Subject: Re: Dual SCSI, single chain?
To: NetBSD/alpha Discussion List <>
From: Greg A. Woods <>
List: port-alpha
Date: 03/30/1999 03:25:05
[ On Monday, March 29, 1999 at 21:31:17 (-0500), Curt Sampson wrote: ]
> Subject: Re: Dual SCSI, single chain?
> On Mon, 29 Mar 1999, Greg A. Woods wrote:
> > Look Curt, you said at least one thing wrong, and gave a whole lot of
> > other rather misleading and incomplete advice.
> No, you did.

Excuse me?

> I'll admit I flubbed on the negotiation between two
> wide devices separated by a narrow cable. But I have personally
> tried pretty much everything else I described and seen it work.
> To say you `can't' do any of that stuff is Just Wrong. If you've
> got some crappy parts combining some fo this stuff may not work,
> but then again, if you've got crappy parts even avoiding these
> combinations does not mean anything will work.

Perhaps you think you're talking only to experts with first-rate
components to work with, but you're not.

Just because it works for you, in your very specific circumstances, has
*NOTHING* to do with whether or not things should be done that way or
not.  If you want to play the cards then that's your perogative, but if
you want be sure things will work then you play by the specs or take
your chances.

> > I've got at least one Digtial Expansion Chassis
> > which will not co-exist on the average bus with any FAST-20 devices.
> I don't see how that proves that you can never combine a slow device
> and fast device, or whatever.

It's a simple matter of economics.  Many "older" devices do not even
approach the minimum specifications required to co-exist with modern
devices -- lengthy stubs, poor cables, poor connectors, etc., which were
perfectly fine and within their design specifications, are not suitable
for use with faster devices.  Period.

The real point is that if people quit giving so-called "advice" that
suggested that anything that works is "OK" with SCSI then the general
impression people have that SCSI is difficult and finicky would be far
less of a problem.

							Greg A. Woods

+1 416 218-0098      VE3TCP      <>      <robohack!woods>
Planix, Inc. <>; Secrets of the Weird <>