NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/mac/misc/15720

Internet Message Format | 1992-08-31 | 5.6 KB

Path: sparky!uunet!noc.near.net!mars.caps.maine.edu!maine.maine.edu!ree700a
Organization: University of Maine System
Date: Monday, 31 Aug 1992 12:03:39 EDT
From: <[e-mail]>
Message-ID: <[e-mail]>
Newsgroups: comp.sys.mac.misc
Subject: Mac's vs. PC's (split for Macs cost too much)
Lines: 93

   Well, I came into this general thread with good intentions.  Obviously
the former subject line (Macs cost too much - not!) has degenerated into
a free-for-all.  So let's try to establish a new subject line.

   First, I will briefly describe myself, so when you do flame me, you
won't have to make up arbitrary slurs & assumed incompetences... Use real
ones.

   I have never owned a computer... I use them too much at work...
   I have programmed in Z-80 assembler (TRS-80), VM/370 assembler, and 6800
assembler.  I have over 12 yrs experience with Fortran and a couple with C.
I have built rudimentary computing machines with no IC's (ie transistors).
OK, they were only 4-bit adders with registers...  I have designed micro-
processor based systems on the 6800 and 68000 (a communications interface
for my senior project - '84 and a joint controller for a robotic joint which
was never built.)  I am a fair (but not good) Windows programmer and understand
the fundamentals of message-based cooperative vs. preemptive multitasking.
   I use Word 5, Cricket Graph and Superpaint on the Mac because they are
unequalled or not quite equalled by their PC counterparts.

   I find the Mac to be the system of choice for producing paper (graphics,
literature, etc.  I never enterred this group waving the PC banner and shouting
"Macs cost too much".  In fact, the $1200 Mac Classic II is a very good match
for a similarly priced (as low as $900 for a reliable one with monochrome)
386SX.  In this case, $300 is noise compared to ease of use.  I would decide
for the SX only if I wanted to do a lot of programming.  To do work with
existing app's, I'd take the Mac in this case.

   I did challenge Eddy to a dual of performance for a $6000 budget, but he
just threatened to get me kicked off the network.  I would like to reissue
this with different objectives, ie MS Word scroll speed, other cross-platform
programs (no- Computer Associates' Hack of Cricket Graph for Windows is not
included in this... It sucks badly).  I'll even have to suffer the overhead
of Windows since Dos is virtually useless & I haven't gone to OS/2 yet.  I
might loose, especially letting the Quadra users choose the Applications...
They must be readilly available, cross-platform things... C-compilers?
Word?  Not sure what others...
   I'll agree that MIPS are meaningless across architectures, MFlops are over-
rated and pixels/second are too esoteric...

   Now, for why I really came to post this...
The 680x0 has a superior memory architecture, what with 32-bit flat mode
and 16-bit relative addressing.  I don't know what the Mac did for partitioning
things like video into the memory map, but if you don't need to know, they
did it right...  I don't know about protection of other application's memory,
but I know it exists.

The 80x86 family of CPU's suffers primarilly from upward compatibility.  The
8088 was useless and the 80286 was marginal.  Many of their limitations
have been passed on to the '386 and '486.  This is not to say that they are
really limitations... The 386 and 486 can address 32-bit flatmode memory.
For that matter, the software can address 4 billion independant segments of
32-bit memory (if someone writes such an OS...) and physically implement it
on disk (if the nations entire manufacturing capabilities went into disk
protection.

The memory limits on the PC are: 640K for DOS (is actually 1080K with no
tricks on a 286 or better, but some of it is used...).  There is a 16MB
limit on some systems due to address lines being limited on the old bus.
There is a 64MB limit of fast (directly addressed) memory on most current
boards, but there is nothing preventing more memory being designed onto
newer boards.

     My point here is that there is no fundamental capability gap
     between the Mac and PC relative to Memory Addressing.  As far
     as software goes, yes the Mac has always had better memory useage
     than the PC, but software advantages are so fleeting...

Although Ed doesn't think cooperative multitasking is a problem, he should
consider that each track and draw operation is a different is a different
message being processed from a different mouse interrupt.  The draw package
is yielding control constantly during that process...  At least, I know that's
how Windows does it & if MacOS isn't the same, what's the basis of Apple's
suit against Microsloth?  The GUI from Xerox (I think it was?).

By the way, Ed, What is your background: Physiology? Optimizing Compiler's?
Color Matching at the paint store?  just wondering...

IMHO both the Mac and the PC have alot to offer & I am planning on getting one
of each!  A Mac Classic II to do my PhD Dissertation on (Acoustic Plate Mode
Biosensors...) and a PC to perform data acquisition, analysis and numerical
analysis.

Since we were on the subject of analogies in the previous thread, here's mine:
I ride a motorcycle because I like it, but I take the car to go get groceries!

As a final statement (in this Post), I think Apple is a little conservative
in the performance department, kind of like IBM was while it had the luxury
of no competition.  I'm sure some IBM patent attorney is saying, "How did
Apple protect it's Toolbox, while our BIOS was copied freely?"

Jeff Andle, Department of Electrical Engineering, U. Maine
"There is no right answer, it isn't even graded.  Now, take your test please!"