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. "
Doesn't perform..and I mean really perform...I'm not sure if AMD will be with us much longer, which would be a shame.
I think this is accurate because of the architecutural choice AMD made--instead of going with an all-new architecture, ala Itanium, they instead blew out the x86 system to 64 bits. That level of division in the CPU market at this time feels like it will have a very significant effect on the balance of power.
No one besides servers can use 64 bit chips right now
This is exactly what makes the Opteron an attractive processor. Rather then being simply a 64 bit proc like the Itanium, it has the ability to run legacy 32-bit instructions. This is a Good Thing. Now I can have a 64-bit proc that can still run all my old apps, but still can take advantage of the benefits of 64-bit architechure.
In fact, as I see it, the only people that won't benefit from 64-bit are Windows users. Until MS makes a 64-bit version of Windows the standard, the only people that will benefit from the Opteron will be the people that run OSes that they compiled themselves.
I didn't think that was a very good article. There seemed to be a lot of guesses in there, none of which appeared to be particularly informed - or at least, they were not explained - and some of it sounded downright childish. Like:
:-)
I don't know what Reserved might mean. One of the reviewers says that maybe in this case the processor turns into DSP. It's a mad idea, but if AMD realized it, this processor would be second to none in some kinds of operations.
or
AMD realizes it, and at present they develop several independent versions of the compiler together with famous software development companies. I won't unveil their names - AMD will do it if necessary. You just should know that at launch the processor will have the required support of the compiler allowing using its architectural advantages.
sorry?
No, i'd rather read C'T, at least they already have one of them chips on the test bench
Well, x86 continues to live because of the absolutely enormous amount of software written for it. Could you even begin to fathom starting at zero again? Itanium failed for this very reason. Sure, one of the holy grails as programmers we are supposed to chase after is full portability, but technical and time constraints often make that a dream. It is unfortunate, but the only way I can forsee a momentous move to a completely new architecture is if Microsoft really monopolizes the entire software industry, and then gets in bed w/ CPU makers and agrees to port all of the apps to this new architecture, and then everyone will make a ton of money selling new hardware and software that is legacy free. I will place my bet that in 10-15 years we will still be having this discussion.
errrr wrong.
More and more people are doing the home-movie dance. Trust me, 4GB of memory will be the deafult on any home PC within 5 years. By then we'd be stuck if not for 64bit computing.
I have 3 compuyers at home : 1 PowerMac with 1GB of memory, one PC with 2GB of memory (parsing 3 or more crosslinked SGML files > 512MB is a pain with less than 2GB) and my tiny game machine with 'only' 512MB. Just by examining the curve of purchased machines, I'll hit the need for 4GB within 2 years. And I'm not even doing video !
When will I end this grieving ? When will my future begin ?
If AMD can deliver this on a desktop level, then Moore's Law can once again be considered applicable...
Think about it - the main problem in terms of pushing computing power these days is electron migration, caused by extremely high clock rates.
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.
This was always the case with graphics cards; the GeForce 256 was a big step up from the Riva chipset, due to doubling the word length.
Supercomputers, such as the SGI Origin series, have been using 64-bit processing for quite some time now (MIPS processors), and while the Itanium series has its flaws (like a lack of backward compatibility), surely it's time to move on from the same old x86 architecture?
We don't all have to wait for Microsoft to make their WinXP 64-bit version mainstream; there's no point in them pushing this until the 64-bit architecture breaks into the home market.
Because the Opteron has this backward compatibility, then the 64-bit architecture will reach the home users, and they can upgrade to the 64-bit version as soon as it is deemed economically viable by Microsoft to release it.
I wonder what kind of performance increase you'd get from a program such as SETI@home or Distributed.net by upgrading to a 64-bit platform...
I married Miss Right. I just didn't know her first name was 'Always.'
The article has some fairly interesting material, but what really amused me was how the reviewer didn't really seem to understand most of what he was writing about. He seems to have alternated between copying stuff directly from some marketing glossy and what he could get from a comp arch textbook...ususally following up with something like "I'm sure this is good for something or someone somewhere, beats me though!". You can almost see him scratching his head. It all starts when he is confused by the 'Resevered' entry in a table of register settings.
Itanium has lots of cool new features that compilers could be using and people could be taking advantage of
Yes, and Intel would really appreciate it if someone would develop a compiler that takes full advantage of the Itanium. Really. Please. Because their own compiler is still struggling with the problems inherent in VLIW... yes, it's much, much, much better than it was a couple years ago, but it's still nowhere near where it needs to be.
And it costs a fortune. But, hey, if you can afford a $9000 chip, you should be able to afford the compiler too.
I like and respect Intel... I've grown beyond the newbie EE stance of "it sucks because it does", and recognize that they have some of the best minds working there, and that their fab processes are second to none. But Itanium has been a massive disaster for them, and they're now caught between a rock and a hard place. They can continue developing future revisions of IA64 and hope that someday their engineers figure out how to make it work well, work cheaply, and work fast with legacy code, or they can commit corporate hari kari and adopt x86-64 from AMD. Or they could do something similar, but different, to x86-64 on their own and just piss off everyone. Bad choices all around.
The only chance Itanium has is if AMD flubs the Athlon64/Opteron launch. AMD will probably pull out of the market shortly after and Intel can gradually increase profit margins to the point where throwing cash at a losing proposition (IA64) remains viable. And eventually force everyone to transition, like it or not (which, admittedly, would probably be a good thing in the long run, but the short term would suck).
Of course the 80386 (they didn't use the 'i' crap back then) was revolutionary. You know that Finn, what's his name, the one with the thing for penguins? He wrote that whole Unix-like operating system because he finally got his hands on an affordable 32-bit CPU for personal use. It's not Intel's fault that the real benefits of the 32-bit design were unavailable with the most common operating system of the time (MS-DOS and Windows 3.0).
Seriously, compare the 80386 to its predecessor, the 80286, and tell me it wasn't revolutionary. Now look at the Pentium (or PII or PIII or P4). A faster 80386, with built-in 80387 and cache memory, and some spiffy additional 'multimedia' instructions. Yes, I'm oversimplifying, but all the improvements I'm leaving out are evolutionary in nature. There's very little code that will run on a Pentium4 that won't on a 80386 - other than the aforementioned MMX/SSE stuff.
Hey kids, there's only 5 days left 'til Yak Shaving Day!