Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!uunet!dg!sequoia!greene
From: greene@sequoia.rtp.dg.com (jon greene)
Newsgroups: comp.sys.m88k
Subject: Re: Data General Aviion 7000, 8000
Keywords: New Data General 88K machines
Message-ID: <1313@dg.dg.com>
Date: 22 Mar 91 00:03:15 GMT
References: <1991Mar14.183029.11714@neon.Stanford.EDU> <1991Mar16.151948.14745@dg-rtp.dg.com> <15515@june.cs.washington.edu>
Sender: root@dg.dg.com
Reply-To: greene@sequoia.rtp.dg.com (jon greene)
Distribution: comp
Organization: Data General Corporation, Research Triangle Park, NC
Lines: 44

In article <15515@june.cs.washington.edu>, robertb@cs.washington.edu (Robert Bedichek) writes:
|> Would someone from Data General give a serious explanation for the
|> claimed 117 MIPS figure for the new workstations?

These systems are servers...not workstations.

|> 
|> Just saying "everybody lies, so do we" doesn't cut it.  The number must
|> come from somewhere.  Even outrageous numbers have some basis.  What is
|> the basis for 117 MIPS from the new DG machine?

It is based on Dhrystone performance relative to the VAX 11/780.  This is consistent with industry practice and is openly acknowledged by DG.

|> 
|> I read that the workstations use four 25 MHz 88100's.  Given the
|> optimistic figure of 17 MIPS for a 20 MHz part (the figure given by
|> Motorola) implies each 25 MHz 88100 yields 21.25 MIPS, x4, that is 85
|> MIPS.  I believe that Motorola claims 60 MIPS for their four-CPU 20 MHz
|> systems.  Given this, that would put the DG machines at 75 MIPS.  And
|> scaling the MIPS figures with clock speed assumes that the memory
|> system can be sped up without problems.
|>
 
I'm not sure where you got the Motorola numbers.  I have in front of me a press release for the Motorola MPC 100 where they claim 27 MIPs for a 20MHz system.

|> And of course, these DG figures are aggregate MIPS, not single CPU
|> MIPS, so it is meaningless to compare a 40 MIPS SUN to a ?? MIPS DG
|> machine.  If the load is single thread, the fastest SUNs, MIPSco
|> machines, and the IBM RS/6000 will win.
|> 

As noted above, the AViiON 7000 and 8000 series machines are servers, not single user workstations.  Therefore, they are intended to run many processes, not one single-threaded process.  For this type of environment, an efficient SMP implementation such as in DG/UX is required.  To my knowledge, none of the vendors that you listed support SMP operation.

|> Someone from DG, please help us understand your company's claims.
|>
 
I hope that I have.

|> 	Robert Bedichek                robertb@cs.washington.edu

Jon Greene
Data General Corp., Westboro, MA

I'm sure it's obvious, but I'll say it anyway....I speak only for myself and not as a representative of my employer.