Re: memory reading and writing
From: Robert Redelmeier (redelm_at_ev1.net.invalid)
Date: 07/18/04
- Next message: Randall Hyde: "Re: Polling, Interrupts, DMA, Synchronous, Asynchronous I/O Definitions"
- Previous message: Robert Redelmeier: "Re: memory reading and writing"
- In reply to: Gyps: "Re: memory reading and writing"
- Next in thread: wolfgang kern: "Re: memory reading and writing"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sun, 18 Jul 2004 18:35:06 GMT
Gyps <gyps_rueppelli@yahoo.com> wrote:
> We all seem to throw out frequencies, but how does, say, 133Mhz
> actually translate into time for a single memory access??
Well, 1/133 MHz = 6.7 nanosecond per clock. After the first
access, subsequent accesses (64bits) in a burst can follow
this closely.
The first access is governed by "latency", some of which is in
the CPU, some in the Northbridge (RAM controller), and lot in
the DRAM chips themselves. Total 70-150ns.
-- Robert
- Next message: Randall Hyde: "Re: Polling, Interrupts, DMA, Synchronous, Asynchronous I/O Definitions"
- Previous message: Robert Redelmeier: "Re: memory reading and writing"
- In reply to: Gyps: "Re: memory reading and writing"
- Next in thread: wolfgang kern: "Re: memory reading and writing"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]