Nvidia Working on a CPU+GPU Combo 178
Max Romantschuk writes "Nvidia is apparently working on an x86 CPU with integrated graphics. The target market seems to be OEMs, but what other prospects could a solution like this have? Given recent development with projects like Folding@Home's GPU client you can't help but wonder about the possibilities of a CPU with an integrated GPU. Things like video encoding and decoding, audio processing and other applications could benefit a lot from a low latency CPU+GPU combo. What if you could put multiple chips like these in one machine? With AMD+ATI and Intel's own integrated graphics, will basic GPU functionality be integrated in all CPU's eventually? Will dedicated graphics cards become a niche product for enthusiasts and pros, like audio cards already largely have?" The article is from the Inquirer, so a dash of salt might make this more palatable.
Heard This One Before (Score:5, Interesting)
What I don't understand is that I thought GPUs were made to offload a lot of graphics computations from the CPU. So why are we merging them again? Isn't a GPU supposed to be an auxillary CPU only for graphics? I'm so confused.
What I'm not confused about is the sentence from the above article: Oh, I've worked with my fair share of DAAMIT engineers. They're the ones that go, "Yeah, it's pretty good but
Re:Heard This One Before (Score:5, Interesting)
a really, really fast pipe. It is a lot quicker to push stuff from CPU->GPU when they are on the same piece of silicon, versus the PCIe or AGP bus. Speed is what matters, it doesn't look like they are moving the load one way or another (although moving some load from CPU->GPU for vector based stuff would be cool if they had a general purpose toolkit, which I'd imagine one of these three companies will think about).
Re: (Score:3, Insightful)
Re:Heard This One Before (Score:4, Interesting)
Re: (Score:2)
That's the whole point here: put the CPU and GPU right next to each other and wire them together. You see, the nearer they are to each other, the less time it takes for electric impulses to travel from one to the other, and the faster the communication is.
And, of course, the reason number one: you get a guaranteed GPU sale for each CPU sale - goodbye pesky competitio
Re:Heard This One Before (Score:5, Insightful)
And vice versa. This might work where someone wants an embeded GPU for low to medium end graphics. However, I doubt gamers would like the idea of having to purchase a new CPU evertime a new GPU comes out and vice versa.
There's something to be said for physically discrete components.
Re:Heard This One Before (Score:4, Informative)
Re: (Score:2)
A cyclic process? (Score:5, Insightful)
Graphics chips seem to have done this cycle at least once; perhaps now we're just looking at the next stage in the cycle? We've had graphics as a separate component from the processor for a while, perhaps the next stage in the cycle is for them to combine together into a G/CPU, to take advantage of the design gains in general-purpose processors.
Then at some point down the road, the GPU (or more likely, various GPU-like functional units) might get separated back out onto their own silicon, as more application-specific processors become advantageous once again.
Re:A cyclic process? (Score:5, Informative)
http://www.cap-lore.com/Hardware/Wheel.html [cap-lore.com]
Re: (Score:2)
I won't dispute that the term in a technical usage was coined by them. But, it's basically a borrowed term from Hindu/Buddhist stuff who have believed in reincarnation and the wheel of life for a few thousand years.
Cheers
Re: (Score:2)
*laugh* That, or I should be less of a pedant.
Cheers
Re:A cyclic process? (Score:4, Informative)
Re: (Score:2)
Frankly, I hope this is the last stage in the cycle - at least until we have some radical changes in how pr
advantage of seperate companies (Score:2)
Re: (Score:2, Funny)
Re:Heard This One Before (Score:5, Interesting)
But I highly doubt that nVidia will be able to get a CPU out that out-performs an Intel or AMD, which the high-performance junkies would want. Intel and AMD put a HUGE amount of money into research, development, and fabrication to attain their performance. This is going to be interesting to watch. Hopefully nVidia doesn't dig themselves into a hole with this attempt.
