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: FJB who wrote (39446)	10/15/1998 10:41:00 PM
From: Tenchusatsu	Respond to of 1573682

Rob, I looked at the article you provided. I have never heard of VCM-DRAM until now. At first glance, it seems like VCM-DRAM holds the key to lower latencies, especially with Peter Gillingham saying that latencies can be reduced from 50 or 60 nsec to 15 nsec. However, that might be a deceiving statement in light of this quote from the article: <VCM, on the other hand, battles precharge latency by allowing memory masters to open and control virtual channels — separate interface blocks, each with its own SRAM page buffer, that work to filter out page misses.> The key word there is SRAM. In other words, VCM seems to follow in the footsteps of CSDRAM (cached SDRAM), which puts some SRAM cache in front of each chunk of DRAM. In effect, that 15 nsec latency is actually due to the SRAM cache chunks, not due to any new architecture in the actual DRAM memory array itself. I'm not sure why VCM DRAM or CSDRAM is necessary in systems which feature L1 and L2 caches. Tenchusatsu