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.
Technology Stocks : CYRIX / NSM
NSM 18.270.0%Jul 31 5:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Scumbria who wrote (27549)6/19/1998 5:05:00 PM
From: milan0  Read Replies (1) of 33344
 
Scumbria

You had a discussion with Steve Porter arguing that the MXi, due to memory latency, will be slower than with an L2 cache.

My favorite issue to beat to death is the fact that CPU performance is limited by memory latency, rather than bandwidth. A 128 bit bus does nothing for latency.


However, Cyrix says the contrary. First, here is Swearingen in this recent article:
techweb.com

The original Cyrix press release said same:
cyrix.com

The processor also integrates the latest system technology, including SDRAM control, AGP and consumer-quality DVD playback, onto the processor. With its low memory latency and industry-leading memory bandwidth - greater than 2GBytes per second - the MXi is expected to significantly outperform competing non-integrated AGP solutions.


Cyrix also has a pdf file explaining the MXi. Since I am not a technical type, I can't say if these memory features could obviate the natural latency of RAM.

ftp://ftp.cyrix.com/developr/mpf97db.pdf

Multi-ported memory control interface with 13 cache lines of write post queuing and tractable prioritized arbitration
256 entry 4-way set associative GLB
128-bit SDRAM interface
Greater than 2GB/sec memory bandwidth
Supports double-edge data rates
Up to 16 open banks and up to 32KB pages
Supports up to 8 pipelined requests


Scumbria (or anyone knowledgeable), do these features balance memory latency in any way? Could the MXi be as fast as with an L2 cache with these memory features?

Mike
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext