Slashdot Mirror
Future of 3d Graphics
zymano writes "Extremetech
has this nice article on the future of 3d graphics. The article also mentions that graphic card gpus can be used for non-traditional powerful processing like physics. A quote from the article, "GPU can be from 10 to 100 times faster than a Pentium 4 and Scientific computations such as linear algebra, Fast Fourier Transforms, and partial differential equations can benefit". My question - If these cards are getting so powerful at computations then why do we need a Intel/AMD processor at all? Just make a graphics card with more transistors and drop the traditional processor..."
Just make a graphics card with more transistors and drop the traditional processor..."
Apple have done this several years ago. The Newton 2000 and 2100 didn't use a CPU but rather the graphics processor.
RST
The head of Nvidia was written about in wired a while ago and he essentially said the same thing.
He was like, our cards ARE the computer, and are becoming far more important then the CPU for the hard core stuff.
It was interesting, but I totally foo fooed it.
obviously he was smarter then me.
Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
... everybody would use his computer for 3D only, but I know a lot of people never do anything with 3D. And I don't think a computer for office-work benefits a lot of the GPU.
This is RiverTonic's sig.
GPUs work with limited precision -- IEEE single precision is typical. This is good enough for 3D graphics -- after all, in the end you'll be limited by the 10-11 bit spatial resolution and 8 bit color resolution -- but not good enough for most scientific problems, which typically require a minimum of double precision.
Simulating higher precision with single precision arithmetic is possible, but the performance penalty is too severe for it to be useful.
Tarsnap: Online backups for the truly paranoid
...This is what the future releases of DirectX is supposed to address: The use of 3D renderers to render non-graphical elements and other work.
:-)
Good for the end user, but going to be a pain in the ass for software developers to take advantage of, is my guess.
alias uptime="echo '5:33pm up 22342352324 days, 6:28, 2124315623 users, load average: 2432.40, 12312.31, 123123.19'"
Why buy a big processor when the only intensive computational tasks are video en/decoding and games, tasks that can easily be farmed off to other, cheaper units?
"To any truly impartial person, it would be obvious that I am right."
Because GPUs are specialized processors. They are only good at a couple of things: moving data in a particular format around quickly, and linear algebra. It is possible to do general-purpose calculations on a GPU, but that's not what it is good at, so you'd be wasting your time.
This is akin to asking why you shouldn't go see a veterinarian when you get sick. Because veterinarians specialize in animals. Sure, they might be able to treat you, but since their training is with animals you might find their treatments don't help as much as going to see a regular doctor.
Not only do some people not play games, but also GPU's do what they do so well because they are specialized. I think that if they were made to do general functions as well their efficiency would decrease. Also the comment from the nvidia guy about the graphics card doing most of the work. In a game there are still physics and AI, and overhead calculations that all need to be done. Not many or none of these are covered by the GPU.
I agree completely that offloading tasks from the CPU is good, look at the Amiga, that was an amazing machine for its time. And the a huge part of its power can be accredited to it's multiple, separate and specialized processors. I think in the future we will see a shift towards that again, as transistor increases become no longer feasible.
But then what do I know..
...Huh?
Graphics? I still use a VT100
CAn'T CompreHend SARcaSm?
GPUs are highly specialized. In graphics processing, you generally perform the same set of operations over and over again. Also, pixels can be rendered concurrently - as such, graphics hardware can be extremely parallel in nature. Also, in graphics hardware, there isn't much (if any) branching in code. Simple shader code just runs through the same set of operations over and over again.
"Normal" code, such as a game engine, compiler, word processor or MP3/DivX encoder does all sorts of different operations, in a different order each time, many which are inherently serial in nature and don't scale well with parallel processing. This type of code is full of branches.
To optimize graphics processing, you can really just throw massively parallel hardware at it. Modern cards do what, 16 pixels/texels per cycle? 4+ pipelines for each stage all doing the EXACT same thing?
Regular code just isn't like that. Because different operations have to happen each time and in each program, you can't optimize the hardware for one specific thing. In serial applications, extra pipelines just go to waste. Also, frequent branch instructions mean that you have to worry about things like branch prediction (which takes up a fair amount of space). When you do have operations that can happen in parallel (such as make -j 4), the different pipelines are doing differnet things.
Take your GeForce GPU and P4 and see which can count to 2 billion faster. In a task like this, where both processors can probably do one add per cycle (no parallelizing in this code), the 2GHz P4 will take one second, and the 500MHz GeForce will take four seconds (assuming it can be programmed for a simple operation like "ADD"). Even if you throw in more instructions but the code cannot be parallelized, the CPU will probably win.
Basically, since you can't target one specific application, a general purpose processor will always be slower at some things - but can do a much wider range of things. Heck, up until recently, "GPUs" were dumb and couldn't be programmed by users at all. I haven't looked at what operations you can do now, but IIRC you are still limited to code with at most 2000 instructions or so.
My server
My question - If these cards are getting so powerful at computations then why do we need a Intel/AMD processor at all? Just make a graphics card with more transistors and drop the traditional processor...
If you'd really like the answer to this question, try programming anything on the GPU and you'll understand. It's hell to do half this stuff. GPUs are highly specialized and make very specific tradeoffs in favor of graphics processing. Of course, some operations, specifically those that can be modeled using cellular automata, map well to this set of constraints. Others, such as ray-tracing can be shoe-horned in, but if you were to try to write a word processor on the GPU, it'd essentially be impossible. The GPU allows you to do massively parallel computations, but penalizes you heavilly for things such as loops of variable length or reading memory back from the card outside of the once-per-cycle frame update, and the price of interrupting computation is prohibitive. Clearing the graphics pipeline can take a long, long time.