Re: (Score:3, Interesting)
Maybe they don't have to. If they can just make something that can accelerate MMX/3D Now (sort of a graphics pre-processor) and plug that into a Socket F [theregister.co.uk] slot, it'd be like a two-stage accelerator: first accelerate the calculations that produce the graphics, then accelerate the display. Maybe they could find a way to do a micro-op translation of MMX instructions into something more RISC-like, and run them on a RI
Re:Heard This One Before (Score:5, Interesting)
Memory size and bandwidth are the usual limitations. Remember that if you want 2x AA, you double your memory usage, and if you want 4x AA, you quadruple it. So, that game that needed 128 megs on the video card, with 4x AA, can suddenly need 512.
steve
Re:Heard This One Before (Score:4, Interesting)
Of course with the GPU integrated into the CPU you wouldn't need card-based RAM at all. You'd process your video on system RAM, and it would be as fast as the GPU accessing its own RAM at the moment is (not shit like shared-memory video cards are at the moment). This results in flexibility: if you're only using 128MB of RAM for your graphics, you can reuse the other 384MB as additional system RAM.
Not so much (Score:5, Informative)
Even fast as it is, it's still slower than the GPU would really like.
What you've suggested is already done by low end accelerators like the Intel GMA 950. Works ok, but as I said, slow.
Unless you are willing to start dropping serious amounts of cash on system RAM, we'll be needing to stick with dedicated video RAM here for some time.
Re: (Score:2, Troll)
Huh? All the systems on this lineup [tomshardware.co.uk] use standard PC3200 (DDR400) RAM. Which is the same RAM that you could use as system RAM with many motherboards (e.g. this one [micom.co.uk]). I don't see why the RAM would be faster on the video card than in the main system...?
Also, a GPU inside the CPU would get to benefit from the CPU's cache, which would usually contain any data that had recently been modified by the main redraw thread, thus eliminating the need to go out to get data from t
What's that got to do with anything? (Score:5, Informative)
Not a small difference, really. My system RAM is rated to somewhere around 10GB/second max bandwidth (it gets like 6 in actuality). The graphics card? 54GB/sec.
Video cards have fast RAM subsystems. They use fast, expensive chips and they have controllers designed for blazing fast (and exclusive) access. You can't just throw normal, slow, system RAM at it and expect it to perform the same.
Re: (Score:2)
Of course, Nvidia could plan to use blazingly fast RAM like that used on video cards now as the system RAM on motherboards supporting their CPU-GPU hybrid, which would solve the problem nicely, though it might drive the price up quite a bit. (Then again, it would improve system pe
Re: (Score:2)
You are correct that those systems are using DDR RAM. But graphics cards (including the cards in those machines) use other, more expensive, faster RAM, like GDDR3.
Actually... (Score:2)
Re: (Score:3, Informative)
I think the first example of this integration we see will use the HyperTransport bus and a single package with CPU and GPU on different dies, though fabbed on the same process. This could be done with an existing AMD socket and motherboard.
Before this happ
Re:Heard This One Before (Score:4, Insightful)
At one time floating point was done by software it still is one some cpus.
Then floating point co-processors became available. For some applications you really needed to speed up floating point so it was worth shelling out the big bucks for a chip to speed it up. This is very similar to what we have now with graphics cards.
Finally CPUs had floating point units put right on the die. Later DSP like instructions where added to CPUs.
We are getting to the point where 3d graphics are mandatory. Tying it closer to the CPU is now a logical choice.
Re:Heard This One Before (Score:5, Informative)
Re:Heard This One Before (Score:5, Interesting)
I was using floating point as an example.
I don't know if Nvidia can pull this off without a partner. Too build a really good X86 core isn't easy. I wonder if they may not do a PPC or Arm core instead. That could make nVidia a big player in the cell phone and mobile video market. At some point there will be portable HD-DVD players.
My crystal ball says.
AMD will produce these products.
1. A low end CPU it integrated GPU for the Vista market. This will be a nice inexpensive option for home and corporate users. It might also end up in some set-top boxes. This will the next generation Geode.
