🚀 go-pugleaf

["The Knights who say Letni demand.....A segment register!!!"]

>I agree that the Motorola chips are nicer in many ways than the Intel chips.
>But to say Motorola has "released" the chip is stretching things a bit. 
>...
>The 80386 is here now, and actually
>has a system available (Compaq) which uses it.  In other words, it's not
>really fair to compare that which Intel has now with what Motorola will have
>in a year.  A more fair comparison is between the 68020 and the 80386.

I agree that you cannot "fairly" compare the 68030 to the 386 by chip. 
You should compare the system as a whole.  For example, Compaq does have
a 386 box, but try one (I have).  You get a copy of 8088 MS-DOS that has been
hacked just enough to boot, so Bill Gates could get his picture in
Infoworld. :-)  The debugger supplied doesn't even know about the 386
32 bit registers or opcodes!  The "386/286" version of MS-DOS that really
uses the hardware is a year away also.  Same thing for XENIX-386 from
Microsoft, which will be available to OEM's sometime in 1987.  (See Microsoft
announcements in Infoworld for their estimates.)

Now when the 68030 part becomes available and is put in a system,
golly gee, it should work with little or no hacking, and all the
assemblers and compilers out there will already work with those
32 bit instructions and registers.  All of the operating systems
out there might, at worst, have to be recompiled.  Sure, the assemblers
and compilers might be modified to take advantage of any new instructions
that pop up, but from what I read, most of the new items are for supervisor
mode, so you wouldn't put too many of them in the compiler. 

So you have the 68030 with a long lead-time till the hardware is
available, but a short (or non-existant) port time for software,
and the 386 which is relatively available now, but has a long lead-time
to get any software (to say nothing of a real 386 assembler or compiler)
that actually uses more than the 8/16 bit part of the CPU.  How soon are
Lotus and DBase going to report their packages to use 32-bit mode?

When you compare the point when you get a functional system that actually
USES the power you paid for, the 68030 and 386 are really tied, with more
risk of delay on the 386 side because EVERYTHING has to be recompiled
(even awk) to take advantage of the 32 bit arch.  (I don't know a single
XENIX/UNIX programmer who isn't ready to recode his/her stuff WITHOUT
segments!)  On the 68030, application programs should not have to do
anything to take advantage of the performance boost.

<The above is the opinion of me and not that of my employer (who thinks
 Intel invented Happy Hour) or my Golden Retriever (who ate all of my x86
 manuals long ago).  Until the EPCA becomes law, one can still have
 opinions in the US.  OPPOSE THIS BILL!!!!>
						
						"Thank you, Uh Clem."
						Frank Durda IV
						@ <trsvax!uhclem>

The 386 is a registered trademark (I guess) of Intel Corp and not NEC or IBM.
The 68030 is a registered trademark (should be patented too :-)) of Motorola
and comes from Austin, which is nearer to "Where the best begins", Ft. Worth.