Path: utzoo!attcan!uunet!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.sys.next
Subject: Re: Next and the competition
Message-ID: <5590@cbmvax.UUCP>
Date: 29 Dec 88 01:42:40 GMT
References: <2405@ncsuvx.ncsu.edu> <5725@polya.Stanford.EDU> <17911@glacier.STANFORD.EDU> <315@belltec.UUCP> <2596@udccvax1.acs.udel.EDU>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 23

In article <2596@udccvax1.acs.udel.EDU> anand@vax1.acs.udel.EDU (Anand Iyengar) writes:
>	I have a report at home which benchmarks a '386 and a 68020
>(similar clock speeds) and finds performance close (each outperformed the
>other in different parts of the test, but not by much (I think less than 20%,
>back and forth)).  I recall reading a report from Motorola which says
>that the 68030 can perform 50% better than than the 68020 (2 cycles/mem-access
>vs. 3 for the 68020).  

	An '030 should run (give or take) around twice as fast as an '020,
especially if you're using the MMU.  The data cache helps as well.

>	I've worked on 16 and 20 MHz 386 machines for about a year (Compaq's).
>Both were quite fast until you went to the disk (granted, we only (<-?)
>had 2.5M of memory, but many times I/O is unavoidable).  If the next
>machine has widened this bottleneck, then it may do well.  

	PClones have pretty dismal I/O rates.  On a 68000 Amiga, 7.xx MHz,
with a SCSI controller, you can get 650K-800K _bytes_ per second through the
file system on large reads, 1.1 MB/s using the driver directly, off good
disks like the Quantum 80S or the CDC Wren V.

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup