SCUMM \ SCI game interpreter?

I've started asking questions over at ScummVM and am pretty sure I'll get laughed out of the forum but I just wanted to know what the techy guys around here thought. I figure that SCUMM \ SCI1 EGA games shouldn't be THAT difficult to get running on the the Mega Drive? I could be wrong the SEGA CD did have all that extra ram for Monkey Island so I could be talking rubbish.

My thoughts are we could possibly develop a SCUMM \ SCI interpreter with some sort of resource conversion tool to get them into a MD friendly format. It would make homebrew point and click adventures for the MD much easier to envision...

Okay well I've starting discussing things in a bit more detail and one of the SCUMMVM leads replied to my message. It's been welcomed as most people want to port it to .NET.

The Sega CD version of Monkey Island does use a version of SCUMM and its resources they advised me that what I needed to do was to take the SCUMM engine and strip it down to the bare minimum.

Start by not enabling HE games and SCUMM_7_8 which are more resource-demanding. Unfortunately a lot of the engine would need rewriting into a non-OO manner for it to work with the C compiler as the source is written in C++.

Quote:

---

Unfortunately a lot of the engine would need rewriting into a non-OO manner for it to work with the C compiler as the source is written in C++.

---

My plate is really, really full lately but that's the kind of think I'm used to do. C++ to C, C to C#, C to Java, C++ to Java, Java to C#, etc.
A first step, IMO, would be to rewrite all the C++ classes as C structs and have the class methods converted to method pointers in the structs.

It would be extremely cool to have a project like this going in full force.

If nobody helps you out now, maybe we can combine forces in the future.
Anyway, let me know if you need anything, I may be able to help you to some extent.

Quote:

---

Originally Posted by Barone

My plate is really, really full lately but that's the kind of think I'm used to do. C++ to C, C to C#, C to Java, C++ to Java, Java to C#, etc.
A first step, IMO, would be to rewrite all the C++ classes as C structs and have the class methods converted to method pointers in the structs.

It would be extremely cool to have a project like this going in full force.

If nobody helps you out now, maybe we can combine forces in the future.
Anyway, let me know if you need anything, I may be able to help you to some extent.

---

I've got one more game I'm in the process of writing. Once I've got that and Mega Dracula to a point of ROM release, I'll definitely come back to this and you're welcome to help.

It would make sense to try and make this a team effort as there is a huge amount of work involved.

Wasn't there a demo a while back of Day of the Tentacle that was posted on SpritesMind? I was wondering if that would be any help for this project of yours

The engine itself could possibly be used but the demo didn't do much more than show day of the tentacle graphics and a character moving via a cursor. We'd want to use the original resource files in conjuction with a custom scumm engine, which is way more work than that demo shows.

Plus day of the tentacle wouldn't be my first choice it has way more than 16 colours on screen. I want to start with games that would be pixel perfect.

I was curious so I downloaded the ScummVM source and built it for Win32. It's way bigger than I expected. Even if you cut everything out that isn't scumm, there's gonna be a lot of work just to pair everything down to support just EGA games. Not to mention the C++ to C conversion. I think it's still possible to do, but I don't think it'll be very fun.

Looking at http://wiki.scummvm.org/index.php/SCUMM/Versions, graphically the Genesis looks like it could support games up to version 4 of the scripting engine. http://wiki.scummvm.org/index.php/SCUMM says the scripting engine was rewritten from scratch several times, so later versions aren't necessarily backwards compatible. And looking at the code tree, each version does indeed have its own interpreter.

You could start with taking the full ScummVM code tree and cut out everything that's not Scumm and everything version 5 and later. Then convert to C and tailor the code for the game(s) you want to support.

That was my thoughts on it. Plus anything that uses the really old text input method could be removed. So you'd find a lot of other code segments of no use.

It's a huge massive task and one that I know would be of great value to the community given how few point and click adventures we got given!

I for one am supportive of this effort, the Genesis really needed more graphic adventure games.

One could always start from the old FreeSCI code, which was mostly C and is still available here: https://github.com/wjp/freesci-archive

Quote:

---

Originally Posted by TheMole

One could always start from the old FreeSCI code, which was mostly C and is still available here: https://github.com/wjp/freesci-archive

---

unfortunately these are keyboard based games! We really need the "point and click" based SCI games but nice find!

Quote:

---

Originally Posted by Guntz

I for one am supportive of this effort, the Genesis really needed more graphic adventure games.

---

Actually, if anything, the Genesis was the one console with more games that had elements of(talking the interface, along with the animation).

Scooby Doo flat out was one, more than anything on any other 2D home console at the time[edit: forgot the NES had a port of Maniac Mansion], and then there were Beavis and Butthead, and Cosmic Spacehead, as a couple full of elements of the genre.

The main thing here is, once you have the interpreter, how large is that going to be, combined with how large the games are themselves?