Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!gatech!rutgers!bellcore!tness7!ninja!sys1!techsup!cpe!neese
From: neese@cpe.UUCP
Newsgroups: comp.sys.ibm.pc
Subject: Re: MCA (was: message <1634624868@r
Message-ID: <12400008@cpe>
Date: 5 Jul 88 14:37:00 GMT
References: <902@isrnix.UUCP>
Lines: 34
Nf-ID: #R:isrnix.UUCP:902:cpe:12400008:000:1683
Nf-From: cpe.UUCP!neese    Jul  5 09:37:00 1988


>>What he didn't tell you the first time around that his machine:
>>						Roy Neese
>>					UUCP @	ihnp4!sys1!cpe!neese
>                                                           ^^^
>is a system owned by Tandy's Computer Products Engineering group.
>
>Come now, Roy...didn't you want to admit that you were just plugging
>your own products?
>
>the tandy favoratism was noticable, but more unfortunate was the false
>claim that 32 bit mca memory expansions exact a 400% performance penalty
>on a model 80.  infoworld compared mca memory expansions several issues
>ago, including the ast advantage/2-386.  they found that board to be equal
>to the planer board memory speed.  my inquiries to other board makers abt 
>this claim turn up only denials of any penalty.
>
>also unmentioned was the newly announed model 70-a71, running at 25mhz
>and using memory cache.  they may be late, but it can't be argued (as it
>was) that ibm can't or won't make performance-competitive products.

Actually, I was not plugging our product, but simply stating the performance
hits we have noticed with our Model 80-111 and the AST MC Memory Board running
SCO Xenix.  Regardless of what infoworld stated, there is a penalty for
accessing memory that occupies an MC slot.  This is by design.  Planar
memory will always be faster than MC memory.
I mentioned the 5000MC because it is the only other MC system I have to compare
to the 80-111.  It does run circles around that system, which happens to be
my opinion and fact as well.  
While the posting may have favored toward Tandy, I haven't seen much to refute
what I stated and still stand behind it.

						Roy Neese
					UUCP @	ihnp4!sys1!cpe!neese