Furthermore, while there have been a few papers published claiming the orders of magnitude increase in speed in these sorts of computations, none actually demonstrate this sort of speed-up. Everyone's speculating, but when it comes to it, results are lacking.
b.c
Using the power of the graphic subsystem to handle other kinds of calculations has been done for years, if not decade(s) by Silicon Graphics.
At least for the demos...
While optimized for graphics, GPUs can indeed be used as general-purpose processors. GPUs are effectively stream processors, a class of devices whose architecture and programming model make then particularly efficient for scientific calculation.
> It might take a real long time, but it is a general purpose processor and so can process anything
The same holds true for GPUs. Like CPUs, they are turing complete.
You mean like: this?
Now, that press release was about two years old, and you can bet that ATI has advanced beyond that point (though I can't provide details).
Also, while not integrating a serious 3D graphics GPU, there's no reason that this can't be done -- except one -- and the same reason that a powerful CPU isn't integrated: heat dissipation.
But, for a "media processor", it sure is sweet.
You could've hired me.
You keep hearing this logic every once in awhile.
Look, for the same price of a $400 graphics engine you can get yourself a dual CPU machine, a cheap graphic card with AGP, and do it in "software" with about the same efficiency, if you know what you're doing.
Because the extra CPU isn't inheritly multi-core like most modern GPUs, you need to compensate with a higher clock speed, and use whatever multimedia instructions it has to the fullest extent (ie altivec, mmx2, etc.)
But of course, the GPU is better suited to the actual drudge work of getting your screen to light up. If there's stuff to be computed and forgotten by it (i.e. particle physics), its probably better left decoupled to exploit parallism in that abstraction.
As you get to a limit in computational efficiency, you start adding on DSPs, and this is where FPGAs and grid computing start looking interesting.
So it shouldn't be considered suprising that these companies will say that; they can see that trend and they want a piece of that aux. processor/FPGA action. The nForce is a step in the right direction. They don't want to be relegated to just making graphic accelerators when they have the unique position to make pluggable accelerators for anything.
But to plan on packaging an FPGA designed for game augmentation and calling it a uber-cool GPU is just a marketing trick. This technology is becoming commercial viable, it seems.
Black holes are where the Matrix raised SIGFPE
Aloha!
You wrote: My question - If these cards are getting so powerful at computations then why do we need a Intel/AMD processor at all? Just make a graphics card with more transistors and drop the traditional processor...
Congratulations! You have just reinvented Ivan Sutherlands Wheel of reincarnation which is exactly about this: Normal CPU:s are enhanced with specific functions to provide acceleration for a common task, the enhancments are getting so big that farming them out into a separate chip/module seems like a good idea. The separate thingy grows in complexity as more flexilibility and programmability is needed. Finally you end up with a new CPU. And then someone says.... You get the idea.
Here is a good take on Ivan Sutherlands story. And here is Myers and Sutherlands original paper.
Read, think and learn.
Unfortunately, the researchers have all inexplicably been rendered deaf.
There's a lot of work being done on reconfigurable computing, which imagines replacing the CPU, GPU, DSP, soundcard, etc., with a single reconfigurable gate array (like an RAM-FPGA). You'd probably have a small control processor that manages the main array. On this array one could build a CPU (or several) of whatever ISA you needed, and GPU, DSP, whatever functionality was called for by the program(s) you're running at the current moment. Shutdown UnrealTournament 2009 and open Mathlab, and DynamicLinux will wipe out its shader code and vector pipelines, and grow a bunch of FP units instead. Run MAME and it will install appropriate CPUs and other hardware.
In the initial case, this would be controlled statically, a bit like the way a current OS's VM manages physical and virtual memory. Later, specialist "hardware" could be created, compiled, and optimised, based on an examination of how the program actually runs (a bit like a java dynamic compiler). So rather than running SETI-at-home your system would have built a specialist seti-ASIC on its main array. There will be lots of applications where most of the work is done in such a soft ASIC, and only a small proportion is done on a (commensuately puny) soft-CPU.
This all sounds too cool to be true, and at the moment it is. Existing programmable gate hardware is very expensive, of limited size (maybe enough to hold a 386?), runs crazy hot, and doesn't run nearly quickly enough.
## W.Finlay McWalter ## http://www.mcwalter.org ##
My question - If these cards are getting so powerful at computations then why do we need a Intel/AMD processor at all?
A development this extreme is unlikely. However, what is very real is the fact that GPUs and CPUs are at least partially competitors.
If you are doing a lot of graphics then you the best computer for your money may be with a great graphics card and a so-so CPU. The better and cheaper GPUs Nvidia can make, the smaller the demand for state of the art Pentium's.
But unless there is a revolutionary development somewhere, we will probably see computers with both kinds of processors for a good while.
Tor
Interestingly he thinks it'll be specialized hardware that will do ray-tracing, etc.
http://www.hardwarecentral.com/hardwarecentral/rev iews/1721/1/
"Is there a future for radiosity lighting in 3D hardware? Ray-tracing? When would it become available?
Gary: Yes, but probably just in specialized hardware as it's a very different problem. Ray-tracing is nasty because of it's non-locality, so fast localized hacks will probably prevail as long as people are clever. Especially for real-time rendering on low-cost hardware. It's interesting that RenderMan has managed to do amazing CGI without ray-tracing. That's an existence proof that a hack in the hand, is worth ray-tracing in the bush.
Oh... and for people who haven't seen it before, here's a cool detailed paper about how the pipeline of a traditional 3d accellerators can be tweaked used to do ray tracing...
http://graphics.stanford.edu/papers/rtongfx/rtongf x.pdf
Reading that shows how programming a graphics pipeline is quite different (more interesting? more complicated?) than programming a general purpose CPU.
Escher was the first MC and Giger invented the HR department.