Is the CPU dead or alive? Nvidia says a little of both
Nvidia's hostility toward Intel is on a high boil these days. In its latest dig against the central processing unit (CPU) and the company that makes the lion's share of CPUs, an Nvidia VP said in a private missive that the CPU is dead and it has "run out of steam."
Nvidia CEO Jen-Hsun Huang
(Credit: Rico Shen)But wait. That's not what Nvidia really thinks. The message cited by the The Inquirer is "not a public statement," said Brian Burke, an Nvidia spokesperson. "The views in (Roy Tayler's) e-mail do not mirror the views of Nvidia." (The author of the message, Roy Tayler, is VP of content relations at Nvidia.)
But is the statement that far apart from Nvidia's public sentiment? "You need nothing beyond the most basic CPU," Burke said. Sounds like Nvidia thinks the CPU is, if not terminal, certainly fading.
(The CPU, or central processing unit, is the main processor in a PC. The GPU, or graphics processing unit, handles much of the visual content on a PC.)
This of course is news to Intel, the largest chip company in the world whose main business is making CPUs. "We believe that both a great CPU and great graphics are important in a PC. Any PC purchase--including the capability level of components inside it--is a decision that each user must make based on what they will be doing with that PC," said Intel spokesperson Dan Snyder.
To be sure, Nvidia and Intel have never gotten along famously. But the acrimony (mostly Nvidia's) started to build at Nvidia's fourth quarter conference call and carried over to the company's financial analyst day earlier this month, when CEO and co-founder Jen-Hsun Huang, alluding to comments from game developer Tim Sweeny, said "Intel is incapable of running modern games. Intel's integrated graphics just don't work."
But the crux of Nvidia's marketing message, vis-a-vis Intel, is focused on the graphics chip maker's perceived limitations of the CPU. In short, buy a high-end GPU, not a high-end CPU, and save money. During the earnings conference call, Huang cited the Gateway P series notebook as an example. One model has an Intel 1.6 GHz processor and a GeForce 8800 GPU. He said systems like this with a "higher-end GPU" and "lower-end CPU" are better optimized for today's users. "Relative to a notebook with a higher-end CPU and lower-end GPU, the Gateway FX is twice the performance and yet $200 lower cost."
Brooke Crothers has served as an editor at large at CNET News, an editor at Dow Jones' Asian Wall Street Journal Weekly, and a senior editor at InfoWorld. His CNET blog covers chip technology and computer systems, and how they define the computing experience. He also contributes to The New York Times' Bits and Technology sections. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. Follow Brooke on Twitter @mbrookec. 
Even for hardcore gamers, the difference from having a core2duo or a core2extreme doesn't mean much in term of performance as of having a 8600GT or a 9800GX2, so I guess Nvidia isn't entirely wrong.
Let's not forget. "Nobody will EVER need more than 640KB of ram" - Bill Gates.
Moore's law is Moore's law, and neither it, or the nature of technology, will ever submit to such limitations.
my blog: www.tech-connections.blogspot.com
On a side note, I would be very surprised if a number of people at the NSA are living in terror at the thought of algorithms designed to use a GPU to factor very large prime numbers.
Large prime number factorisation is executed in an exponential number of operation depending on the number to factorize. Adding more core doesn't really change things as even the world most powerful computer would take several centuries to factor a 4096 or even a 2048 bit RSA key...
The only thing they could fear would be quantum computing, if it can exist in a stable form. (No, D-Wave does not count)
Unless someone came up with an algorithm that would be easy to implement on a GPU (Lets just say that this non existent algorithm might even work better on a Cell). The only problem with such an algorithm would be that something like that would probably be very memory intensive.
As long as you still need a regular CPU then there isn't any point in saying you don't need a fast CPU. Yeah I can buy a cheap CPU to save money, but I buy a fast CPU because I want a fast CPU. Not because I need one. Right now if you buy a cheap computer with a cheap CPU, you're probably going to get a cheap GPU too. Last I checked out a cheap computer it didn't come with SLI boards.
We already know the x86 architecture is not the best architecture. We don't use it because it is the best. Windows isn't the best either. They're just the most popular.
The problem is as long as the computer needs that part it has the potential to become a bottleneck. Slow CPU will slow you down. Slow GPU will slow you down. Slow ram will slow you down. A fast system means a fast everything. I've sense gotten rid of them, but I had two systems with 128 meg. One with a 500 mhz CPU and one with an 800 mhz CPU. The 800 mhz system had an extremely slow hard drive that was fragmented and actually going bad.
The 500 mhz system had a newer faster hard drive. The 500 mhz would run circles around the 800 mhz because it didn't matter the 800 had a faster CPU. The drive was broke so you couldn't get any data to the CPU to begin with.
Now if they can build a machine that can do non-graphical tasks as fast as graphical ones regardless of the CPU connected to it then awesome, but it's probably going to take more than just building the machine. You wouldn't see the real difference until you modified the compilers to take advantage of the GPU then recompiled all the software. So, while they might be right we're still going to need a good CPU for now.
- by Mam00th April 29, 2008 5:50 AM PDT
- @ralfthedog, I still do not see why a SIMD processor would be albe to factorize in a non exponential time. Still, the NSA don't use RSA, they use a proprietary algorithm design by their cryptographers and I higly doubt it's based on the product of two large prime.
- Like this Reply to this comment
-
(7 Comments)