Slashdot Mirror
More Drooling Over The Opteron
bradv writes "I havent heard much about the new 64bit chips from AMD lately and was excited to find this article to satisfy my appetite for a little while longer. Probably more info than most people will ever care about. "
SPEC results linked from The Inquirer ... here
Well, Mr Cringely thinks the Opteron will be a huge win for AMD and makes quite a nice argument about it (based on hearsay though): link
/.ers...)
(top sci/tech link from memigo currently; yes it's the holidays but a few things are happening
Nice try. Microsoft has already publically announced 64-bit Windows support for x86-64.
Relevant quote:
AMD's newly named Opteron server processor will get its own 64-bit version of Windows, and the 64-bit desktop Athlons will not be forgotten either
Linux is ready as well.
Now, if we can just get MacOS X.... =)
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
We are evaluating a dual Itanium2 machine right now. It's screaming fast, but with the price point I doubt we'll buy any.
Electron Migration? what are you talking about. Processors continue to get faster and faster due to improved processing technology and increased parallelism. Leakage and electromagnetic interference from the clock signal are major problems today but who knows what scientists are working with nowadays.
> By doubling the word length to 64-bits, you can reduce the clock rate of the chip, and will still be able to perform more instructions per second than your top-of-the-range Athlon/Pentiums.
That is absolutely not true. Having 64 bits allows you to access a larger amount of total memory, and it lets you put more information in each instruction. The amount of data you can work on in any given clock cycle is proportional to the cache access and bandwidth and the register size (Neither of which inherently need 64 bit long instructions).
To perform more instructions per second (or instructions per clock cycle) you need instruction level parallelism (ILP). This has been a major goal of processor manufacturers for many years now. Intel had two main ways of trying to increase ILP.
You may be interested to read about the HyperTransport capabilities of the chip at http://www.hypertransport.org
One thing I found particularly interesting was the SMP abilities of the AMD, through the use of Hypertransport. It allows multiple chips to be used on the same board without all the glue logic normally associated with SMP setups, so you can have arrangements like the Power4 and suchlike, without enormous amounts of additional circuitry.
Funky stuff
Say it with me: There is no such thing as electron migration. There is, however, something called electromigration and it has nothing to do with clock rate. The problem is that as electrons flow in a conductor, they collide with lattice ions and push these ions around a little bit. This isn't a problem in the macroscopic world since wires are so big, but in a microscopic (or nanoscopic) scale this can lead to melting and diffusion of the conductor into the surrounding medium. The copper atoms slowly diffuse into the silicon around them, almost like a gas (a very slow moving gas).
Since these motions are caused entirely by the force of electrons colliding with the atoms, they are completely determined by the kinetic energy of the electrons -- i.e., how fast they move. And that in turn depends on the mean-free-path length (a property of the conductor) and the electric field within the conductor. It has absolutely nothing to do with clock rates.
Newer, high-speed chips may suffer more from electromigration than slower chips, but this is only because the new chips have much thinner wires and are therefore subjected to a greater current density at a given voltage. I.e., more electrons flow per unit area, so the number of electron-atom collisions goes up.
Well, in Q1 2002, desktop processors accounted for 76% of AMD's revenue (according to their S&P stock report). Thats a little more than just a "cog in the AMD wheel" if you ask me. Their flash memory products only accounted for about 17% of their revenue during that same time period.
The fact is AMD really isn't that big of a company. I think they have around 13,000 employees (compare with ~80,000 for Intel), and their revenues have been dropping like crazy ($500 million in Q3 of this year, compared to ~$6 billion for Intel). I don't think either AMD or Intel could really survive if they lost thier PC processor revenue. If you want to see a company that could survive losing a major product, look at Motorola or IBM. Now those are truly HUGE companies.
"The defense of freedom requires the advance of freedom" - George W Bush