2 A family of medium and high end video products that use Hyperchannel to interface with Opteron and Athlon64 line.
Intel will
Adopt Hyperchannel or reinvent it. Once we hit four cores Intel will hit a front bus wall.
Intel will produce a replacement for the Celeron that is Duo2Core with integrated graphics on one die. This is to compete with AMD new integrated solution.
Intel will not go in to the high end graphics line.
nVidia will fail to produce an X86+GPU to compete with AMD and Intel.
nVidia produces an integrated ARM+GPU and dominates the embedded market. Soon every cellphone and media player has an nVidia chipset at it's heart. ARM and nVidia merge.
Of course I am just making all this up but so what, electrons are cheap.
Re: (Score:2)
Your CPU isn't going to work well at the 200-400 MHz of a GPU, and you're not going to make a huge GPU die run at 2 GHz to get your CPU to work well. I think that their CPUs will be closer to Via's C3 than a P4/Athlon 64.
Re: (Score:2)
Re: (Score:2)
Re:Heard This One Before (Score:4, Informative)
Re:Heard This One Before (Score:4, Informative)
I believe the last option (option 7) is what x86/87 CPU/FPU combo actually used. That's why there is a coprocessor-prefix in front of the FPU instructions. They are not just unused opcodes.
Option 5 (and sometimes even 3) is commonly used for MMX/3dNOW/SSE/SSE2/SSE3/whatever instructions.
Unless they *really* need nonportable features, most programmers tend to go with option 2.
Re:Heard This One Before (Score:4, Interesting)
Then people started using floats for the convenience, not because the accuracy was needed, and performance suffered greatly as a result. Granted, there are a lot of situations where accuracy is needed in 3D, but many of the calculations that are done could be better done in integer math and table lookups.
Does it often matter whether a pixel has position (542,396) or (542.0518434,395.97862456)?
Using a lookup table of twice the resolution (or two tables where there's non-square pixels) will give you enough precision for pixel-perfect placement, and can quite often speed up things remarkably. Alas, this, and many other techniques have been mostly forgotten, and it's easier to leave it to the MMU or graphics card, even if you compute the same unneccessary calculations and conversions a million times.
Fast MMUs, CPU extensions and 3D graphics routines are good, but I'm not too sure they're always used correctly. Does a new game that's fifty times as graphically advanced as a game from six years ago really need a thousand times the processing power, or is it just easier to throw hardware at a problem?
Re:Heard This One Before (Score:5, Informative)
Yes we live in an upside down world where floats are faster than ints some times.
Re: (Score:3, Informative)
Yes. It absolutely matters. It makes a huge difference in image quality.
It matters when we go to sample textures, it matters when we enable AA, it matters.
Re: (Score:3, Informative)
No, it doesn't. Note that I said pixel, not coordinate.
The coordinates should be as accurate as possible, but having a pixel more accurate than twice the resolution of the display serves very little purpose.
Re: (Score:3, Informative)
You'd be mistaken [virginia.edu]. See the slide on Texture Mapping.
Perspective divide is performed before texture sampling. This is necessary to get proper texture step sizes, for correct sampling of the texture onto the pixel.
Fractional pixel locations are also used in antialiasing.
Some things you forget (Score:4, Informative)
2) For many things, it DOES make a difference. You might ask why do we need more than 24-bit (or 32-bit if you consider the alpha channel) integer colour? After all, it's enough to look highly realistic. Yes well that's fine for a final image, but you don't want to do the computation like that. Why? Rounding errors. You find that with iterative things like shaders doing them integer adds up to nasty errors which equals nasty colours and jaggies. There's a reason why pro software does it as 128-bit FP (32-bits per colour channel) and why cards are now going that way as well.
3) In modern games, everything is handled in the GPU anyhow. The CPU sends over the the data and the GPU does all the transform, lighting, texturing and rasterizing. The CPU really is responsible for very little. With vertex shaders the GPU even handles a good deal of the animation these days. The reason is that not only is it more functional but it's waaaaay faster. You can spend all the time you like trying to make a nice optimised integer T&L path in the CPU, the GPU will blow it away. You actually find that some older games run slower than new ones because they rely on the CPU to do the initial rendering phases like T&L before handing it off, whereas newer games let the GPU handle it and thus run faster even though having higher detail.
