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: Tenchusatsu who wrote (80933)	11/24/1999 3:10:00 PM
From: Cirruslvr	Read Replies (1) \| Respond to of 1574096

Tench - RE: "Is it reasonable to assume that Athlon's on-die cache will just come straight out of the gate at 1 and 2 MB? I don't think AMD can start off with one or two megs of on-die cache. Most likely, Thunderbird will feature 512K of on-die cache. A respectable amount, for sure, but not enough to attack the markets that Cascades will be serving. That's why I don't think AMD can compete against Cascades until Mustang." I may have jumped the gun in assuming Thunderbird will be the derivative which will have 1 and 2MB on chip cache. I forgot about Johan's article at Ace's which says Thunderbird will have 512K on chip cache and that it will be Mustang which will have the MBs of on chip cache. So I guess you are right that it will require 1 and 2MB L2 Mustangs to compete with 1 and 2MB Cascades. But I still think 512K Thunderbird will be faster than 512K Cascades (assuming no core changes) based on what I wrote in my previous message. "My WAG is that they're enhancements for the mobile market." According to the roadmap I linked to previously, there will be a mobile version of Mustang as well. "I'm still wondering how AMD is going to find the capacity to build all these Athlons, especially those with on-die cache. Athlon derivatives are all good and swell for bragging rights, at least, but it's hard for me to imagine that these new cores are not going to hurt unit shipments any." Athlons made on the .18 process with up to 512K on chip L2 cache will still be smaller than the Athlon on the .25 process was. But I guess with on chip cache the chances of the processor being bad are greater, but Johan has a theory in that linked articl of how AMD might combat that. "And AMD said that Dresden will soon become the sole manufacturer of Athlons, which makes me wonder even more." Where did AMD say that? Speaking of Dresden, since AMD said they will be seeing revenue from Fab 30 in Q2 or the 2H of '00 I wouldn't be surprised if Dresden ramps up on Thunderbirds and Spitfires while Fab 25 continues to make current K75s. Does anyone know if there any way using copper can enhance yields on processors with on-chip cache? Or is it speed related only? BTW, Johan is supposed to have an article which will supposedly have info about Mustang's core enhancements and Willamette soon. Also, I wrote the message you replied to, not Chuck.