mmm… I don’t have a repository (and I think isn’t a thing whose can be put into if-archive…), but anyway, these are the (simple) changes done to the “root file” (checkpoint.zil, whose is identical to i1.zil)
Interesting, there is a ZAP executable in the zork-german repository (under Zork 1). Unfortunately, the only associated file with it seems to be a zstring.c. The ZAP binary have been compiled using a m68k based GCC compiler. Maybe someone else here to extract more info out of it. No signs of a ZILCH binary.
In the later days of Infocom they did try to move from the PDP-10 to Macs, so most likely that ZAP binary is for Mac Classic. Possibly it could be tempted to run under an emulator.
That ZAP binary, a COFF object, is intended to run on a Unix system, not a Macintosh. Given that the Infocom file cabinet documents mention Sun hardware, I strongly suspect that it was intended to run on a pre-Sparc Sun computer of some sort.
There are other Unixy telltales in the binary that can be seen by running the “strings” command on it. I’m intrigued by the fact it was clearly written in C and this tidbit of text:
Copyright (c) 1987 Apple Computer, Inc.,
1985 Adobe Systems Incorporated,
1983-87 AT&T-IS,
1985-87 Motorola Inc.,
1980-87 Sun Microsystems Inc.,
1980-87 The Regents of the University of California,
1985-87 Unisoft Corporation, All Rights Reserved.
Yes, after the Cambridge office was shutdown it’s known that the Muddle interpreter was ported to run on A/UX on the Macintosh II. Zilch remained Muddle-based the whole time but they’d previously ported ZAP (which was originally written in Muddle) to PDP-10 assembly language for speed. It seems reasonable that they’d have a ZAP to run natively on A/UX too.
I’m trying to build and bugfix this code with full change histories here. Bugs can be reported on the individual game repositories and bug reports/feedback would be welcome.
I don’t know much about GitHub. I may or may not file bug reports, but I really hope you’re aware of the large amount of work that has already been done on finding Infocom bugs. Many of these are only in older versions and got fixed by Infocom, but many remained. See Graeme Cree’s list and mine for the vast majority of known Infocom bugs.
Bug reporting is still valuable for these new builds, though - there may be new bugs that were introduced as part of a fix, or introduced by Infocom after the last commercial release, or caused by bugs and/or incompatibilities in ZILF.
Yes thanks, I’m aware of these lists and they will, I hope, end up as problem/bug reports on the corresponding games. Any that aren’t submitted by others, I’ll end up submitting myself.
I start to think that Mr. Cree ought to release his TUIC scripts, even if are old DOS .BAT scripts, because now can be useful for tooling the research environment.
on a different matter, I think that because the minizorks shows a definite evolution of the source organisation and layout, is important to figure what was the zork version at the base of the two minizork I.
researching minizork I “1982” I doubt on the dating, albeit the idea was already sound in 1982 (exploiting the rather genial bankswitching mechanism of the C= 64, esp. the “ram-under-rom”)
example: the bat swooping don’t refer to wumpus, and albeit one can’t exclude that was part of the excising done, this point toward a Zork I codebase after release 30, serial 830330.
