Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!lll-crg!topaz!husc6!harvard!panda!genrad!decvax!tektronix!uw-beaver!ubc-vision!ubc-cs!manis From: manis@ubc-cs.UUCP (Vince Manis) Newsgroups: net.micro.atari16 Subject: Re: The 68020 Suggestion Message-ID: <326@ubc-cs.UUCP> Date: Wed, 23-Jul-86 23:31:59 EDT Article-I.D.: ubc-cs.326 Posted: Wed Jul 23 23:31:59 1986 Date-Received: Thu, 24-Jul-86 21:49:46 EDT References: <860720202512.001@Juliet.Caltech.Edu> <5311@sun.uucp> Reply-To: manis@ubc-cs.UUCP (Vince Manis) Organization: UBC Department of Computer Science Lines: 42 In article <5311@sun.uucp> cmcmanis@sun.uucp (Chuck McManis) writes: in response to Lingus (CCL @ CITROMEO): > >While we all admire what Atari has wrought with bringing the price of >68000 machines down to earth, your $3000 number is a bit absurd. Mostly >because the monitor alone to display 1200 X 800 monochrome graphics is >on the order of $1000 wholesale. Add to that the $200 price of *each* >68020 chip, multihundred dollar price of 68881 chips, 4 Meg of ram >(which at $3 per 256K chip is still $384) Add to a new box, bigger >power supply, you have to include a hard disk to make it run at any >speed. No, I don't think $3000 will buy you a Sun workstation ($8000 >will :-)) I think the best that can be said is that in 5 years it >will probably be possible. Ah well. This is no swipe at Sun (whose products I use and admire), but Atari would probably encounter a major marketing disaster if it tried to bring out a Sun-compatible right now. Why? Well, first of all, you need a pretty hefty disk (let's say a good fraction of an Eagle) to store Berkeley UNIX, along with the necessary goodies to make it useful. Second, installing a Sun is a non-trivial job (take a look at the installation manual some time): I'm not sure Atari's market in general is ready for that level of complexity (sure, lots of UNIX hackers have ST's, but lots of ST owners aren't UNIX hackers). Somebody could make a killing with a lean, mean, Berkeley UNIX clone which was aimed at low cost personal computers like the ST. (At the risk of getting into the interminable Berkeley vs System V argument, I'd just point out that much of the interesting graphics etc has been done on Berkeley Unix machines). The result would be a full-featured, multi-tasking (and even multi-user, to support bulletin boards) operating system which didn't require a networked file server. Maybe OS-9 is the answer; since I've never used it (though I did once read some of the manuals), I couldn't answer the question. In any case, without a resolution of the system question, I think Atari would be foolish to announce a box which required considerable support and was only of interest to techies. I wouldn't be surprised to hear that the TT's arrival awaits only two things: lower-cost 68020's, and some good application/system software to make people want it.