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: Petz who wrote (48274)	7/19/2001 5:03:03 PM
From: fyodor_	Read Replies (1) \| Respond to of 275872

Petz: But keep in mind, a 2 GHz Athlon with DDR doesn't have twice the performance of a 1 GHz Athlon with PC133 Well… that depends on the benchmark. There are certainly benchmarks where that is the case, although, yes, for the majority of applications, scaling isn't going to be nearly as good. However, when these guys use words "performance is going to be X times product XYZ", it's usually a safe bet that that'll be pretty much on the nose… in a few select benchmarks. Take what many here believe about Hammer (btw, the slide doesn't specify if it's Clawhammer or Sledgehammer): - longer pipeline - SSE2 - larger (and/or faster) L2 cache¹ - integrated 128bit DDR memory controller Just using these 4, I can easily imagine a 50% IPC increase in some cherry picked benchmark. Remember, the Hammer will have prefetching as well, which the 1GHz Athlon did not. -fyo ¹ The slide only says "Perfoamnce normalized to 1GHz Athlon" (sic!), so the reference could be to either Athlon Classic or Athlon Thunderbird. And the (in)famous pink dot is not centered at 3, but slightly above. I should mention that the original AMD slide (used in the AMD Analyst Conference on November 9 of last year), which is the one I am referring to, does not have the "dot" placed nearly as high as the c't slide referenced elsewhere on the thread (http://www.heise.de/ct/01/10/016/bild.jpg). The latter has the dot centered at something like 3.4, whereas the original was much closer to 3.1.