Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 alpha 4/15/85; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!cbosgd!ucbvax!info-vax
From: info-vax@ucbvax.ARPA
Newsgroups: fa.info-vax
Subject: Re: microVAX III and busses
Message-ID: <7791@ucbvax.ARPA>
Date: Tue, 4-Jun-85 17:56:38 EDT
Article-I.D.: ucbvax.7791
Posted: Tue Jun  4 17:56:38 1985
Date-Received: Thu, 6-Jun-85 03:21:44 EDT
Sender: daemon@ucbvax.ARPA
Organization: University of California at Berkeley
Lines: 34

From: rna!dan@nyu-cmcl2 (Dan Ts'o)

> I find the rumors about things truly amazing, especially the microVAX
> III. A recent note mentioned that the next microVAX will not have a
> Q-bus, but might use a CI or a UNIBUS. This is very unlikely.
> The Q-bus is a pretty fast bus, almost twice as fast as a UNIBUS in

	I have heard enough rumors about the uVAX III to believe its more
than fantasy. Much of the industry speculates that the uvax III will be
the introduction of the new BI bus, a DEC 32bit bus designed to compete
with VME and Multibus II. A year or two ago the Multibus I, Qbus each shared
about 30% of the market. Since then, VME has really grown while the Qbus
has shrunk. DEC really needs a better bus to compete.

	The Qbus is not a pretty fast bus. Most SMD controllers on the Qbus
still need to do sector interleave (as much as 3:1) to slow down the transfer
rate onto the Qbus. True, the new block mode DMA protocol and the PMI for the
uvax II help somewhat. But the Qbus is plain slow when compared with VME or
Multibus II.

	So the uvax III may well be the 78032 chip set on the BI bus. The
chip is sure plenty fast. The I/O bus need help. There is also supposed to
be yet another VAX chip set in the works, a multichip set so called VLSI VAX
which has somewhat better performance than the 78032 of the uvax II. Anyone
know anything about it ?

					Cheers,
					Dan Ts'o
					Dept. Neurobiology
					Rockefeller Univ.
					1230 York Ave.
					NY, NY 10021
					212-570-7671
					...cmcl2!rna!dan