s fans are very quiet, even when set at 100% in Catalyst. However, when playing BC2 maxed out @ 2560×1600, either the SYS1 fan or the CPU fan are kicking up louder it seems, and seems like a lot of hot air is now exiting from the top of the case near the power supply instead of formerly (when using the stock cooler) at the bottom of the case, where the 6990 is. I realize I should have a better case w/ better ventilation, but seems like the cooler is keeping more hot air in my case, or at least significantly changing the air flow. I guess I could switch out some of the Dell stock fans, too. Anyway, it’s still a big improvement in noise over the leafblower stock cooler, even with the other fans kicking up.
I bought one too, and run it at 100% in CCC as it is quiet even at that. You do need good case cooling as more heat gets trapped inside the case.
Several good exhaust fans at rear of chassis are important to help with airflow with one of these installed. HD6990 produces a ton of heat.
@Will
If your case has top mounted PSU it fails from the start. Also did i understand right that you’re using the stock cooler for the CPU? Seems to me like you have 2 more “upgrades” to do: case and CPU cooler.
Yeah, the case sucks ventilation wise but the stock MB has some proprietary connections and I’d rather just get a whole new non-locked MB if I’m going to switch cases, and I’d rather not spring for a new (end of life) X58 board or both a new 1155 board and processor at this point. I think my best option now w/o getting a new case is to replace the non-CPU cooler stock fans. There’s a 92MM fan sitting right below the power supply that I think I’ll replace with a Kama Flow 2 fan, and a fan in the front bottom of the case that draws air into the case.
Re the stock CPU fan, I think the MB’s screws for the stock CPU fan are slightly off standard spec. Also, the stock MB locks the i7 920 to 2.66 ghz, so the processor isn’t running terribly hot.
Update in case anyone is installing this in a 435t/xps 9000. I replaced the stock cpu cooler with an arctic freezer 13. The push pins would not go down into the MB far enough, so I used regular screws on the mounting bracket instead (similar size to the stock cpu cooler’s screws), and this worked fine and eliminated a lot of noise.
Also, my second GPU was overheating when even barely used (97 degrees celsius easy), and I figured out this was caused by there being nothing but bare case floor beneath the bottom 3 PCI slots where the arctic 6990 must go, and the card was leaning too far down because of its weight. I solved this by using a twist-tie on one of the power cables to create a lifting effect on the card, thereby giving the arctic 6990 clearance from the floor. Temps are back to normal ranges (max of about 62 degrees after running furmark benchmark), although the 2nd gpu still runs about 5 degrees hotter at full load.
Fantastic article. I’m going through several these difficulties.
Hi there! I could have sworn I’ve been to this website ahead of but soon after searching by means of a number of the article I recognized it is new to me. Anyways, I’m surely pleased I discovered it and I’ll be book-marking and checking again regularly!
Your type is so exclusive in comparison to several others. Thanks for publishing whenever you possess the option,Guess I’ll just make this bookmarked.two
I see that your VReg temps went up under Furmark from 102.7C with the stock cooler to 110.5C with the Accelero. That doesn’t seem too good. What VReg temps are other people with this cooler seeing?