home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Subject: Re: What happened to the UCSD p-System?
- Path: sparky!uunet!ornl!rsg1.er.usgs.gov!darwin.sura.net!zaphod.mps.ohio-state.edu!sdd.hp.com!think.com!unixland!rmkhome!rmk
- From: rmk@rmkhome.UUCP (Rick Kelly)
- Organization: The Man With Ten Cats
- Date: Sun, 22 Nov 1992 07:31:24 GMT
- Reply-To: rmk@rmkhome.UUCP (Rick Kelly)
- Message-ID: <9211220231.26@rmkhome.UUCP>
- References: <1992Nov17.162428.15881@coe.montana.edu> <1992Nov21.055139.18590@foretune.co.jp>
- Lines: 64
-
- In article <1992Nov21.055139.18590@foretune.co.jp> trebor@foretune.co.jp (Robert J Woodhead) writes:
- >uesu03@giac1.oscs.montana.edu (Lou Glassy) writes:
- >>Is any work going on in this area? Or does performance pay
- >>more than portability?
- >
- >The UCSD p-System had many advantages for its time. Recall, back in
- >those dark ages, machines were slow, memory was tight (64k was huge)
- >and hard discs were non-existant. Even floppies rarely had more than
- >150k on them.
- >
- >So packing a decent OS that could run reasonable programs into such
- >environments was tough. The p-System attacked this problem in several
- >ways:
- >
- >1) p-Code instructions were highly optimized both for space and ease
- >of interpretation, which meant that the interpreters were small, and
- >the p-Code programs themselves were smaller than equivalent machine
- >code programs; so much so that usually OS+Interpreter+p-Code program
- >was smaller than the native machine code program alone!
- >
- >2) That smaller size really helped when you consider the size and
- >speed of the storage devices.
- >
- >3) p-Code was easy to compile for, which meant that the Pascal compiler
- >was small (ie: it could actually run on a 64k machine!)
- >
- >4) It was portable, which meant that it was easy to support all the
- >gazillion different machines out there.
- >
- >For it's time, the UCSD p-System was a great achievement, but like
- >all things, its time has passed. I myself used the Apple Pascal
- >version to write Wizardry (14,000+ lines, originally done on a
- >64k, 2-floppy machine; full compiles took an hour!).
- >
- >Wierd side story: When Wizardry was translated into Japanese, the
- >Japanese company that did the work (and from whose premises I am
- >writing this note) simply wrote p-System interpreters for all the
- >target machines, and we made minor modifications to the Wizardry
- >program to handle Japanese characters and syntax (the final version
- >let you choose at boot time if you wanted to play in English, Kana
- >or Kanji). After we were done, we back-converted their interpreter
- >to run either standalone or under MS-DOS; I wrote Wizardry IV on a
- >NEC9801 MS-DOS machine that, after a suitable command line incantation,
- >promptly said "Welcome to Apple Pascal!"
- >
- >(And yeah, I bought an extra copy of Apple Pascal for the NEC, too!)
-
-
- When I read "4)" I almost hit 'F' to say that Wizardry I on the Apple II
- was a great example of the efficiency of UCSD Pascal. :-)
-
- Luckily, I read further before I put my foot in my mouth.
-
- Along with all my other computers, I still have an Apple IIe here, and I
- boot up Wizardry every now and then and remember back to the days of
- excitement.
-
- Wizardry on the Apple is/was very responsive considering the speed of the
- floppy media. It should probably be considered as a significant stepping
- stone in the evolution of computer games.
-
- --
-
- Rick Kelly rmk@rmkhome.UUCP unixland!rmkhome!rmk rmk@frog.UUCP
-