Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.sys.amiga
Subject: Re: Commodore gets serious?  You judge.
Message-ID: <14338@cbmvax.commodore.com>
Date: 10 Sep 90 23:00:10 GMT
References: <29164@nigel.ee.udel.edu> <6891@vax1.acs.udel.EDU> <14251@cbmvax.commodore.com> <1990Sep7.021855.2789@uncecs.edu>
Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 60

In article <1990Sep7.021855.2789@uncecs.edu> urjlew@uncecs.edu (Rostyk Lewyckyj) writes:

>Dave would you please then tell all of us why the A1000 has such
>great difficulties with its bus and expansions? 

Mainly because folks are hooking up stuff in a way never intended.  The A1000
should have ONE, count 'em, ONE add on device hooked up.  That's either a
SOTS box or an Expansion box.  Anything that goes inside, other than perhaps
a daughterboard replacement ROM kind of thing, is going to be bad news.
Anything that involves chaining together multiple SOTS boxes is also asking
for trouble.

>Why the problems aand fixes to terminate the bus and ground some of the 
>chips?

Well, I have heard that the original daughterboard was noisier than one
would have liked it.  I hacked around with it way back (in '86, in conjunction
with this Chris Erving memory hack thing) and didn't notice it being 
particularly noisy, but perhaps it was.  The quoted fix was running an extra 
ground wire to one of the PAL devices.  I guess the basic problem here, if 
any, was that the daughterboard was a very late development and could have
been done better.  In fact, it was in the very last A1000s, which 
incorporated that section into the motherboard.

On the other hand, a number of the problems this is supposed to fix are related
to multiple SOTS boxes and other doohickeys working together.  You can hardly
blame the A1000 for not working in situations it's not especially designed to
be working in.  So, while some of these things are looked on as fixes by the
folks hooking up too much stuff, I would prefer to call them hacks to get
that stuff, sometimes, working.

>What about the fact that the bus is (not buffered?)

Well, with only one SOTS box hooked up, that's not a problem.  The loading on
the A1000 edge connector, like the A500 edge connector or A2000/A3000 
Coprocessor slots, is expected.  Within limits.  The A1000 expects exactly
one set of buffers to be hooked up to it, and is well within its right to
get annoyed at multiple things hanging there.

>Is this because you were not involved then?

No.  I was working on the C128 then.  You can't hook cartridges up to the 
C128 and expect it to work, either.  And there's no standard for an external
C128 expansion backplane.  I'm pretty happy about most of what they did back
in the A1000 days, especially on the expansion bus, which I did pretty much
inherit responsibility for.

>I also seem to remember there were some problems on the 2000
>(with memory?)

Memory problems?  There were problems with the very first 2000-A memory
daughterboards, but I think they licked those problems before production.

>  Reply-To:  Rostyslaw Jarema Lewyckyj


-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
      Get that coffee outta my face, put a Margarita in its place!