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: tejek who wrote (93869)	2/17/2000 2:45:00 PM
From: Scumbria	Respond to of 1571218

Ted, the Athlon has a 10 stage pipeline; how difficult would it be to extend its pipeline to improve clock speed and would that require redesigning the entire chip to accommodate the longer pipeline. It would require redesigning the whole chip, unless there is some major imbalance in the current pipeline (not likely.) Scumbria

To: tejek who wrote (93869)	2/17/2000 3:10:00 PM
From: kash johal	Read Replies (1) \| Respond to of 1571218

Ted, re:"Willamette questions" Will do my best: if the Willamette is to debut at 1.3 G; at maturity where would you expect it to be vis-a-vis clock speed. My best guess is at up 1.5-1.8Ghz on 0.18. But the chip should really hit its stride with 0.13 micron/cu. By end of next year 2-2.5Ghz would not surprise me. the Athlon has a 10 stage pipeline; how difficult would it be to extend its pipeline to improve clock speed and would that require redesigning the entire chip to accommodate the longer pipeline. Not a chance. It would take a complete redesign etc. **De Gelas seems to flip-flop on whether the T-bird will be a real competitor to the Willamette....what do you think? I think willy will beat thunderbird hands down - assuming thunderbird is simply athlon with on board cache. The new mustang core may be more of a match - but thats speculation as AMD is very quiet on Mustang. The extra pipelining and dual integer ALU's should allow willy to win all the business benchmarks convincingly. The overall FPU of AThlon seems superior. But the new SSE instructions seem to be MUCH more than just MMX rehashed. They will likely enable willy to win most of the gaming benchmarks if these get SSE2 enhanced due to enormous benefits. Seems to me Intel did a great job overall. AMD may well have to do a sledgehammer "lite" to one up willy. Such a chip would be dual Athlons on a die without the x64 stuff. A brute force chip would be in the 250mm2 range in 0.18 micron going down to 130-150mm2 range in 0.13 micron. Having said that willy is being lumbered with a very expensive plateform of dual rambus channels. So real machines with 256Mb of dual channel RDRAM are likely to cost in the >$2500 plus range. Fortunately they have shot themselves in the foot with the rambus approach. regards, Kash