Nvidia As x86 Manufacturer
There is usually some truth behind every rumor. For years, there were rumors of Apple’s Marklar division, a team dedicated to porting OS X from PowerPC to x86. Though this seemed crazy during the Intel Pentium 4 era, this was a rumor that never died. Turns out, it wasn’t a rumor.
Along those same lines, I don’t believe it is simply an unsubstantiated rumor that Nvidia is making an x86 CPU. This has come up too often, and the company's recruiting of x86 validation engineers and licensing of "other Transmeta technologies" besides LongRun hint at a bigger picture.
Without having any inside information, there are two areas where the x86 investment could prove worthwhile. First, I’ll tell you what it’s not. It’s not a high-end CPU. Although AMD and Intel are both working on integrating CPUs and GPUs on the same die, neither company will be able to integrate flagship performance graphics on the same die. This is because the increase in die size from incorporating a high-end GPU and CPU will result in exponentially greater manufacturing costs, and the challenge of thermal management for such a large chip will be another engineering challenge (Ed.: just look at the time Nvidia is already having with GF100, and that's a GPU-only).
The first option is that Nvidia is continuing development of the Transmeta Crusoe CPU. Though the Crusoe was not a commercial success, on a performance per watt level, the CPU was highly competitive against even today’s Intel Atom. A newer version of the VLIW architecture of the Crusoe, augmented by improvements in manufacturing technology and the code-morphing algorithms could be a competitive low-power device. When combined with an embedded GPU, Nvidia would have a product that competes against AMD Fusion and Intel embedded products. This could be a desktop version of Tegra.
The second option, which is more likely, is that Nvidia will incorporate a simple CPU on future versions of the Tesla or Quadro. Currently, one of the most computationally-inefficient portions of GPGPU is transferring data back and forth between the graphics card and the rest of the system. By incorporating a true general purpose CPU on the graphics card itself, "housekeeping tasks" can be performed on the GPU with local graphics memory, thereby improving performance. It could be an intermediate to better manage asynchronous data transfers from the GPU to this mini CPU. This device would not need to run x86; it could apply code morphing to work with Nvidia PTX instructions or have some efficient combination that makes it worthwhile.
Hardware REYES Acceleration?
Remember all that talk about Pixar-class graphics? Pixar’s films are rendered using Renderman, a software implementation of the REYES architecture. In traditional 3D graphics, large triangles are sorted, drawn, shaded, lit, and then textured. REYES divides curves into micropolygons that are smaller than a pixel in size, along with stochastic sampling to prevent aliasing. It’s a different way of rendering. At SIGGRAPH 2009, a GPU implementation of a REYES render was demonstrated using a GeForce GTX 280. Though more work will need to be done, Nvidia appears to be headed in this direction with Bill Dally in the position of VP of research. I’d be surprised if we didn’t see an Nvidia implementation of REYES in the future.
In fact, Nvidia already has an investment in Hollywood. Late last year, it announced iRay, hardware-accelerated ray tracing for use with the mental ray suite. Mental ray is a global illumination/ray tracing engine that competes against Renderman/REYES, and has been used by feature films such as Spiderman 3, Speed Racer, and The Day After Tomorrow. Oh, and Mental Images is a wholly owned subsidiary of Nvidia.
Nvidia’s Outlook
Nvidia’s corporate philosophy and track record is consistent with the goal of providing hardware-accelerated graphics to consumers, hardware-accelerated rendering to Hollywood, and throughput computing to the scientific community. The hardware and software expertise required to produce this is available within Nvidia’s walls. Whereas AMD has the track record with CPU and GPU hardware, and Intel has the deepest pockets, Nvidia has built the strongest portfolio of software technology. Software is what made the iPod. Software is what made the iPhone. Nvidia’s vision is coherent, but the company’s success requires timely execution of both its hardware and software milestones (Ed.: notable, then, that this is currently an issue for the company).
Conclusion
The next few years will be an exciting time for computing. We have a bona fide three-horse race with AMD, Intel, and Nvidia. Perhaps more important, each company has non-overlapping talents and a unique approach toward success. The next generation of products will not simply be "me too" launches, but instead reflect a world of new ideas and paradigms. These technologies will enable new areas of entertainment, science, and creativity. And games will look pretty sweet, too. At least, that’s the way I see it.
