Path: utzoo!attcan!uunet!aplcen!uakari.primate.wisc.edu!sdd.hp.com!elroy.jpl.nasa.gov!usc!jarthur!nntp-server.caltech.edu!tybalt.caltech.edu!toddpw
From: toddpw@tybalt.caltech.edu (Todd P. Whitesel)
Newsgroups: comp.sys.apple2
Subject: Re: Optimizing
Message-ID: <1990Jun2.063316.9473@laguna.ccsf.caltech.edu>
Date: 2 Jun 90 06:33:16 GMT
References: <15800132@ux1.cso.uiuc.edu>
Sender: news@laguna.ccsf.caltech.edu
Organization: California Institute of Technology
Lines: 26

brownfld@ux1.cso.uiuc.edu writes:

>     On a related topic, does CPU speed effect the optimum interleave of
>your harddrive?  In other words, if you went from a IIe to a IIgs, would
>a new interleave be necessary?  This comes down to the question of "Do
>the slots speed up when the CPU does?"

The best answer you can get on that one is "it depends."

When using a DMA SCSI card (any kind, Apple's or not) CPU speed should make
virtually no difference because all the time critical stuff is handled by
the card. The driver itself makes more difference because it may or may not be
designed to handle 1:1 interleaves.

When the card is being passive (i.e. DMA not in use or unavailable) then the
CPU is doing the work and the speed makes a _big_ difference.

My preferred solution to the question is 'buy a card that either coprocesses or
has optimized drivers.' Apple's card, with (to my sense of aesthetics at least)
proper drivers, is the most cost-effective solution for the GS since the track
buffer and cache can be pulled from the system memory pool. Coprocessed cards
are better in the //e because they support 1:1 interleave drives out of the box
and Apple's card needs more buffer memory to really perform under everyday use.

Todd Whitesel
toddpw @ tybalt.caltech.edu