Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: RDM who wrote (47888)	1/30/1999 1:07:00 AM
From: Yougang Xiao	Respond to of 1571043

RDM: <<Sorry for such a long opinion.>> I, and believe many others on this thread, enjoy greatly of reading it! Thanks again for the great piece!

To: RDM who wrote (47888)	1/30/1999 1:42:00 AM
From: Scumbria	Read Replies (3) \| Respond to of 1571043

RDM, I agree that Alpha ownership would be an extremely valuable investment for AMD. Merced is going nowhere real fast. The chip is huge and will not be manufacturable in quantity. The performance will probably be a great disappointment as well. I have a few technical comments on your comments: 2. Will they would produce an X86 enhanced Alpha (more speed running X86 apps)? No way. This has been studied extensively by many companies and nobody has come up with a worthwhile idea yet. Messing with the architecture would probably slow down the clock speed. The current Alphas run X86 apps by emulation and are not going to be super fast at running x86 apps True and not true. FX32 is primarily a translator, rather than an interpreter. It actually converts the memory image of the program into an Alpha executable. Only the non-translatable parts are emulated. As a result, x86 apps generally run >50% as fast as native Alpha apps. On servers, x86 code compatibility is largely a don't care anyway. The vast majority of the server market is Sparc machines. Scumbria

To: RDM who wrote (47888)	1/30/1999 2:31:00 AM
From: Tenchusatsu	Read Replies (2) \| Respond to of 1571043

Thanks for your extensive thoughts on Alpha and the server market. Here are some comments of my own ... <Personally I wonder about the prospects for the Intel Merced and successors. The Intel 64 bit chips may not be the fastest X86 processors when they arrive.> Actually, the Merced will most probably be the fastest x86 processors out there, mainly because Willamette is being released about six months after Merced. It is unknown whether Foster (derivative of Willamette for servers) will be faster at x86 apps than Merced, but I wouldn't be surprised if it was. <One million chips per year may be a lot for this market. Even at $2,000 per chip the early 64 bit CPU market may not be a big growth market for either AMD or Intel.> With all this talk over the impact of Merced and that blasted six-month delay (you can tell I'm not happy over the Merced delay), the fact is that Merced is going to such a low volume market compared to Pentium or K6/K7. <Probably AMD wants the K8 to extend the addressing of the K7. Perhaps the Alpha people and patents may help with this.> Actually, I don't think extended addressing is that hard to achieve. It was done for Xeon without too much trouble (I think), and it can't be too hard to come out with a K7-derivative that has extended addressing. <I my opinion the Alpha is an awesome chip and a real contender for being one of the most powerful processors on the planet. Does it make sense for a mass production chip maker to produce it? That depends upon lots of things, including the price.> We'll see. Alpha's biggest threat is Merced and McKinley, but of course, it isn't here yet. Tenchusatsu