Subject: Re: Oooooh... IDE to Q-bus/Unibus adapters.
To: None <port-vax@netBSD.org>
From: Allison J Parent <allisonp@world.std.com>
List: port-vax
Date: 07/07/1999 19:11:17
Please keep in mind I'm used to dealing with physically SMALL PDP-11 and 
VAX systems not the 11/70, 780 or 8550 class monsters.

<You certainly do need disconnection if you want to run more than one drive
<per bus.  On the other hand if you only want a few drives, it's not that bi

Disconnection is not a inherant function of IDE.  The interface can provide 
that however.

<a deal; all IDE controllers that can handle UltraDMA provide two buses per
<chip anyway.

Asked and answered.

<That's true for *one block*.

IDE protocal will transfer many blocks sequentially.

<When's the last time you saw a preponderance of single-block I/O requests
<on a Unix machine?  Command queueing lets the drive -- which, with a moder
<drive, is the only thing that really can -- accept N requests, sort them

And the current PIO scsi protocal is...?

<And if you have multiple drives on the bus, that's even more true, since
<with IDE when you ask one drive for a block, you *can't ask the other driv
<anything at all until the first drive has finished* -- essentially halving
<your available disk bandwidth.

Having muliple busses is very cheap with ide..

<If you only want to use a small number of drives, I think IDE is a good
<choice.  It keeps cost and complexity down.  On the other hand, as a
<general-purpose solution it's vastly inferior to SCSI -- you shouldn't

Never said IDE was better overall for SCSI only pointed out that for build 
it yourself and be able to write a driver for it it's got SCSI beat more 
often than not.  It's biggest advantage is for those whose income can't 
swing those costly used SCSI boards for qbus or Unibus or for those of us 
that find were swimming in small (under 1gb!) drives that are useless for 
W9x, NtX, or the current crop of Linux (with Xand apps).  I must have a 
dozen of those 120, 420, 525mb drives that are looking for good use
other than a swapping drive in a PC.  Even new IDE is 1/2 to 2/3rds the
cost of similar SCSI offerings.

<think that all SCSI host adapters are as braindamaged as a 5380 with no
<DMA, certainly not if you're comparing them to brand-spanking-new PCI
<IDE interface chips.  A more reasonable comparison would be the a 
<53c[7,8]00 series SCSI host adapter.

Sure, but do the VS2000, 3100/m10/m30/m38 and M76 have them? NO.  Qbus 
vaxen have very nice SCSI controllers that even speak MSCP for a not so 
nominal fee.  Many of the systems that have SCSI do use the brain damaged 
5380 though.  In that context IDE stacks well, compared to the latest
PCI offering, well... when did VAX get PCI?

SCSI is a good robust bus, well designed and often underutilized.  But 
could you take a DRV-11 drawing and make an interface to a SCSI disk... 
only with the greatest difficulty. Why? all the nice scsi chips are plcc
or other nasty many legged flat packs (or very bus incompatable).
IDE that would be a fairly small adaptor, as out Ukrane friends have shown.
Could it be done better, yep.  

Also the bus length issue for the Qbus set is a trivial limit as the 
modern 3.5" IDE (or better yet 2.5") is small enough to plant on a card
in a bus slot!  

Allison