videofx
Member
Hi,
I have Zork I, II and III. Anyone have a zip with all th other Infocom adventures?
Thanks
Bill
I have Zork I, II and III. Anyone have a zip with all th other Infocom adventures?
Thanks
Bill
I don't think it is possible. Infocom never released a CP/M interpreter for anything newer than V3. I am not aware of any third-party interpreters (for any version) for CP/M either.Whats needed is a LOADER to load and run .Z5 files in CP/M, is that possible ?
Looks like the source is available - it probably wouldn't be too difficult a job to cross-port it to Z80 CP/M machines.Well, ZXZVM has Z3, Z4 & Z5 files running on a Spectrum +3 and all the 8bit Amstrads. Check it out at https://www.seasip.info/ZX/zxzvm.html
I've heard people say that so many times, I specifically address it in the porting guide.Looks like the source is available - it probably wouldn't be too difficult a job to cross-port it to Z80 CP/M machines.
I have a Microbee 512k unit that runs CP/M 2.2 (actually ZCPR2) that gives me a 414k RAM drive (M: ) So reading your Porting Guide -I've heard people say that so many times, I specifically address it in the porting guide.
The more you know! XDI've heard people say that so many times, I specifically address it in the porting guide.
Also have a look at other people who have ported John's excellent work (including myself). Someone has done a page-file based CP/M version, and I've done reduced game support TRS-80 versions that run inside a 64K or even a 48K footprint.The more you know! XD
I would suggest the main limitation is proper screen control. To run V4 and higher, you need a lot of control over cursor positioning and scroll protect for split-screen, and to get the most out of it, highlighting as well. For some games yes the memory model is a challenge too, but I'm able to run Arthur: The Quest for Excalibur, a version 6 game, on my TRS-80 model I or model 3 (2MHz Z80, 64x16 display, 48K RAM).I don't think it is possible. Infocom never released a CP/M interpreter for anything newer than V3. I am not aware of any third-party interpreters (for any version) for CP/M either.
My gut feeling tells me that CP/M machines simply lack the ability to run newer adventures efficiently, due to lack of memory and address space.
There are quite a few post-infocom z3 games, and now thanks to the PunyInform library there are very recent ones. Hibernated 1 (Director's Cut) is commercial quality (including available for some platforms on actual media) which you can check out at https://8bitgames.itch.io/hibernated1Oh, a recent Z3 text adventure is Tristam Island from https://hlabrande.itch.io/tristam-island for both a demo & payed version.
Interesting idea. I've seen a few cut down versions of fweep. Does yours support the status lines, highlighting and all versions from the one binary? I'm interested because it may solve another use case.The cut down Fweep I use in Fuzix will run almost any v1-5/8 game in 32K of RAM. The approach used would also work for ZXZVM I think if you wanted a portable CP/M + RAMDISK version. You'll need a good fast ram disc and some patience to run the bigger Inform ones though as they have a very bloated working set.
FUZIX/Applications/games/fweep.c at master · EtchedPixels/FUZIX
FuzixOS: Because Small Is Beautiful. Contribute to EtchedPixels/FUZIX development by creating an account on GitHub.github.com
Rather than fetching bytes from banked RAM it keeps a cache of roughly least-regularly-used blocks and fetches from disk as needed, as well as keeping a backing file for the writable parts of the game state. Providing that disk is fast (RAM, CF, etc) then all but a few Inform games run at a nice speed. If you've got more memory (it doesn't dynamic size which it should) it'll work even better
So I'd guess if you nicked the algorithm from fweeplet and pasted it into ZxZVM and used a ram disc you'd get the result you needed for a CP/M 3 version using the ramdisc via the CP/M API. You might need 512 byte blocks not 256 so that you avoid any extra copies (physical block size, and the CP/M 3 fast paths for block copies) but that shouldn't reduce performance too much. Anyone who wants to do so has my permission
On my RC2014 systems the CF card appears to be faster than the RAMdisc 8)