Slashdot Mirror
NVIDIA Makes First 4GB Graphics Card
Frogger writes to tell us NVIDIA has released what they are calling the most powerful graphics card in history. With 4GB of graphics memory and 240 CUDA-programmable parallel cores, this monster sure packs a punch, although, with a $3,500 price tag, it certainly should. Big-spenders can rejoice at a new shiny, and the rest of us can be happy with the inevitable price shift in the more reasonable models.
A video card I can't use on XP32 since it can't properly allocate that much VRAM & system RAM at the same time.
There is a war going on for your mind.
Does this mean we can finally run Crysis now?
Kwisatz Haderach
Sell the spice to CHOAM
This Mahdi took Shaddam's Throne
Yes, AMD's Stream technology. I don't think it is used as much as CUDA in practice.
... "the most powerful video card in history", it's "the most powerful videocard yet".
[/pet peeve]
Or maybe there are companies that need high end cards with 4GB of RAM. This isn't some trick to get consumers to pay more for a low end card. This is now Nvidia's highest end workstation card.
I read that as 4*MB* video card.
I fucking hate the beginning of work weeks.
Working hard, I see.
Swedish plasma phys. PhD student; MSc EE; knows maths, programming, electronics; finance interest; seeks opportunities
I don't believe anyone claimed this was a gaming card.
This is a scientific number cruncher. Its use is in visual computer modeling for anything from weather models to physics models.
How about folding@home? this does it faster than any computer on the block.
All of you kids making jokes about crysis are missing the point. This might run games, but it's a science processor first.
They're using their grammar skills there.
excuse me but this is total bullshit. oldest trick in the book. if you are behind in technology, pop out a card with huge ram and try to get some sales.
lets face it. nvidia has fallen behind ati in the chip race. you can place any number of 4870s in a setup as much as you like to equate the power of any monolithic nvidia card and they always kick the living daylights out of that nvidia card in terms of cost/performance per unit of processing power.
In case the $3,500 price tag didn't tip you off, this isn't a gaming/enthusiast card. This is a Quadro - a professional card for high-end 3D rendering. Stuff like generating film-grade 3D or insane CAD stuff. Actually, due to the design of the card, it'd be pretty horrible at playing games.
This thing is aimed at high-end scientific calculation and professional-grade rendering.
ATI may, or may not, have something comparable. ATI may even have something better. I don't know, I don't follow the GPU industry very closely. But claiming that they're just slapping a bunch of RAM on a card to drum up sales is just plain wrong. Hell, the blurb here on Slashdot even mentions the fact that it has 240 cores.
"Work is the curse of the drinking classes." -Oscar Wilde
These were being sold in the first half of August for 10500$ - containing 2 of those cards. Only 3 months late.
"Violence is the last refuge of the competent, and, generally, the first refuge of the incompetent" - Thing_1
Really, people. If you're going to buy such an expensive professional card, you're going to go with a professional-grade operating system, which will of course be 64-bit.
Hail Eris, full of mischief...
E pluribus sanguinem
Coder Hate like that brought by the shitty, bug filled drivers that ATI has a long history with?
I think ATI/AMD is on the right path, but they have a long history of being on the wrong path, while NVIDIA has always been more towards the middle (Not completely right, but not too badly wrong). It'll take some time before I jump to the ATI Bandwagon as completely as you obviously have.
That's times more powerfull and flexible than CUDA.
I like how statistics are so meaningless we're not even putting the numbers in anymore.
Haida Manga
There is no upper limit on the amount of memory required for tasks like volume visualisation, where you have a nice big 3D cube of data in 16-bit format. A cube 1024 voxels in each dimension with a single channel of 16-bit data (2 bytes) is going to be 2 Gigabytes. You will need at least two such cubes to do any sort of image processing work.
Even a digital movie can be considered to be a cube if you consider time as the 3rd dimension.
Rather than having cards with a fixed amount of VRAM, which can't manufacturers just put a bunch of memory card sockets on the card and allow users to add memory when they want?
Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
In case the $3,500 price tag didn't tip you off, this isn't a gaming/enthusiast card. This is a Quadro - a professional card for high-end 3D rendering. Stuff like generating film-grade 3D or insane CAD stuff.
Cm'on, we are all grown ups here. You can say it clearly:
It's for high-detailed 3D virtual porn.
factor 966971: 966971
meaning you can code directly the hardware
Guess what CUDA and Stream have been designed for? Yes: for programming the hardware. What you suggest is pure insanity. NEVER EVER touch hardware directly from an userland app. And once you start writing a kernel module, you end up with something like CUDA/Stream anyway.
I am a coder, and quite frankly I couldn't care less about nvidia drivers being closed source. They are MUCH better than the ATI ones, especially in the OpenGL department. nvidia whipped up a beta GL 3.0 driver in less than a month since GL3 specs were released. ATI? Nope. New standardized feature X is added to the registry. nvidia adds it pretty quickly; ATI adds it months, even years later. nvidia drivers are also pretty robust; I can bombard them with faulty OpenGL code, and they remain standing. With ATI's fglrx, even CORRECT code can cause malfunctioning.
THESE are the things I care about. Not the license.
This sig does not contain any SCO code.
But claiming that they're just slapping a bunch of RAM on a card to drum up sales is just plain wrong. Hell, the blurb here on Slashdot even mentions the fact that it has 240 cores.
Umm, the GeForce GTX 280, a gamer card released last summer, also has 240 "cores" (as Nvidia counts them; actually stream processors).
This workstation card, as you might expect, is essentially the same thing as the consumer card, just tweaked towards the professional market (more RAM, different drivers). It's nothing especially innovative.
This is a new age of statistics. Instead of putting numbers up there that could be misinterpreted, the author has chosen to take the politically correct route and allow the user the decision that best pleases them.
This way you get your message out, and the person on the other side is happy with their decision. It's a win-win!
Every time I start to have faith in humanity, I ruin it by driving to work between 7 and 8 am.
Two reasons:
One is simply that the cards use memory that isn't available normally. They don't use normal DDR RAM, they use special RAM for graphics cards, called GDDR. It is similar but not the same as memory in systems. Thus you can't just go out and buy sticks of RAM for it. So they'd have to be made special for the cards (and each gen of card uses different RAM), and thus would be expensive.
The bigger one is that the RAM is really pushed to the limit. You start to run in to all sorts of shit you never thought about. The electrical properties of the connection are highly important and there is a difference between what you get soldered on to traces and in a socket.
It's a nice thought, but not practical these days. Graphics cards are heavily dependent on high RAM bandwidth and you get that by really pushing the envelope. That means new RAM technologies all the time and the chips being pushed to the max.
Do you realize that for computers 12+ years is several GENERATIONS?
I had always been using ATI for Windows boxes and laptops, since my main concern was almost always video performance and TV-Out capability and I could not even get a video overlay work over TV-out with nVidia cards for years.
Of course, when I had problems with linux drivers I built nVidia (I admit, even intel) linux boxes. But that is a thing of the past, I am back to ATI for linux, they are good and even getting better with each release.
Anyway, long term loyalties is pretty silly. I bought my K6 233 at the same price my friend bought his MMX 166, in retrospect we all know how those two compare. I kept on buying Athlons when others were paying more for their crap P4's (they weren't called crap back when it was the best intel had to offer). But, hey, I am now buying Core 2 for non-low end systems, until AMD can come up with something better.
Fanboyism gets you bad deals at least half of the time. You buy hardware, you don't marry it. Ok, I know this is slashdot and the last statement might generate some debate, but anyway you get the point.
Violence is the last refuge of the incompetent. Polar Scope Align for iOS