Guru3d has posted an article on taking four GeForce GTX 480 graphics cards and putting them into a 4-way SLI setup. Nvidia has recently released a new driver that enables 4-way SLI. However, there is just one small problem. The only “game” that supports 4-way SLI is 3DMark Vantage, which is not really a game. To prevent the CPU from being the bottleneck, they needed to overclock the Core i7 980 Extreme processor. But then they ran into a new problem. What motherboard would support 4-way SLI? They finally realized that they would need two Nvidia NF200 integrated circuits embedded on the motherboard in order to be able to accomplish this feat. How did they do it and what was their final score? Click the link below to find out.
Tom Peterson has the world record for the Supersonic Sled Demo from Nvidia. He has posted a video on YouTube showing some of the secrets of the demo. Anyone want to take him on and try to beat it?
Here’s a video showing you how to get the best scores on the new GTX 480 demo game, Supersonic Sled. The demo uses tessellation, PhysX and is, overall, just a cool game. After you download and play the demo, you have a chance to show up on the leaderbaord where we’re posting the top run scores automatically.
You download the demo here. You can see the leaderboard here.
Guru3d is at it again. In the past they have added extra voltage to a GeForce GTX 480 and a GeForce GTX 470 to see how far you could overclock them. Today, they take three GeForce GTX 480 cards and put them into triple SLI mode to see how well they perform. Take a look at that picture to the right and realize that it required a 1.2 Kilowatt power supply to power the system. Under load, the system consumed over 1 Kilowatt of power. Of course, overclocking the Core i7 965X to 3.75 GHz also adds to the power load.
But how did it perform? Was 3-way SLI worth it? Well, you will have to read the article to find out, although this little teaser should give you an idea.
Far Cry 2 did a little better, though 3-way SLI again is massively limited by the processor it at the very least scaled fairly nice. And 2-way SLI, well just look at 2560×1600… that’s roughly 1.9x performance over the single GPU GTX 480.
Nvidia’s GeForce GTX 480 has a default core clock speed of 700 MHz. The shader clock speed is 1400 MHz, while the memory clock speed is 3696 MHz. However, it was rumored that the clock speeds that Nvidia were trying to hit were much higher. What if you could attain those clock speeds? What would its performance have been? Guru3D takes a look at adding more GPU voltage to the GeForce GTX 480 and overclocking the GPU core. The problem with this is that it adds a considerable amount of heat. Therefore, they recommend either setting the fan at 100%, or to watercool the card. What did they achieve? They were able to achieve a core clock of 857MHz, a corresponding shader clock of 1713MHz and a memory clock of 4512MHz. Hit the link below to see how that translates into high framerates in games.
The GeForce GTX 400 series definitely has been topic of much discussion. Though extremely powerful this graphics card runs hot and is rather loud. Now these sure are lacklustre factors .. but that doesn’t mean the product doesn’t have any further potential. Contrary as today’s article will show there’s in fact a lot of additional performance to be found in the GeForce GTX 480, that’s if you know how to tweak it properly.
Nvidia recently held the Nvidia Game Festival (NGF) in Shanghai, China. This festival was to promote the launch of its GeForce GTX 480 and 470 graphics cards. Digitimes is reporting that NVidia is happy with the yields that they are getting. Of course, they do not say what those yields actually are.
Commenting on recent market rumors the Nvidia’s GTX 480 and 470 GPU yields are only about 20%, Drew Henry, Nvidia general manager of MCP business, refuted the rumors saying yields have met company expectations.
Guru3d has some pictures from EVGA running a setup of four, count them, four GeForce GTX 480 graphics cards running in quad-SLI. Can you imagine the power supply that would be needed to drive that thing? Let’s see here. Four GeForce GTX 480′s at 250 Watts apiece equals one kilowatt of power. And that does not include the rest of the system. While they can and do make power supplies up to 1400 Watts, I imagine that they actually used two different power supplies. That is just sketched out using a back-of-the-napkin approach. Click through the link below to see another image of the setup.
As you can see the setup is up and running, but no benchmarks are shared just yet. It certainly will be very interesting to see how well the scores scale. The motherboard used was a eVGA 4-Way SLI classified motherboard. Anyway have a look at some photo’s.
[H]ard|OCP is back with a video showing the real world temperatures and sound for a GeForce GTX 480. Click the link below to read their conclusions about the latest and greatest card from Nvidia.
Our goal here is to show you what Fermi should deliver in terms of heat and sound in a real world gaming scenario, so what we did was take our GTX 480 cards off the open test bench, and installed them into a chassis that I used previously for my personal system. We show you what GTX 480 SLI and single card GTX 480 do in terms of temperatures and fans speeds in a real chassis playing real games. Below we will share our thoughts on noise levels experienced and our GPU system temperatures. The video below covers our test setup and how we went about gathering our data and will let you see what we saw.
Since the GeForce, Quadro, and Tesla products are all based on the same graphics chip from Nvidia, one of the things that people have done is to buy the lower cost product (GeForce) and use it like a higher cost product (Quadro and Tesla). Sure you may have to give up on a few features, but not everyone needs them. Sure you may have to give up some of the extra memory, but it may be worth the price. This is a trade-off that purchasers have to make. Hexus has found yet another trade-off, and one that I had not heard of yet. The GeForce GTX 480 will have 1/8th of the speed as it does in single precision. That may not be a deal breaker to some people. After all, the GeForce GTX 285 had a similar problem. But for those expecting the double precision speed to have increased to ½ that of single precision, it does make a difference.
As much of a general-purpose computer as a GPU, the parallel architecture is also designed for the high-performance computing segment in mind. The enhanced cache structure, detailed above, helps with general computations, and GF100’s adherence to the IEEE 754-2008 floating-point standard means that it can run high-accuracy tests (double-precision support) at an increased rate when compared to anything NVIDIA has designed before.
Delve a little deeper, handily not mentioned in any briefing, and NVIDIA is limiting the double-precision speed of the desktop GF100 part to one-eighth of single-precision throughput, rather than the one-fifth speed of the Radeon HD 5000-series. We’ll have to wait for the Tesla parts before that’s restored to the one-half speed the GF100 is capable of.
Nvidia has released footage of their keynote presentation where they announced the GeForce GTX 480 and 470. The second part of the footage is after the jump.
LegitReviews is reporting that XFX has designed a limited number of Radeon HD 5970 graphics cards with 4GB of on-board memory. This is simply a Eyefinity6 version of the Radeon HD 5970. The production run of the cards is to be numbered in the hundreds, with a retail price of around $800.
But there is some other news in the post from LegitReviews as well. This time it concerns XFX and its support, or lack thereof, of the Nvidia GeForce GTX 480.
This afternoon we received confirmation that XFX, a division of PINE Technologies, will not be releasing any GeForce GTX 400 series graphics cards to the market when the cards become public next month. XFX was not listed as a launch partner for Fermi and did not issue a press release about the upcoming cards, which might come as a shock to many of our readers as they are one of the largest NVIDIA add-in board (AIB) partners in the world! XFX said that the decision not to carry this series of GF100 graphics card was their decision and that they will still be carrying NVIDIA products.
Comments