[net.micro] 32xxx bux cycles and CPU speed

jackson@jaws.DEC (I'm all right jack, keep your hands off my stack) (03/25/85)

> The VAX memory access speed relies on write back cache -   ...


The VAX 11/780, 750, and 730/725 all have write through cache.  The VAX 8600
is the first VAX processor that has write back cache.  (for reasons of 
complexity and also Multi-processing capability (although the 780, 750
and 730 weren't really multiprocessor)

Doug Pardee writes:

> my experience is that a 10 MHZ 32016 w/MMU and FPU is in the 
>same ballpark as (or slightly faster than a VAX 11/750.

This may be true in single or even small user mode.  The VAX 11/750 
is a SYSTEM product which can support many users.  I really doubt that any
micro system can really handle the kind of load that the 11/750 can 
(this has been discussed before about the 68K and the 11/780)
More things make a system than a CPU.  Sure the CPU can go fast on
a cycle by cycle basis.  How about Disk I/O?  How about memory controllers?





The 68K family and 32K family are impressive chips.  They can go FAST if 
used in the right environment.   They can also go SLOW in some applications


Bill Jackson
Digital Equipment Corp.

ARPA:	Jackson%JAWS.DEC@decwrl.arpa
USENET:	{allegra|decvax|ihnp4|ucbvax}!decwrl!dec-rhea!dec-jaws!jackson
E-net:	JAWS::JACKSON
USmail:	77 Reed Rd; Hudson MA 01749; MS HLO2-1/N10
Phone:	(617)568-4083