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: Joe NYC who wrote (64983)	12/5/2001 2:29:30 PM
From: TGPTNDR	Read Replies (1) \| Respond to of 275872

Joe, Re, <But can 2 decoders be used in Kap's code?> You're never going to know, IMO, by looking at C code. If you get to the .asm you have a chance. if you get to the object code and analyze it you have a better chance. At the C level you're trying to look at the moon through the wrong end of the telescope.You can guess it's not a streetlight but that's about all, IMO. tgptndr

To: Joe NYC who wrote (64983)	12/5/2001 3:33:36 PM
From: combjelly	Respond to of 275872

"but about whether the trace cache runs at full speed or not." Decoding those x86 instructions is a very complex process. It is hard to take those instructions, align them, decode them and take all the special cases in consideration, without having a lot of stages. Since it seems that the decoder was lifted straight from the PIII, it would almost half to run at half speed, otherwise it would need to have more stages to the decoder pipeline. If the trace cache is big enough, this isn't a real penalty, unless there aren't any loops or the size of the loop exceeds the size of the trace cache.