- 1 / 2
- Next
-
Latest Graphics Cards News
- 20/05 – Qualcomm Hires Former AMD CTO Demers
- 18/05 – Nvidia Responds to GeForce 600 Series V-Sync Stuttering Issue
- 18/05 – Behold: Here Are The First Unreal Engine 4 Screens
- 17/05 – Nvidia Debuts GK110-based 7.1 Billion Transistor Super GPU
- 17/05 – GE Announces Kepler Graphics Card for Military and Aviation
Latest Graphics Cards reviews
- 18/05 – Best Gaming CPUs For The Money: May 2012
- 10/05 – Best Graphics Cards For The Money: May 2012
- 10/05 – GeForce GTX 670 2 GB Review: Is It Already Time To Forget GTX...
- 09/05 – Video Teaser: Radeon HD 7800 Series Remains A Solid Value
- 08/05 – Benchmarking AMD's 768-Shader Pitcairn: Not For Public...
Couldn't agree more on upgrading speakers before sound-card. I use a Denon amplifier and a pair of Mission floorstanders with my PC and all of my friends tell me that they didn't realise a PC could sound so good. And this is with a P35 motherboard so it's a good few years old. I will admit though, that I am now looking at getting an Asus Xonar Essence...
Fantastic artical!
Really insightful, and with really interesting lloks back at the past. All we need now is someone to proof-read articles before they go live!
"Given that Pixar films still require 5 to 6 hours to render a single frame on large supercomputer clusters..."
If movie is all about effects, Transformers 1-2, Avatar...
2 hour movie = 7200 seconds = 518400 frames (72p frame rate)= ~3 million hours of render time with 6 hours per frame = about 350 years to render a 2 hour movie. Correct me if I'm wrong, but I think the 5-6 hours per frame is little too much isn't it?
Fantastic artical!Really insightful, and with really interesting lloks back at the past. All we need now is someone to proof-read articles before they go live!
+1 rep for having spelling mistakes in a comment having a go at someone for making spelling mistakes
+1 rep for having spelling mistakes in a comment having a go at someone for making spelling mistakes
haha! yeah, I noticed that as soon as I posted it but couldn't be bothered to rectify it. And in my defence, I'm not doing this for a living!
How many things can you get wrong in 1 article?! The iPod generation might be happy listening to their horribly compressed MP3s with crappy onboard sound - the rest of us will stick with dedicated sound cards and proper speakers. If the author thinks that an el cheapo chip on a motherboard that costs less than £40 is anywhere near as good as a £80 X-Fi card then he has a screw loose. Never mind the price/performance ratio - it isn't even in the same ballpark. To Asus & Creative: Please don't take this article as a template of what we all think - its not!!!
As for the Fermi architecture, it is dead in the water. It is like the author has not read any tech news from the last few months and is still going off the marketing cr*p from Nvidia produced 1 year ago.
How many things can you get wrong in 1 article?! The iPod generation might be happy listening to their horribly compressed MP3s with crappy onboard sound - the rest of us will stick with dedicated sound cards and proper speakers. If the author thinks that an el cheapo chip on a motherboard that costs less than £40 is anywhere near as good as a £80 X-Fi card then he has a screw loose. Never mind the price/performance ratio - it isn't even in the same ballpark. To Asus & Creative: Please don't take this article as a template of what we all think - its not!!!
As for the Fermi architecture, it is dead in the water. It is like the author has not read any tech news from the last few months and is still going off the marketing cr*p from Nvidia produced 1 year ago.
Whilst I disagree totally with the authors opinion on the quality of onboard sound (my NAD amp / Mission speakers come alive with my DJ friends M-Audio card compared to my onboard), the rest of the article is an interesting opinion. I do hope Fermi comes good - if only to provide competition, I'm not toally convinced by the 'ATI / AMD drivers are as stable as nVidias' arguments put here - my 4870 was awful and I ended up flogging it in a week just to go back to properly written drivers.
You don't get it david__t. The difference between a onboard sound chip and external are first of all not discernable for a large portion of the population, secondly; most of the general population do not care about the tiny relative difference between a 192MP3 to FLAC enoguh to spen the extra £200+ on teh equipment for it.
Graphics will head to a point where the differences between engines and generations are so small many people will end up not caring. The relative difference between 3000 to 500,000 polygons compared to 5million to 7million is so large eventaully we will head towards many billions of polygons with tesselation where the difference at the high levels is so small it isn't worth it.
Note the word relative used in this comment. The differnce between first generation tapes/cd players and computer systems and today's MP3/AAC players and onbaord sound chips is so large over a similar period of time it takes for us to move towards lossless and exact replication eventually we don't bother.
