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: Scumbria who wrote (120890)	7/25/2000 10:59:30 AM
From: Joe NYC	Respond to of 1572086

Scumbria, What would be the point of a single processor Sledgehammer? Sledgehammer is a new core, and it needs to be proven. It includes many enhancements including enhanced FP, 64bit extension to x86, maybe some new instructions for 3DNow. The chip will also likely have even better scalability and probably some enhancement to caches. The complexity of putting 2 cores on a single die (plus the Northbridge) will likely cause a slip in the schedule. But why not go for a coup and possibly release a stripped down version of Sledgehammer in 2000. While the work on dual core Hammer continues, the single core Hammer would create space and market acceptance for the new family. Also, a single core Sledgehammer will be a basis of future Athlons, so the more mass market versions could be built on the single core product. Joe

To: Scumbria who wrote (120890)	7/25/2000 11:30:26 AM
From: TimF	Respond to of 1572086

What would be the point of a single processor Sledgehammer? If it was faster then any other AMD chip and had a large fast L2 cache it might have some point. If a single processor version is ready 6 months before the dual processor version it might help get some support for 64 bit OSs or software. Tim

To: Scumbria who wrote (120890)	7/25/2000 2:30:12 PM
From: Gopher Broke	Read Replies (2) \| Respond to of 1572086

Scumbria, As I understand it the Athlon x86 core is implemented as microcode on top of a small fast risc core. (Apologies for my laymans terminology here) Presumably it was known when Dirk started on Athlon that AMD would need to produce a 64 bit system, and he was a 64 bit guy before. So I would be very surprised if he did not design the Athlon at the outset to be enabled for a painless migration to 64 bit, at least at the level of the risc core. Also, given that Sledge is an evolution of Athlon, I see no reason why AMD cannot produce interim silicon that is operational as far as external interfaces but does not necessarily have all 64 bit codepaths in place yet. Maybe even a processor that will only run 32 bit code but uses the Sledge bus architecture? This would allow for parallelism of CPU core development and system infrastructure would help in time to market, which has been AMD's real problem with Athlon and SocketA launches. After all 286 to 386 was simply a question of renaming AX to EAX so presumably Sledghammer just introduces EEAX :^) Seriously, the specification of the instruction set can probably leverage a lot of prior experience of the 16->32 bit migration which should help the design enormously. Sod it. We all know it is fantasy don't we :(