🚀 go-pugleaf

In article <nevin-0809950153300001@mac-nevin.cs.arizona.edu>,
Nevin ":-]" Liber <nevin@CS.Arizona.EDU> wrote:
> I have a little secret for you:  the only difference between CMT and PMT
> is how tasks get swapped; ie, the whether scheduling is under application
> control or OS control.

In other words, whether sceduling is done by code written bu OS developers
or by application developers.

In a general purpose computer, preemptive sceduling is a complete win. There
is no downside to the user. It makes a little more work for application
developers who aren't used to a multitasking environmnet, but those are the
ones who screw up and steal the CPU for way too long under cooperative
multitasking.

The only place cooperative multitasking is a win is on a dedicated computer
with a very small mix of applications (usually just the one, replicated over
and over again).

And if you want to compare operating systems, why don't you at least compare
similar systems, instead of comparing a multiuser/preemptive/fair share
scheduler with a single user/cooperative/demand driven one. Change one thing
at a time... compare Apples and Amigas.

Like the Mac, the Amiga is a single user/low security system with a demand
driven scheduler. The biggest difference between it and the Mac is the
preemptive multitasker. For modern systems, 68020 and above, the blitter
is pretty much irrelevant: a 68020 can swamp the bus.

The Amiga O/S provides a much more responsive environment than the Macintosh
does, even on much smaller and less powerful systems. The first comparison
I did was between an Amiga 1000 (7 MHz 68000, plus blitter) with a Mac SE
(68020 or 68030 at 16 MHz). For ordinary window operations the Amiga was
faster, for application startup the Amiga was faster, and overall it was
more responsive.

It was, of course, slower on CPU-intensive tasks... but where the O/S and
scheduler was involved this much less powerful machine put the Mac to shame.
When running CPU-intensive tasks in the background the Mac was almost
unusable. The Amiga gave no perceptible indication that the entire CPU
wasn't at the user's disposal.

A later test with Windows on a 386 at 16 MHz was even more impressive. The
Windows box was much slower than the equivalently powered Mac.

> So, why don't you tell us *how* PMT is going to make our lives oh so much
> better, instead of emptily asserting it without proof or justification
> that it will.

On an even playing field, it's the clear winner.
-- 
Peter da Silva    (NIC: PJD2)      `-_-'             1601 Industrial Boulevard
Bailey Network Management           'U`             Sugar Land, TX  77487-5013
+1 713 274 5180         "Har du kramat din varg idag?"                     USA
Bailey pays for my technical expertise.        My opinions probably scare them