Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!lll-crg!seismo!gatech!akgua!akguc!mtune!mtuxo!hsc
From: hsc@mtuxo.UUCP (h.cohen)
Newsgroups: net.unix,net.arch,net.micro,net.micro.pc
Subject: Re: Unix V vs. Xenix V Performance
Message-ID: <1734@mtuxo.UUCP>
Date: Wed, 2-Jul-86 18:14:54 EDT
Article-I.D.: mtuxo.1734
Posted: Wed Jul  2 18:14:54 1986
Date-Received: Fri, 4-Jul-86 07:40:25 EDT
References: <400@galbp.UUCP>
Organization: AT&T Information Systems Labs, Holmdel NJ
Lines: 12
Xref: watmath net.unix:8507 net.arch:3648 net.micro:14990 net.micro.pc:8978

We ran the AIM suite on the AT&T PC6300 PLUS with AT&T
UNIX/Simul-Task (UNIX SVR2) and with SCO XENIX V.
Our SVR2 was notably faster than XENIX V on "logic" and notably
slower on "tty."  The other subscores were roughly comparable.
As a matter of policy, we do not publish the exact numbers.
They are too dependent on factors that change relatively frequently,
like the disk speed, swap space size, etc.
Also, we are working hard to improve our weakest areas, and
I'm sure the XENIX vendors are doing the same.
The main point is that you can not chose between XENIX and UNIX
on some global speed or efficiency basis; you need to consider 
your own specific applications and conditions.