Path: utzoo!attcan!uunet!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.sys.amiga
Subject: Re: 2090 HDdisk.device
Message-ID: <4176@cbmvax.UUCP>
Date: 2 Jul 88 22:46:18 GMT
References: <8806280524.AA12822@jade.berkeley.edu> <2320@antique.UUCP> <4158@cbmvax.UUCP> <2323@antique.UUCP>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 44

In article <2323@antique.UUCP> vax135!cjp (Charles Poirier) writes:
>In article <4158@cbmvax.UUCP> jesup@cbmvax.UUCP (Randell Jesup) writes:
>>With no real fast mem ($c00000 mem is as slow as chip),
>
>Actually I do have 2meg of fast mem, but I had not been using FastMemFirst
>so as to allow my VD0: to recover.  Having just replaced VD0: with VDK:,
>and then turning on FastMemFirst, voila, my hard disk works fine!

	Glad it worked.  That's the reason FastMemFirst is the first thing in
all the startup-sequences.

>>	The last thing that is happening is that we're not talking just a
>>Read() request, we're probably talking an EA IFF reader, which isn't blazing
>>itself.
>
>We-l-ll, I don't see how you would get bus contention on the file
>access due to the same process that is making the file request.  It's
>either waiting for the next Read() to complete, OR it's processing the
>last one.  Though in the case of Deluxe Paint, known to indulge in busy
>waiting in at least some circumstances, they may have found a way.

	I just meant that even after you get a sector from the HD, the CPU
has to decode it.  With the code, data, and destination all in slow memory,
combined with the fact that the EA IFF reader isn't really CPU efficient,
that's extra overhead.

>>>Commodore!  Yoo-hoo, Commodore!  What do I do NOW?  Help!

>>	Please don't yell.

>Sorry, I feel better now.
>Everything
>I could think of about my setup was identical to that of people whose
>SCSI drives worked fine.  I did forget about FastMemFirst though.

	It's often the simplest things that are hardest to find.  If in doubt,
go to a true plain vanilla WB to test.

>Thank you Randell, DaveH, Andy, and all others who helped me through this.

	Glad I could help.  Pass the word about FastMemFirst/hd34.4.

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