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: Dan3 who wrote (67405)	8/2/1999 8:14:00 PM
From: Charles R	Read Replies (1) \| Respond to of 1574179

Looks like I was not the only one surprised by AMDs 64-bit comments from Drew. techweb.com The only alternative to Alpha that makes sense is extending x86 instruction set to 64 bits and that would be very interesting.

To: Dan3 who wrote (67405)	8/2/1999 10:12:00 PM
From: grok	Read Replies (1) \| Respond to of 1574179

RE: <Why would you not need the whole DRDRAM module component set for each channel? Controller, 128 sets of DRAM cells (that would be integrated into far fewer die) and connections to them. Without the whole set, how are you going to talk to the other end of each channel on tinma?> I'm really confused by your questions. Try this: Imagine two different Timnas, one with one channel (timna1) and one with two channels (timna2). Timna1 has four drdrams on its only channel and timna2 has two drdrams on each channel. Timna2 has enough internal smarts to send each memory transaction to the correct channel since there would be some interleaving scheme such as all even cache lines are on one channel and all odd cache lines are on the other channel (based on phyical addresses). Does that work for you?