is SCUMMVM looking for an source game creation tool?

[Chris_Tolworthy]

Would the ScummVM community be interested in a proven adventure game creation tool becoming open source? This is my first post, so forgive me if I misunderstand the situation.

According to a recent thread on the AGS forums, some people here on ScummVM were hoping that AGS could become open source, so that AGS games could run on multiple platforms via SCUMMVM. I couldn't find the thread in question, do forgive me for starting this new one.

AGS developers were excited about this, but Chris Jones (the creator and king of AGS) said he can't make AGS open source due to private promises made to commercial users of AGS, though he left other possibilities open (e.g. an AGS-lite or something).

Anyhow, is it true? Is there a general interest in making SCUMMVM compatible with an established freeware game creation tool? Or is AGS a special case due to the large back catalog of AGS games?

It seems to me that if SCUMMVM supported a game creation tool, even an obscure one, it would attract a lot of developers to make new games, in order to gain cross platform support. Which brings us to Sludge and today's bombshell announcement.

I'm a great fan of Hungry Software's Sludge development kit. It's an excellent set of tools, as seen from the various Sludge games. It's also relatively simple. It does not use DirectX or DLLs (except for a sound engine), so it works well in emulators, and should be relatively simple to port to other systems. Last year Tim Furnish (its creator) made the engine freeware. He's becoming busier and busier in his real-world job, so doesn't have the time to devote to Sludge any more. A few of us dedicated Sludge users suggested that he make it open source.

So here is the point of this post: Today Tim said yes, he does plan on making "Open Sludge" when he gets the time.

Wow! My immediate thought was "what if Sludge was adopted by ScummVM - a match made in heaven!" Like the song says, Sludge and Scumm.... Scumm and Sludge... go together like a horse and carriage..."

Please note that I have no authority to speak for Tim or anyone else. All I know is what he said today. I just thought people over here might be interested.

I am a great fan of both Sludge and ScummVM, so an Open Source cross-platform Sludge would be a dream come true. But I wonder what other people think? Is there any demand for it? If Tim goes ahead and makes Open Sludge, would anyone care enough to work on it? Do the ScummVM folks really want to attract new game developers? Am I wrong to think that an interest in AGS could transfer to an interest in other engines? Was the interest in AGS blown out of proportion?

[eriktorbjorn]

Anyhow, is it true? Is there a general interest in making SCUMMVM compatible with an established freeware game creation tool? Or is AGS a special case due to the large back catalog of AGS games?

(I'm speaking only for myself here.)

Every now and then, someone will suggest that we support AGS games. I've always found this rather puzzling, since at least some of these suggestions seem to come from people who assume that we'd be able to do a better job without any prior knowledge of AGS (not to mention the source code) than Chris, who's been maintaining it for the past decade.

I think the general consensus among the developers is that we don't want to compete with any actively maintained project. It would just be a waste of time trying to keep up with them. This would include projects like AGS and (though I just know someone will disagree with me on the "active" part) FreeSCI. On the other hand, if they want to work with us, they are of course welcome to do so as long as their project fits with ours. In fact, that has happened a few times already.

It seems to me that if SCUMMVM supported a game creation tool, even an obscure one, it would attract a lot of developers to make new games,

There does seem to be development tools for the AGI engine, though I haven't used any of them myself. Of course, that's probably not the most exciting adventure game engine around.

Wow! My immediate thought was "what if Sludge was adopted by ScummVM - a match made in heaven!" Like the song says, Sludge and Scumm.... Scumm and Sludge... go together like a horse and carriage..."

Right now, it would probably be a pretty bumpy ride. From what I understand, Sludge uses 16-bit graphics while ScummVM assumes all games use 256 colours. It's possible that someone will eventually implement support for 16-bit graphics (particularly since there are a few SCUMM games that use it), but there's no telling if or when that will happen.

Do the ScummVM folks really want to attract new game developers? Am I wrong to think that an interest in AGS could transfer to an interest in other engines? Was the interest in AGS blown out of proportion?

