SI
SI
discoversearch

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: Marco Polo who wrote (60385)6/3/1999 1:37:00 PM
From: Tenchusatsu  Read Replies (2) | Respond to of 1571204
 
<I see no mention of C400/100. Only C500/100, which is of course possible with the C333.>

Look here: www4.tomshardware.com and check out the fourth row from the bottom. Then check here: www4.tomshardware.com and look right underneath the AMD K6-3 450.

<You say the 200 MHz "bus" of the K7 is useless unless it uses a certain type of memory, but a 7200rpm hard disk would be even faster with the 200 MHz "bus", no?>

Not really. Large data block transfers is done with DMA these days. It's way too time consuming to have the processor transfer blocks of data directly to a hard drive controller (HDC). Why? Because the ultimate source of that data is SDRAM. So either the processor reads that data from SDRAM and then writes it to the HDC, or the HDC reads the data directly from SDRAM using DMA. Either way, the channel to SDRAM is the bottleneck.

So as of now, the 200 MHz K7 bus isn't necessary unless the memory system is composed of two parallel PC100 SDRAM channels or one RDRAM channel. It's not that hard to do the parallel SDRAM channels, and I don't see why initial K7 systems shouldn't feature it, even if it is more expensive. As for RDRAM, I guess AMD will just let Intel blaze the trail with Camino and Carmel, then follow along in early 2000 with K7 motherboards supporting RDRAM.

Tenchusatsu