| ASUS ENGTX480/2DI/1536MD5 GeForce GTX 480 |  |
| Reviews - Featured Reviews: Video Cards | |
| Written by Olin Coles | |
| Friday, 01 October 2010 | |
|
Page 1 of 19
ASUS ENGTX480 GeForce GTX 480Benchmark Reviews tests the GF100 Fermi GPU six months after its launch, and is reminded that refined construction process and mature firmware can make a difference.Back in late March (2010) when Benchmark Reviews revealed NVIDIA's GeForce GTX 480 video card, Fermi fever was hot and expectations were unreasonably high. After months of wild rumor and inaccurate speculation, the enthusiast community was finally able to purchase the worlds most powerful single-GPU graphics card. As this article is written, it's been half a year since that launch and the landscape has hardly changed. NVIDIA still sells the most powerful graphics processor made, and the competition has resorted to a low and middle-market focus on value over pure performance. Fortunately for NVIDIA, their launch of the GeForce GTX-460 and GTS-450 product lines have made the value play extremely difficult for AMD. Yet, for a select few with money set aside for elite-level graphics, the top-end GeForce GTX 480 video card still offers good reason for purchase. In this article, Benchmark Reviews tests the ASUS ENGTX480/2DI/1536MD5 GeForce GTX 480 video card. This GF-100 GPU Fermi-based graphics card is the key to many things: high-performance 3D-Vision and 3D-Vision Surround, unmatched PhysX potential, and one part of an unbeatable SLI set. NVIDIA Forceware driver optimizations and firmware refinements have turned the GTX 480 into the video card it should have been from day one, reducing operating temperatures over the original production samples and improving power consumption and fan noise. Priced at $486, the ASUS ENGTX480/2DI/1536MD5 empowers DirectX-11 video games to deliver unmatched geometric realism. Our benchmark tests compare 3D frame rate performance with the ASUS GeForce GTX 480 in single-card and SLI-modes against some of the most powerful graphics products on the market.
ASUS ENGTX480/2DI/1536MD5 Video CardAt the center of every new technology is purpose, and NVIDIA has designed their Fermi GF100 GPU with an end-goal of redefining the video game experience through significant graphics processor innovations. Disruptive technology often changes the way users interact with computers, and the GeForce GTX 480 graphics card is a complex tool built to arrive at one simple destination: immersive entertainment, especially when paired with NVIDIA GeForce 3D Vision. Benchmark Reviews tests graphics frame rate performance of the ASUS GeForce GTX 480 using several of the most demanding PC video game titles and benchmark software available. Old favorites such as Crysis Warhead and PCMark Vantage are all included. New to the scene are Aliens vs Predator, Battlefield: Bad Company 2, Mafia II, Metro 2033, BattleForge, and the recently announced Unigine Heaven 2.0 benchmark. TSMC, the largest semiconductor foundry on the planet, has had a great deal of difficulty shrinking their business. Originally intended to feature 512 CUDA cores, NVIDIA was faced with limited yields from TSMC, and decided to end the consumers wait and offer gamers and PC hardware enthusiasts a 480-core solution. Powered by 48 ROPs and 480 unified CUDA (shader) cores, the GF100 Fermi GPU has 3.2-billion transistors to help process DirectX-11 commands and render some of the most detailed graphics ever seen on the PC platform. Tessellation is the word for 2010, and DX11 brings movie-quality graphics to life on consumer-level video games.
Manufacturer: ASUSTek Computer, Inc. Full Disclosure: The product sample used in this article has been provided by ASUS.
|
|
Comments
I'll probably go for a pair of GTX460's in SLI configuration myself. So far, two of them cost less than a single 480 does, and their performance together is knocking on it's door. Also, two of the 460's draw far less power than one 480 does. I'll take a chance and assume that they'll also produce less heat because of their low power usage.
Heat and power is important to many of us as you said above, I feel that it's a lot of money to buy one and will also cost a lot to use it over the lifetime of the card. It is a truly impressive video card though. Thanks for another detailed and informative review.
Surprised ASUS didnt do anything with the cooler though.
And point out all the strong points of nVidia, not that i have anything aginst those strong points, I have my self a 3x SLI GTX480 + dedicated GTX280 for PhysX + 3D Vision, on a 3 screen setup.
But not because the 480 is the best card, but because it scale's mouths better in SLI 3x then CF-X 3x.
And even tho they are really fast, they ware also very irritating loud and hot before i installed water cooling on them.
That said, if i would buy a single card i would for shore go for ATi as they run cooler and uses less power.