Re: (Score:2)
Per definition, no.
But does it matter whether the programmer has to create and populate lookup-tables, with lots of manual tweaking to find the right compromize between memory usage (cache issues), few branches (lookup must be fast), accuracy (lookup must be correct), etc... when the alternative is to simply call sin(x)?
Re: (Score:3, Interesting)
Obviously this is not going to be ideal for high end gaming rigs; but it will improve the quality of integrated video chipsets on lower end and even mid range PCs.
Re: (Score:3, Informative)
Do y
Re: (Score:2)
Re: (Score:2)
GPU's are so powerful now, that some of the latest high-end scientific visualisation applications will actually do calculations on a supercomputer, transfer the data across to a PC, and then use the CPU to process the data so it can be visualised on the GPU in real-time. Similarly for
Re: (Score:3, Informative)
You're partially right. GPUs were made to execute the algorithms developed for graphically-intensive programs directly in silicon... thus avoiding the need to run compiled code within an operating system, which entails LOTS of overhead. Being able to do this directly on dedicated h
Re: (Score:2)
Now that processors have multiple cores, many of which are left looking for a job to do - it makes sense to bring the GPU back to the main die.
The result will produce an immediate performance boost for Joe Sixpack, at lower manufacturer cost.
Re: (Score:3, Interesting)
What I don't understand is that I thought GPUs were made to offload a lot of graphics computations from the CPU. So why are we merging them again? Isn't a GPU supposed to be an auxillary CPU only for graphics? I'm so confused.
You've already gotten some good answers here, but I'll throw in something that I haven't seen anyone else mention explicitly: GPUs aren't only being used for 3D animation anymore. GPUs started because, in order to make graphics for games better, you needed a specialized processor to
Re: (Score:2, Insightful)
As other have replied its all about the bus speed. The amount of time it takes to move data from chip to chip can insert enormous overhead.
Think back a little to the DEC Alpha. Now the ALPHA chip in and of itself was not really that remarkable. What was so VERY remarkable about the Alpha system was its bus switching. It was blazingly fast and could handle monster amounts of data from main memory to CPU, GPU, etc. The reason ( mostly ) that its now sitting in HP's vault is that the bus switch was/is re
It's a cycle (Score:2)
Then, there came minicomputers. Fewer components meant smaller machine with smaller prices. Banks, small colleges, and even some high schools got into the game. My high school had an AS/400.
Then came the micros. These are what we call PCs now. They started as hobbyist toys. Then, the chips got more powerful and more memory was put in them, and they started
nVidia don't have a good chance with this. (Score:3, Interesting)
I think the better option would be to have a graphics chip fit into a Socket 939 on a dual socket motherboard, with an AMD chip. It will have a high-speed link through hyper-transport, and would act just like a co-processor. I'm no chip designer, so I have no idea what the pros/cons of this are, or if it's even possible.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
One Question: (Score:2)
I can't imagine it is that much
(since they mostly suck)
With integration.. (Score:3, Interesting)
I'm thinking way too much. It did alleviate boredom for about a minute though...
Out of their league? (Score:5, Insightful)
Nvidia is a fantastic graphics card company - they should continue to innovate focus on what they're good at rather than try to play follow the leader in an arena they know nothing about.