"Given that Pixar films still require 5 to 6 hours to render a single frame on large supercomputer clusters..."If movie is all about effects, Transformers 1-2, Avatar...2 hour movie = 7200 seconds = 518400 frames (72p frame rate)= ~3 million hours of render time with 6 hours per frame = about 350 years to render a 2 hour movie. Correct me if I'm wrong, but I think the 5-6 hours per frame is little too much isn't it?
I thought that to until I realised that that only 1 super computer what if they had say 40, that would be around 8 years, still a very long time but I think that's how long it takes for Pixar movies. Correct me if I'm wrong
Electron tunneling is already making things a considerable challenge. Gate oxides are already in the sub 10 nm range, which is what gives CMOS it's idle leakage current. What people usually aren't told is that it's not enough to just scale one thing between process nodes; gate length and gate metal pitch are just two of many things that need to be scaled down for each process jump. A transistor with a 5 nm channel wont work if it's gate oxide is 5 nm also. It's very difficult to make a continuos film that is only a few atoms thick.
Some roadblocks have already been reached, leading to the rise of such things as parallel computing rather than just continuing with higher GHz. I really don't see conventional CMOS transistors getting to 5 nm. We'll probably see gradual advancements in other directions which require software to adapt, etc.
I'm not a specialist in these matters but I think that the console market is having a big impact on the rate of progress in graphics technology nowadays. Usually when a console comes out a pc is almost immediately putting out better graphics and by maybe 3 years down the line pc graphics completely blow the consoles out of the water (for instance the PIII, 32mb tnt2 graphics system we got at my house in 1999 could play Quake 3 a lot better than PS2 could a year later, and by 2005 when that console generation was coming to a close Xbox was getting titles like HL2 and Doom 3 but was clearly WAY behind in terms of graphics)
In the current generation though, with rampant piracy on PCs and the continued success of consoles it is more profitable for developers to make sure their games are playable on 360 or PS3. If you consider that the current benchmark for PC graphics is still Crysis from 2007, it's clear that the PC has a lot of capability but developers would rather hold back to the spec of current consoles (for instance MW2's graphics are nothing like crysis' but it has sold much better) I even have a feeling I read that crytek have vowed never to make a pc exclusive again, and that the sequel to "it's so good it could never come out on consoles Crysis" will be on both PS3 and 360.
In this light I feel the pc graphics industry is being held back by the popularity of consoles and sony and ms's refusal to upgrade after huge investments in the current generation that is slowing down the growth in pc graphics as much as anything else. From what I've read, DX11 features should allow the pc to start pushing ahead again in graphics while still allowing the games to be console quality quite easily, but this remains to be seen and is just further proof of the supervenience of the pc graphics market on that of consoles.
Oops just signed up, first comment I made and somehow it came out 3 times. Anyone know how to delete it?
Oops just signed up, first comment I made and somehow it came out 3 times. Anyone know how to delete it?
Haha!
Welcome to Tom's hardware, the most sketchy forum/comment system in all the land!
Don't think you can delete or edit comments. Don't worry about it!
ok, you can but you have to do it through the forum rather than the comments section.
for all you guys saying this article is crap coz u run your own dedicated sound card...... there would be less than 1 in 1000 PCs out there with a dedicated sound card...
The numbers dont care about how 'passionate' you are about your soundcard.... This is a huge drop from the 486 days of requiring a sound card to play sounds....
use your brain.
wow, what an extensive article! You must've done a lot of research to get the numbers and facts to prove you points so hats off to that!!!
for all you guys saying this article is crap coz u run your own dedicated sound card...... there would be less than 1 in 1000 PCs out there with a dedicated sound card...The numbers dont care about how 'passionate' you are about your soundcard.... This is a huge drop from the 486 days of requiring a sound card to play sounds.... use your brain.
I wasn't saying the article was crap - merely the analogy
There's still a huge difference between onboard and dedicated sound cards, it's just fewer people care about it. Personally I find it very odd that someone will spend £500 on a great gfx card / monitor and £30 on speakers, but each to their own!
"There's still a huge difference between onboard and dedicated sound cards."
Please elaborate - how a SPDIF on an onboard card is worse than the SPDIF of a dedicated card? If you go thru DAC then yes, dedicated cards win. But nowdays you can just use the digital pass-thru (with an optical cable) and still have excellent sound quality...