Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!brutus.cs.uiuc.edu!lll-winken!ames!ncar!ico!ism780c!randvax!ucla-an!stb!gendep!craig
Newsgroups: comp.sys.ibm.pc
Subject: Re: Speed in MIPS
Message-ID: <1JXZD8w160w@gendep>
Date: 3 Feb 90 02:46:23 GMT
References: <881@watserv1.waterloo.edu>
Organization: General Depravity 1 213 447 1543
Lines: 29

ssingh@watserv1.waterloo.edu ($anjay "lock-on" $ingh - Indy Studies) writes:

> 
> BTW, based on Intel's and Motorola's past record and their present design
> techniques, is it possible to say which is the better microprocessor, or
> are they aimed at different markets, where the 486 will smoke an 040
> on one application, while the 040 will hammer the 486 on something else?

"The better microprocessor" is kind of an unrealistic concept.  Both
Motorola's and Intel's chips work, work fast, and can do what you need
done; beyond that it's largely a matter of taste and convenience.
The Motorola's are arguably better for writing compilers, for
instance, because the instruction set is much more regular than the
Intel chips.  Then again, Motorola's multiply and divide instructions
take an amazing amount of time due mainly to the fact that Motorola
is into microcode and tends to minimize on transistors, whereas Intel
seems to like putting things in hardware whenever feasible.
 
Coming from a software development perspective, I would have to say
that the Motorola architecture and assembly language are much friendlier
than the Intel.  The only dedicated registers in the 68000 are flags,
the program counter, and the stack pointer, and with few exceptions
you can do as you like with the other 15 32-bit registers.

I must say, though, that with the 386 Intel finally did segmentation
right, and in protected mode segments are actually a help rather than
an annoyance as they are in the earlier 8086-family processors.  The
386's memory-mapping abilities are just what you need for writing
multi-etc. operating systems.