Re:Out of their league? (Score:5, Insightful)
Re: (Score:2)
Especially if they just shove several of them together onto the die. Everyone is going to be focusing more on software that can take advantage of multiple CPUs for the next couple of years, and nVidia can ride the coattails of that with a nice, simple in-order-execution design. Put 16 or so of those onto your chip with a good GPU, and you might get pretty good perform
No not so much (Score:2)
Even after you've offloaded all the graphics, there's still a ton to be done. AI, physics, and sound to name just three. Fire up a copy of Oblivion some time and you'll be amazed how hard it hits the CPU. It even makes use of dual cores, if they are available. Yes, a prime factor in game speed is having a GPU that
Re: (Score:2)
Yes but those (Score:2)
The problem is the GP seems to think that nVidia would have a viable market combining their excellent graphics chips with shitty processors. Nope, not so much. If you want a good graphics card, you pretty much by definition also want a good processor. There are actually plenty of people who can use a good processor and have no need for a good video card
I smell a pattern (Score:3, Interesting)
We had separated math co-processors, that later were integrated in the CPU.
Then the separated GPU, which will soon be integrated back too.
Re: (Score:2)
When you look at the 80387, it was a lot harder to get enough onto a single chip, and the bus speed between chips was a lot faster, relative to the CPU speed, than it is now (you might have a couple cycles of latency, but
Math co-processors, anyone? (Score:5, Insightful)
patents (Score:2, Insightful)
I would very much doubt that they could compete with AMD and Intel who have already patented many x86 cpu concepts.
It's a shame that Intel has decided not to buy nvidia, and go it alone with it's own design staff.
Thank MicroSoft (Score:5, Interesting)
No.
The major driving force right now in GPU development and purchase are games.
The major factor that they have to contend with is DirectX.
As of DirectXv10. A card either IS, or IS NOT compliant. None of this "We are 67.3% compliant".
This provides a known target that can be reached. I wouldn't be surprised if the DirectX10 (video) featureset becomes synonymous with 'VGA Graphics' given enough time.
Yeah, sure, MS will come out with DX11, and those CPUs won't be compatible, but so what?, If you upgrade your CPU and GPU regularly anyway to maintain the 'killer rig', why not just upgrade them together?
Re: (Score:2)
Don't know about you, but for me graphics and CPU still aren't in sync. I imagine for most gamers it's still that way. More often than not you want to swap your graphics card, and your CPU *only* if it's holding back the GPU, which is far from always. I'd say you can at least live two generations of GPU cards on the same CPU. Still, for the "all-in-one" market that wouldn't go off buying a se
Re: (Score:2)
While this is a deliberate feature of DirectX, it's nowhere near as useful as you suggest. What is specified and complied with is the set of instructions which the card will accept. What is not specified is what it will do with those instructions - a card is considered to be DirectX compliant even if it has many rendering errors in the output. GPU makers can and do take a great many liberties with this, and that's
Re: (Score:2)
Re: (Score:2)
My comment about DirectX10 was just that it was being presented as an absolute standard to adhere to, unlike its predecessors, if you want to get certified, you have to have certain features. This provides a much more concrete standard that people can look at and say 'can that graphic Card/Chip do X?'.
I'm confused how this would make the CPU beholden to MS. Somehow the CPUs current
Re: (Score:2)
Not until M$ looses its grip it holds with every game developer as hostage to cheap development tools M$DevStudio for M$Widn0ze and Xbox.
As of now, they are obediently do what M$ tells them. Earlier Xbox release? - Pay them promotional fees and consider it done. Poor/late MacOS/X support? - No problems. Non-existent Linux support? - Give them another rebate for M$DevStudio and they would forget abou
Pointless without documentation. (Score:4, Insightful)
This isn't a good thing unless they also release documentation for it!
Just my preference . . . (Score:4, Funny)
I prefer my articles with a dash of accuracy.
Niche market? (Score:4, Insightful)
Haven't they already???
The vast majority of machines (at least, from my experience, which could not be broad enough) seem to be shipping with integrated graphics on the motherboard. Certainly, my last 3 computers have had this.
Granted, I buy on the trailing edge since I don't need gamer performance, but I kind of thought most consumer machines were no longer using a separate graphics card.
Anyone have any meaningful statistics as opposed to my purely anecdotal observations?
Re: (Score:3, Insightful)
I really doubt the CPU part is going to compete with the latest super-quadcore chips from AMD or Intel, so no-one will us
Re: (Score:2)
What about the latest quad-core chips are mainstream??? Those are specialty products if there ever was a specialty product. Except for high-end gamers and people doing really specialized tasks, who actually needs one of them? I bet I couldn't tax one for anything I do.