Personally, I have no strong feelings for or against. We're not in any danger of running out of old games that people think we should support. On the other hand, it's nice that people are still writing adventure games.

[Chris_Tolworthy]

Thanks for the reply.

Offhand I can think of several reasons why it might make sense for ScummVM to include game creation abilities.

1. It might make it easier to port new games across. If I understand correctly, every time you support another old game, the engine has to become even more complex, with more fixes and hacks. At some point it might save time to have a standardised way of creating a game, then all you need to do is write a new clean game that looks like the old one and uses its resources.

2. It might attract new talent to the ScummVM project. Some of the people who make games would also enjoy tinkering with the engine.

3. It would raise the profile of ScummVM by bringing it to a much larger user base, and this in turn would attract more developers.

4. I suspect there is a large overlap between people who love old adventure games and people who care about adventure games in general. Some of the existing supporters probably want it.

Just thinking aloud. As a newbie I could be totally wrong.

Every now and then, someone will suggest that we support AGS games. I've always found this rather puzzling, since at least some of these suggestions seem to come from people who assume that we'd be able to do a better job without any prior knowledge of AGS (not to mention the source code) than Chris, who's been maintaining it for the past decade.

This would be an argument in favor of Sludge, as it is no longer maintained.

From what I understand, Sludge uses 16-bit graphics while ScummVM assumes all games use 256 colours. It's possible that someone will eventually implement support for 16-bit graphics (particularly since there are a few SCUMM games that use it), but there's no telling if or when that will happen.

Might that make your job easier? If it's going to happen anyway, why not do it in a way that adds huge extra benefits?

Of course it's easy for me to say this. I'm sitting in my armchair suggesting that other people do a lot of work to make life more pleasant for me.

[LordHoto]

1. It might make it easier to port new games across. If I understand correctly, every time you support another old game, the engine has to become even more complex, with more fixes and hacks. At some point it might save time to have a standardised way of creating a game, then all you need to do is write a new clean game that looks like the old one and uses its resources.

I don't quite understand what you mean by "the engine", but ScummVM has support for different engines and not just one big engine which handles all games. Since ScummVM is (originally) meant to support 'old' games, I wonder why this would be an benefit to ScummVM in case of 'old' games, but maybe I just don't get your point?

2. It might attract new talent to the ScummVM project. Some of the people who make games would also enjoy tinkering with the engine.

Well ScummVM is more than just "the engine", as I said for 1., we support different engines as you can see here.
So I wonder what use it would be for the other engines to get new people working on a 'new' engine.

[Arantor]

The one real benefit I can see for having game creation tools for ScummVM is that the game's author instantly has access to a multiplatform distribution.

AGS is pretty cool but most of the extra features are written as plugins which aren't supported outside of Windows.

ScummVM, however, is available for a multitude of ports - you may even start to see some of the AGS authors move to ScummVM if it had an authoring tool, which can only boost the user base.

[sev]

Let me express my own vision.

First, do not expect that some of existing ScummVM developers will perform the porting task. Rather, you should try to find an interested party elsewhere. Of course, those new folks will get answers to all their tech questions.

Second, it could be really a good thing to have some state of the art engine with an authoring tool, but speaking of Sludge, someone has to make the editor multiplatform as well, otherwise you will lower attractiveness of this solution.

AGI Studio is nice, but AGI engine itself is rather primitive.

As of 16 bpp support, adding that to OSystem will be quite trivial, as we in fact already render GUI on 16-bits surface. Of course, some platform do not support it such as.... well it looks like all current platforms do support 16bits gfx .


Eugene

[Chris_Tolworthy]

I don't quite understand what you mean by "the engine"

I just mean the common aspects of ScummVM. Unless ScummVM is an unrelated collection of hundreds of programs (one per game per platform) then there must be a common way to do stuff.

I wonder why this would be an benefit to ScummVM in case of 'old' games

Because the process of adding an "old" game sounds a lot like the process of creating a new game: create a plan of how the game works (or should work), identify the resources, then script it.

