Slashdot Mirror
Nvidia 480-Core Graphics Card Approaches 2 Teraflops
An anonymous reader writes "At CES, Nvidia has announced a graphics card with 480 cores that can crank up performance to reach close to 2 teraflops. The company's GTX 295 graphics cards has two GPUs with 240 cores each that can execute graphics and other computing tasks like video processing. The card delivers 1.788 teraflops of performance, which Nvidia claims is the fastest single graphics card in the market."
It's an allusion to Back to the Future, where Marty goes back to the 50's and tells the doctor that the Delorian time machine requires "one point twenty-one gigawatts" to make the leap. Only back in 1985, the SI prefix "giga" wasn't well known, so presumably the actors or directors in the film arbitrarily, or by following french language convention, decided to pronounce giga with a soft g, hence the line "1.21 jiggawatts" which sounds a little out of place in 2009.
I can run Crysis/Warhead at 30fps maxed out at 720p. I have a single 4850.
The problem with video card review is they don't bother testing anything lower than 1920x1080 which is 2.25x bigger than 720.
Crysis takes a lot to run but it has already been tamed as long as you aren't running at 2560x1600 or some other absurd resolution.
It's 12" long and 6" around and it's going to go straight up your ass.
218 GFlops
http://www.realworldtech.com/page.cfm?ArticleID=RWT072405191325&p=2
A single 8800 kill the cell and the video processor in the ps3 combined
One of the benefits of the technology war is that it produces good midrange and low end technology as well. This is particularly true in the case of graphics cards since they are so parallel. They more or less just lop off some of the execution units and maybe slow down the clock and you get a budget card.
Whatever your budget is, there's probably a good card available at that level. Now will it be as fast as the GTX 295? Of course not. However they'll be as fast as they can be at that price/power consumption point.
Don't pitch because some people need/want high end cards. Enjoy the fact that they help subsidize you getting good, cheap midrange cards.
If you want serious suggestions, tell me your budget range and what you want to do and I'll recommend some cards.
Because this card can only do 1.788 tera-multiply-adds per second. Try instead to have it build a parse tree, then run transformation algorithms on it (chasing pointers all over the place) and so on, like you would while compiling code, and this thing will make the Atom look great.
CPUs are optimized for general computing, GPUs are optimized for stream-oriented numeric computing. Both have their uses, and the ideal is probably a combination of both, as is currently done.
What the fuck are you smoking? It's a $500 card.
The specs are very specific (lol, get it?).
I take it you havn't seen full-length graphics cards yet? 280's, 8800 GTX's, GX2's, etc, aren't full length cards, but they're close.
These are full length cards: http://management.cadalyst.com/cadman/Review/AMD-ATI-FireGL-V8600-and-FireGL-V8650-Graphics-Car/ArticleStandard/Article/detail/526886?contextCategoryId=6631
You can tell the difference by them not only being longer, but having that retention connector at the end (right side of the pictures) which helps steady the card.
61C on a video card isn't much to worry about. Using RivaTuner I used to watch an 8800gts creep up to 90C and it never died. Unfortunately, taking comfort in knowing your video cards won't get cooked isn't very useful when you're worried about the other nearby devices. For what it's worth my old temps were in an antec sonataII case. When I switched over to the antec 900 my 8800gts temps dropped to the 55-60C range. Maybe a new case is the aftermarket part you're looking for?
Apart from, you know, link length.
The most important thing to understand is that these aren't actually 'cores' in the same sense that your Core 2 Duo has two of them. They're shader units. It works more like SIMD than parallelization, only instead of something like SSE that can perform a single operation per clock across 4 packed floating point values it performs the operation on thousands of them.
If they could slap a billion or a million or even a thousand shader units on a card without actually reducing performance they would, but they can't. At a certain point the bottleneck becomes link length. You can overcome it by increasing voltage but then heat becomes the issue. This is a large part of the reason transistor count is tied to transistor size. NVIDIA isn't "failing" in this respect, they're just succumbing to the laws of physics.
If they could improve performance by slapping 20 or 4 or even 2 of the *actual* cores on each card they would, but they can't. Because it's not an actual processor, it doesn't have fancy features like three levels of cache and a TLB and branch prediction and out-of-order execution. But even if they were engineered to work this way, you can't improve PC performance by slapping in a thousand Core 2 Quads either. A part of the reason Xeons have so much cache is so you can mitigate the penalty of having 8 processors using commodity RAM, but eventually you run up against that bottleneck. Shared resources become saturated much faster than most people expect.
The most efficient way of improving graphics performance is with SLI because you are replicating all of the hardware, the memory and the bus the *actual* core depends on. For the exact same reason, you can extract the most performance out of each CPU core by putting each one in a different machine.
Apparently the US National Bureau of Standards decided in the 1960s that Jiggawatt was the one true pronunciation.
http://en.wikipedia.org/wiki/Giga
And jif is only correct for the same reason, the developers decided "Choosy developers choose Jif" was a hilarious slogan they could use internally for the gif format.
http://en.wikipedia.org/wiki/Gif#Pronunciation
So yes, Jigabit, Jigabyte, Jigawatt, those are how we are legally supposed to pronounce them, atleast in the US.
I don't suffer from insanity, I enjoy every minute of it!
Not even close. A single ATI HD4870X2 card has 2.4 TFLOPS or processing power: 2 (instr/clock with MAD) * 800 (Streaming Processors) * 750 (MHz) * 2 (GPUs) = 2.4 TFLOPS.
That's gross; you're gross.
everything in moderation
I thought the Radeon HD 4870 x2 did 2400 GFLOPs, i.e. 2.4 TFLOPs?
I'm not saying the Radeon is more powerful, but in terms of FLOPs, it is (I thought), faster.
http://hothardware.com/Articles/NVIDIA-GeForce-GTX-295-Unleashed/
The spec and the creator says you're wrong.
Do you have sources or do you just like telling people they're wrong without any data to back it up?
Here's another source if you trust wikipedia more than random webpages that can't be edited by half the world: http://en.wikipedia.org/wiki/Gif#Pronunciation
I touch computers in naughty places
Most likely your HVAC unit has a heat pump and not just a resistance heater. If so, it pulls heat in from the outside. They usually have a CoP of between 2-3 which means that for every kWh of electricity used it puts between 2-3 kWh of heat in the house. If you use just a resistance heater, though, then you might as well run your computer.
See http://www.pcworld.com/businesscenter/article/155242/inside_tsubame_the_nvidia_gpu_supercomputer.html, for example.