Nowadays, so many motherboards have video, lan, sound, IDE controller, possibly RAID, an
It's a logical extension of the NVidia NForce line (Score:5, Interesting)
I've been expecting this for a while, ever since the transistor count of the GPU passed that of the CPU. Actually, I thought it would happen sooner. It's certainly time. Putting more transistors into a single CPU doesn't help any more, which is why we now have "multicore" machines. So it's time to put more of the computer into a single part.
NVidia already makes the nForce line, the "everything but the CPU" part, with graphics, Ethernet, disk interface, etc. If they stick a CPU in there, they have a whole computer.
Chip designers can license x86 implementations; they don't have to be redesigned from scratch. This isn't going to be a tough job for NVidia.
What we're headed for is the one-chip "value PC", the one that sits on every corporate desk. That's where the best price/performance is.
Re: (Score:2)
Something that's interesting about this, if t
Re: (Score:2)
Re: (Score:2)
Re: (Score:3, Interesting)
If using larger chips means I can get 2GB combined RAM for the price of 1GB system RAM and 256MB video RAM? Absolutely.
Re: (Score:2)
Re: (Score:2)
Jen-Hsun Huang: A True Asskicker (Score:3, Informative)
It's actually a very good article for those interested in nVidia's history and Huang's mentality. Paul Otellini [intel.com] ought to be afraid. Very afraid.
Niche (Score:2)
It already is.
Other uses? (Score:2)
Duh. Gaming consoles. Add memory, a sound controller, and some sort of storage, and you're in business.
Here's the bottom line. (Score:2)
Every once in a while an unrelated tech innovation comes around that benefits (A) or (B) in some indirect fashion. It could be faster bus speeds, more sophisticated GPU instruction sets, etc etc etc. Doesn't really matter what they are, but they happen all the time and each m
Nothing New In The World (Score:2)
L2 cache used to be external. Then they integrated it when technology and performance allowed. L3 cache then became external while L2 was integrated; now you can buy processors with all this inside. Put the memory controller inside the CPU and you no longer need to spread out high (er than CPU core voltage) IO lines with nasty length requirements between Northbridge and CPU, and can clock the bus faster. Put the ethernet
Despite doubters, this seems a good idea to me (Score:2)
What if you could put multiple chips like these in (Score:2)
They'd probably be obsolete in three months, as opposed to one month
Nvidia is the odd-man out (Score:2)
It's pretty clear that both Intel and AMD are intent on swallowing up the lower 3/4 (hand-waving guess) of the GPU market over the next few years. And I believe that ATI will still be fighting it out at the top end over that remaining 25%.
That would leave Nvidea as a niche player in the uber high end, making GPUs strictly for graphics professionals and gamers with t
How about a Xeon-socket GPU? (Score:2)
Re:Should Slashdot really insult other news outlet (Score:5, Insightful)
What NVidia eventually does may not bear much resemblance to the story.
Re:Why multiprocessor units suddenly most efficien (Score:3, Informative)
multiple processors (CPU, GPU or otherwise) are a way to add more 'cycles' based on current technology. This has the advantage of being able to get more out of your cur
Re: (Score:2)
Traditionaly, when you move to a more compact production process, your parts are closer together, so it takes less time for the electric signals to move through them (propogate), so you can get faster clock speeds without really changing the design much. When Intel reached the 90nm process (or maybe the one before -- 130nm?), they were startled to discover that that effect just didn't work anymore. The c
Re:Why multiprocessor units suddenly most efficien (Score:2)
I don't know the 3dfx history all that well, but I'd *guess* that their cards were getting hard to dig more performan
Re: (Score:2)
Re: (Score:2)
No, it wouldn't. It would have to be either cheaper for OEM's to consider it, or offer some other selling point that AMD or Intel wasn't offering, or, preferably, both; which is how AMD (and Cyrix and