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 : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: wsp who wrote (21183)	6/1/1999 1:15:00 AM
From: Scumbria	Read Replies (1) \| Respond to of 93625

Scott, Could you describe the latency issue as mentioned in the article that MileHigh has submitted Memory latency refers to the average length of time between a data request, and the return of the requested data to the CPU core. This is the key factor in determining Winstone (and some other benchmarks) scores. Memory in a typical computer system consists of (from fastest to slowest) the level 1 cache, the level 2 cache, DRAM, and virtual memory on the hard disk. Dram latency is a very complicated subject. The article you referred to: aceshardware.com made the coarse assertion that DRDRAM latencies are higher than SDRAM latencies. This is true with the simplest possible memory controller, but the argument has been made that the large number of open banks in DRDRAM provides the opportunity for a sophisticated memory controller to reduce the average latency, and make up the gap. I don't want to delve into the subject in too much detail, but I will make the suggestion that the latency difference is probably not worth worrying about. The big issues with DRDRAM will be centered around memory bandwidth and cost. Scumbria