When an old game is particularly complicated, it might save you guys time to simply create a new game that looks and plays the same.

[clem]

Because the process of adding an "old" game sounds a lot like the process of creating a new game: create a plan of how the game works (or should work), identify the resources, then script it.

No, this is not how it works. The scripts are (mostly) in the game data files, and are just ran by ScummVM. See also http://wiki.scummvm.org/index.php/About

[Chris_Tolworthy]

No, this is not how it works. The scripts are (mostly) in the game data files, and are just ran by ScummVM. See also http://wiki.scummvm.org/index.php/About

[edit]
Let's see if I understand. Imagine if the Sludge game Out Of Order was added to the ScummVM list of games, then ScummVM would gain the ability to read Out Of Order's script files, right?

Then anyone could then write another Sludge game that used the same scripts and commands, so it looked to ScummVM like just another variant of Out Of Order?

Bingo! If that happened we would have the Holy Grail of game development: a way to write games on one platform and have them work on all major platforms.

Technically it sounds even simpler than I thought.

[Chris_Tolworthy]

speaking of Sludge, someone has to make the editor multiplatform as well, otherwise you will lower attractiveness of this solution.

Ah, but that's the beauty of it: the sprite editor and floor editor are very basic. Most of what Sludge does is through plain text files and of course the compiler. In that respect it's completely different from AGS (for example).

[fingolfin]

And here is *my* personal take on this:

In my eyes, ScummVM is for running old adventures games. Also, I think a cross platform adventure toolkit would be cool. And if somebody would want to create one (incluing the editor and other tools required), and wants to make it open source, and wants to benefit from the ScummVM code base -- then I applaud that and welcome them. Since our code is under the GPL, you are already allowed to use it freely in your own GPL code anyway.

But I don't think it would make all that much sense to then integrate the requied new run-time engine into ScummVM itself. Well, maybe if you want to use our Launcher. But for new games, why burden yourself with the full ScummVM? Why waste memory, esp. on low mem devices, with the rest of it??? So I do not at all see why it would be so attractive for the toolkit maker to use ScummVM as runtime engine. Although I do see why he might want to benefit from our experience.

On the other hand, I don't see either why ScummVM would benefit from this so much. Yeah, yet another game system supported. But "gaining marketshare" is not really something we are interested in. We are a hobby open source project, not a company trying to "sell" something. So why "raise our profile" ?
The only reason would indeed be to "attract more experienced and capable developers". But I am highly skeptical of the success of that just because of adding support for an "adventure toolkit".

All in all, I just don't buy that this is as great an idea as you seem to think. So my answer you original question "is SCUMMVM looking for an source game creation tool?" would be: Nope . However, ScummVM is licensed under the GPL, so everybody is of course allowed to write engine plugins for it.

[Chris_Tolworthy]

Fair enough. I guess that answers my question. Thanks for all the feedback.

[Chris_Tolworthy]

Just in case anyone was interested, OpenSludge should be on SourceForge any day now. Sludge already has some award winning Windows games under its belt. The first open source port is likely to be for the Mac. Happy New Year everyone.

[rivadolmo]

...At the moment it seems AGI is the only creation tool avaible...I feel that the possibility to increase the creation possibilities of SCUMMVM may dissapoint someone in the Game Industry....anyway I'm trying to develop something interesting with AGI (time is a critical resource)....

[Chris_Tolworthy]

I feel that the possibility to increase the creation possibilities of SCUMMVM may dissapoint someone in the Game Industry

Yes, I think this is probably the true underlying reason. ScummVM walks a tightrope of not offending the industry, and it works, so why rock the boat?

I'm trying to develop something interesting with AGI (time is a critical resource)....

Keep up the good work! Most people like me just talk about it. But you're actually doing it.

Ona related noter, I am very interested in what was said about ScummVM simply interpreting existing commands contained within each game. I wonder. Does this mean that, if I found the script format for a supported game, and wrote a new game that only used those existing commands, then the new game would automatically work in existing versions of ScummVM?