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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Paul Engel who wrote (82167)	5/30/1999 8:41:00 PM
From: grok	Read Replies (2) \| Respond to of 186894

A possible good use for rambus would be as an L3 cache in a server. Imagine a slot card with Coppermine with a drdram port serving as an 8MB L3 cache. All the 800 MHz stuff is on the slot and off slot traffic is reduced due to the large on-slot L3.

To: Paul Engel who wrote (82167)	5/30/1999 8:46:00 PM
From: Scumbria	Read Replies (2) \| Respond to of 186894

Paul, However, my intuition is telling me that Intel will be moving the DRDRAM controller on to the CPU in the very near future - as indicated in the upcoming Timna. Then, your suggestion will work. This will achieve the reduced latency and deliver the performance "promised" by the DRDRAM approach. Having the CPU core, memory controller, and graphics on the same die has the advantage that the memory accesses can be managed more carefully between the CPU and graphics, to prevent contention. The raw latency will only improve by a very small amount, but the average latency can be improved by precharging/opening banks early. Scumbria