Looking at all the articles it looks like they ware put together whit the help of a nVidia PR guy.
And if BMR want to be taken seriously, more balanced reviews would be welcome.
That's the way the computer industry works, when new designs get released, they are generally a LOT better thatn last year's model. Ever hear of Moore's law? When ATI brings out their next generation of video cards, I expect them to be better than what is available today. And when we test them, if that is indeed the case, that's what we will report. And then someone will complain again that we're getting paid off by ATI. As if...!!!!
How can you seriously expect me to review an AMD Radeon product when the last video card they offered was the Radeon HD 5550 almost five months ago? You might also go back and look over all of MY article, and count-up who has received more awards.
If you want your remark taken seriously, perhaps you should be more constructive. Otherwise, you just come off as another fanboy troll with poor spelling.
Of course he's a fanboy or a troll, but you don't need to be a Grammar Nazi about it!
It's very hard to take you seriously when your post is filled with simple spelling errors and grammatical mistakes. If you disagree with the results of the tests, then it is up to you to test for yourself. Many of the benchmarks used are free or have free versions. Anyone who can afford a setup with 3 GTX480s in SLI can certainly afford to do their own testing.
Hank
Tinfoil hat time.
LOL
Great review. I wanted to know if you guys measured the GPU voltage (in MSI Afterburner or Asus Voltage Tweak software) to check if the more mature 40nm process resulted in lower GPU voltage at load? If so what was the difference?
Also, since this is only 1 representative sample, how can you be certain that the more mature 40nm manufacturing process is the reason for the reduced demand? Could it be an outlier videocard?
Yes, it could always be one lucky sample or it could be this way for every sample. It's very difficult to verify, but the mere fact that one card could ever reach temperatures or power consumption this low is surprising.
I have not conducted the GPU voltage tests you mention.
NVIDIA GeForce GTX 480 Engineering Sample
MSI Afterburner reported 0.962V GPU at idle, and 1.025V under load.
GPU-Z reported 0.953V 15.0A 14.3W idle VDDC, and 0.980V 75.0A 74.0W at load.
ASUS ENGTX480
MSI Afterburner reported 0.962V GPU at idle, and 1.075V under load.
GPU-Z reported 0.955V 11.0A 10.5W idle VDDC, and 1.033V 70.0A 72.3W at load.
Still what you measured is an observable result. I look forward to future GTX480/470 reviews, where you can provide more data points :)
No mention of voltages either.
In no way have your test proven that these cards are running cooler now compared to before.
Hope no one makes the mistake of buying one of these cards now if they wouldn't before thinking the heat and noise issues are resolved, because this 'review' proves nothing of the sort.
I will pull the article offline and start taking orders from you when this becomes your website. Until then, your opinion matters as much as the next anonymous post.
Voltage is absolutely something you need to give in this article, because voltage is directly related to heat output. When GTX 480 launched there was quite a bit of voltage binning (same thing with HD5870 even), where some cards were binned higher or lower. Specs remained the same, except for voltage, heat, and power consumption. Assuming absolutely no change, it's entirely possible to randomly get three cards successively binned with lower voltage.
Now, that's not saying that TSMC isn't making advances on 40nm, I'm sure they are, and obviously any kind of decrease in defect density will be amplified by larger dies, but when you already had such variability in volts, heat, and power at launch, you need a lot more than three data points to draw a solid conclusion. Someone could have easily gotten the same data from three random cards at launch.
Obviously the more cards you use the stronger the probability is (again assuming those added cards support your original data), however as I said you could have randomly taken three cards at launch and gotten the same results thanks to the large amount of voltage binning.
About the only way that I could see for easily proving that GTX 480's are getting cooler is if new chips start using lower voltage than the lowest voltage that was available at launch. Otherwise with this type of problem you're dealing with statistics. You don't "prove" anything so much as provide probability. With a high enough probability you can begin to draw conclusions.
"but it doesn't appear that AMD has any surprises for the upcoming holiday season"
Umm, Cayman, Barts etc. These are coming up in the next couple of months and as we dpont have benchies, regardless of performance, they'll be a surprise (bad or good). And if Cayman improves like it's 'rumoured' too, it will eclipse the 480.
Also, there was a 512 core GTX 480 out in the wild ##brightsideofnews.com/news/2010/8/13/512-core-geforce-gtx-480-reviewed-any-surprises.aspx and it wasnt very impressive.
Gf100 good- yes no doubt, comment on no surprises from AMD - quite ignorant.