NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/mac/misc/15606

Internet Message Format | 1992-08-29 | 3.6 KB

Path: sparky!uunet!noc.near.net!mars.caps.maine.edu!maine.maine.edu!ree700a
Organization: University of Maine System
Date: Friday, 28 Aug 1992 10:09:10 EDT
From: <[e-mail]>
Message-ID: <[e-mail]>
Newsgroups: comp.sys.mac.misc
Subject: ewright is full of Bull.....
Lines: 66

+++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++

If you're crunching floating-point numbers, the Quadra's 68040
will be significantly faster than the 486/50.  No matter what
computer salesmen may tell you, you can't can't just look at
the clock speed (MHz) of a chip and tell which one is faster.
I don't know why you believe adding data-acquisition is difficult
or expensive -- a Nubus board takes about 30 seconds to install.
And 16 bits, while it might be pretty good by PC standards, isn't
close to true color.

+++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++
    Now, E. Wright has some points here... And of course the Motorola
family of chips has traditionally used a different clock scheme than
the Intel family, Yes, 16 bits is incapable of properly rendering a
good photograph with lots of flesh tones, but as I replied to him ( I
should have saved outbound mail as well ), 24 bit graphics is.  see last
section as I don't know how to cut & paste on this screwy mainframe
editor... ( I will never speak in support of the IBM VM/370 !)


    My reply to the comments of E. Wright to the above (listed last)
 follow below...

  I'm a 29 yr old Electrical Engineer with course background in cognition,
psychology, computer architecture, etc, along with my specialty of Microwave
Engineering.  If all you can do is spout nonsense denigrations of your
adversaries on no basis, then you are not worth debating with.  The human
eye, with the possible exception of yours, cannot distinguish between two
colors in a 24 bit RGB format (on average).
  As to lab benchmarks, if it is possible that a 68040 could outperform
an 80486, it was not in a MacIntosh.  I for one (see post ...best of
both worlds..), will readilly agree that the relative addressing format
of the 680x0 series is far superior to the segmented memory model of
the 80x86 family.  You, unfortunately, never make this highly relevant
point, nor (although you bit heads about a PC Jock knowing it) did you
ever make the correct comment on the resource toolbox:  Mac code is, on
average, half the size of the corresponding PC code on disk (Word 5 vs
the Windows port of it, Word for Windows 2.x.  This is because of the
shared resources that each & every application doesn't have to share....

     Research Talks, Bullshit Walks.    Have a good walk...
     Jeffrey C. Andle
     Laboratory for Surface Science & Technology
     Department of Electrical Engineering
     University of Maine
     Orono, ME  04469

+++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++ +++++++


Sorry, but all the laboratory benchmarks show that the 68040
is about 50% faster than the 80486, at the same clock speed,
for floating-point calculations.  As far as 16-bit graphics
being "true color" is concerned, this is nonsense.  In fact,
even 32-bits isn't enough.  I suggest that you find a book
on computer graphics and read the chapter on color theory
or ask one of your professors to explain it to you.  I'm just
getting too damn old to argue these things with kids who didn't
bother to do their homework.

     So, Unless E. Wright starts throwing facts & figures instead of
juvenile temper tantrums and personal slurs (which, unlike this one,
are unfounded) I, for one, will not even read his posts.  I encourage
others to do the same and save this debate from a lot of hot air.