SI
SI
discoversearch

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.
Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?


To: Daniel Schuh who wrote (2077)7/24/2000 4:52:50 PM
From: ScumbriaRead Replies (1) | Respond to of 275872
 
Dan,

It wouldn't surprise me a bit if Intel was doing a Sledgehammer type product, though I have no specific information to confirm that.

Scumbria



To: Daniel Schuh who wrote (2077)7/24/2000 5:42:05 PM
From: pgerassiRead Replies (1) | Respond to of 275872
 
Dear Daniel:

The whole point of Sledgehammer is that it does not require any changes to run current software fast. Recompiling opens up a further performance increase. This is more like the P2 to P3 or K6 to K6/2 transistion rather than P3 to IA-64. You do not have to use the 64 bit extensions for you to get native speed. But if you want maximum speed, you must recompile or get x86-64 software. IMHO, gcc will be able to get 80-90% of the maximum speed within a year (probably less since it supports the flat FPU of RISCs like Alpha).

The closest historical parallel would be the 286 to 386 transistion. All software ran. Unix variants ran using 32 bit segments within a few months. Microsoft took many more years to get to full 32 bit compliance (some say it is not, even now). This time, IMHO, Linux will run with it when shipped, and Microsoft (or what ever it is by that time) will run 64 bit extensions within a year after it starts shipping.

LDT's performance gains do not need software changes to show up. Optimizing for the LDT bus might gain at most 10% except for highly specialized apps.

Pete