Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!cs.utexas.edu!sun-barr!newstop!sun!imagen!qmsseq!pipkins
From: pipkins@qmsseq.imagen.com (Jeff Pipkins)
Newsgroups: comp.sys.ibm.pc
Subject: Re: Compuadd Disk Controller?
Message-ID: <59@qmsseq.imagen.com>
Date: 4 Dec 89 18:58:02 GMT
References: <201@sixhub.UUCP>
Reply-To: pipkins@qmsseq.UUCP (Jeff Pipkins)
Distribution: na
Organization: QMS Inc., Mobile, Alabama
Lines: 14

Like Mikey, I've tried it and I like it.

I don't believe in fairy tales like benchmarks, so my test was to
do a complete make (make -a all) on a commercial software product.

With the cache on the ESDI controller disabled, it took 3 minutes
(even without cache it's still a pretty fast ctrlr, w/1:1 interleave)

With the cache enabled, it took 1 minute & 10 secs!

So which is faster: 386/25 with ESDI cache, or 386/33 without?
I have a 386/33 WITH  @8-)

(Not related to CompuAdd, my opinions are mine, std disclaimers, etc.)