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.

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

To: combjelly who wrote (87489)	8/23/2002 3:35:18 PM
From: brushwud	Read Replies (2) \| Respond to of 275872

I haven't looked at the SPECint code, but it likely has 32 bit constants which get padded out to 64 bits, hence the extra space... There's no need to do that in the software, because 32-bit constants get padded out to 64 bits by the hardware. There are a fair number of bytes added to express the extended x86 instructions and such. The big downside of x86-64 is the average instruction size goes up. According to AMD's presentation at the Hot Chips conference, the average instruction size goes up by a fraction of a byte, and is still less than 4 bytes. Most 32-bit processors had instructions of a fixed length of 4 bytes. The x86 architecture was fairly unique in having variable-length instructions, and now x86-64 still has an average of less than 4 bytes. That seems pretty good to me and they had to put extra bits in somewhere to address the extra registers.