Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site pertec.UUCP
Path: utzoo!utcs!lsuc!pesnta!pertec!mb
From: mb@pertec.UUCP (mike burg)
Newsgroups: net.games.hack
Subject: Re: Hack on a PDP-11/44
Message-ID: <300@pertec.UUCP>
Date: Thu, 7-Feb-85 01:31:40 EST
Article-I.D.: pertec.300
Posted: Thu Feb  7 01:31:40 1985
Date-Received: Thu, 7-Feb-85 05:07:16 EST
References: <552@ukma.UUCP> <248@turing.UUCP> <412@ark.UUCP>
Organization: Pertec Computer Corp., Irvine, CA
Lines: 31

> In article <248@turing.UUCP> play@turing.UUCP (The MC-funhouse) writes:
>It is not impossible; I have had a 11/44 version long ago. You'll have
>to start by throwing out vaults&guards, options, news, help, engravings,
>tracking properties of monsters, the choice of character at the beginning
>of the game, the tombstone, the camera, the scroll of punishment and all
>references to ball and chain, the icebox, the whistles and magic whistles,
>the commands 'a','o','=',')','[','/','\\','?'; if that is not enough (and
>it won't be) throw away all mention of gems, the special treatment of
>unicorns, the long worms (in the lincolnsudbury version 'w' was a wumpus -
>restore that situation), the wand of wishing together with the routines
>that parse a typed in object name, the wizard mode, the ring of warning.
>Throw away the routine docorner, and let inventories be given one item a
>time at the topline. Trim down the routines in topl.c. Omit the possibility
>to ask for repetition of recent messages.
>By now it should almost fit, and throwing away a few other small things
>you should be able to make it work. It is a lot of work.
>By the way, if I remember correctly (but this might have been on an 11/45)
>it is not enough to have the program down to precisely 64K, because some
>bug in loader or kernel will still prevent it from running. It should be
>64K minus 64 bytes.

	I have hack running on a PDP-11/44 running 2.9bsd. 2.9bsd has a
new feature that allows programs to be overlaied (I think maybe someone
has tried to put this feature in V7, also). Anyways, the whole program is
roughly over 220KB (34K of it which is the data, leaving around 25K to 30K
for run-time allocation, the highest the data segment has gotten was around
45K at run-time).

		Mike Burg
		decvax!trwrb!pertec!mb
PS, has anyone else try this?