Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!dptg!att!chinet!saj
From: saj@chinet.chi.il.us (Stephen Jacobs)
Newsgroups: comp.sys.atari.st
Subject: Re: Empire v.3 / Empire II Information
Summary: You'd have to pay me to do stuff on Intel chips I do for fun on 68000
Message-ID: <9514@chinet.chi.il.us>
Date: 8 Sep 89 04:42:56 GMT
References: <8908291701.AA18947@ucbvax.Berkeley.EDU> <480025@hpdml93.HP.COM>
Organization: Chinet - Chicago, Ill.
Lines: 36

In article <480025@hpdml93.HP.COM>, rona@hpdml93.HP.COM (Ron Abramson) writes:
> 
> 	I don't know about the rest of you Atari guys, but I'm beginning
> to look just a little bit closer at those clones.  I'll be walking
> around the electronics section at some department store and there's 
> some "Clones 'R' Us" computer for $799.99999999999... and I start
> thinking "Gee, imagine having a choice of 14,000 games without driving
> more than a mile from my home.  Imagine getting on any one of 17,000
> BBS's with programs for my machine without dialing long distance..."
>

It's all in what you like to do.  I've gotten pretty friendly with the owners
of a PC-Mac type computer store.  I'll eliminate the Mac as a hobby machine
because of price and bizarre programming rules.  PC clones give you a very
decent level of performance per dollar now, and there are a lot of good 
programs, some of them at reasonable prices.  For a developer with a unique
product, it's the ONLY ballpark to play in.  But for a guy who likes to 
write the odd bit of useful code, maybe modify some PD source to bring it
closer to the heart's desire, the ST is still a very good choice.  You have
a complete jungle of OS calls at various levels on PC clones, and then there
are memory models to worry about.  Even in 80386 'tiny' memory model, it's
hard to -ignore- the segmentation.  And if you want to write graphics oriented
stuff, YIKES!  The various not-real-comparable graphics modes!  The developers
kits that cost more than the computer!  Not fun.  And heaven help the developer
with a better program that isn't alone in its niche.  A genuine publisher of
an outstanding ST program that would be the best (but not only) in its
category if ported to the PC commented on the obstacles in my hearing--nothing
that a million dollars front money couldn't fix.
   The 68000 is near as doesn't matter a ripoff of the PDP 11 architecture.
The PDP 11 is doing something unheard-of for a computer this year: passing
the 20-years-in-production mark.  It's lasted because a lot of right tradeoffs
were made.  The ST, and the successors-to-ST should remain as platforms for
a lot of adventurous and experimental programming, because they're based on a
good processor architecture and because good development tools are affordable.
                                   Steve J.