Subject: Re: Rotational delay for IDE drives
To: None <amiga@NetBSD.ORG>
From: Michael L. Hitch <firstname.lastname@example.org>
Date: 04/25/1995 19:26:20
On Apr 25, 5:59pm, Bartholomew Niswonger wrote:
> It seems that in the discussion so far of disk performance, people
> have been sticking to SCSI drives. Since my drive is an IDE, I was
> curious if similar performance gains might be had by changing the
> rotation delay to 0, rather than the current 4, or if IDE drives were
> signifigantly different in the way they wrote to disk to make it not
> such a great idea. It seems to me it should make no difference
> whatsoever whether it is SCSI or IDE, but I was curious.
Since IDE drives have a read cache, I would think the same thing
applies to them as well. Writing 10MB to my IDE drive yielded a rate
of 300K with the rotational delay at 4ms, and 403K for a rotational
delay of 0ms.