SCI Games

[Illuminerdi]

FreeSCI is about as active as a comatose slug, it's taken them years just to get SCI0 games working, I doubt we'll see any of the SCI1 (VGA) games anytime before 2010 at this rate. I know it's not easy, and these people are donating time, but I've been watching FreeSCI for almost half a decade, and it's just crawled along.

Meanwhile, ScummVM has perfected Scumm and moved on to multiple other engines in around the same time period. ScummVM has an active community that could probably make it happen in a much faster timeframe. Perhaps even a merger of the two projects, depending on where FreeSCI is at currently?

More to the point, why not merge them? If I were a FreeSCI dev, I'd be banging on ScummVM's front door to let me in - I see no reason to have two separate programs with the same goal. Just doesn't make any sense.

Am I the only one who'd like to see this happen? Perhaps a ScummVM dev/admin should reach out to FreeSCI and see if this interests them at all?

[clone2727]

Perhaps you could use the search feature of the forums?

[PsYcO]

no. as for the 'goal' argument, it simply doesn't apply since scummvm's initial goal was to support lucas art games, recently this has changed, more because i feel the devs run out of things to do.

they've simply done an excellent job, and fast.

[bramvandijk]

It is experimental, but there is not only SCI1 support, but also SCI1.1 support

[md5]

It's not an issue of "goal". The REAL issue that this is not as straightforward as some people want to make it is that ScummVM and FreeSCI code games in a different manner and follow a different philosophy (i.e. loads of things tend to be rewritten from scratch in FreeSCI). Do a search in the forums and you'll find out this question has been answered in the past.

Loads of people (including me) would like to see a merge of those two projects, but there are some fundamental things which are different, plus sometimes merging two projects doesn't really make any difference at all to their progress. The people who are most familiar with SCI are the FreeSCI programmers, and I doubt that a merge with ScummVM can make things go faster for any reason. Sometimes, developers who are interested in a certain engine appear out of the blue, but this is a rare occasion

[clem]

Also there seems to be the underlying impression that if there was work done on SCI (or any other engine) in ScummVM a team of 20 people or something would jump on it to make it real, just because ScummVM is a "big" project. Looking at the credits shows that most engines are maintained by 1-4 people however.

[Kaminari]

More to the point, why not merge them?

/me slaps myself on the forehead.

Of course! Brilliant suggestion. And to think that nobody ever thought of this.

[Illuminerdi]

Every thread I found on the subject seemed to end rather abruptly, and consisted primarily of someone saying "hey when is ScummVM going to support SCI" and the responses were always "Go to FreeSCI, and go away".

I found little actual discussion on the topic of how slow FreeSCI's progress was, or the idea of merging the code - hence the reason I brought it up. FreeSCI has seen very little in the way of updates in over a year - sure they've added beta SCI1.1 support, nevermind that SCI1 still barely works. In that same timeframe, ScummVM has added full or partial support for 3 engines without source code, with a fourth on the way. Meanwhile, FreeSCI is showing all the signs of a project slowly going through its death throes.

Plus nobody's even scratched the topic of portability - FreeSCI is primarily Linux based, and the Win/Mac ports get updated when the porters feel like it - to say nothing of the near ubiquitous porting of ScummVM, which is currently officially and unofficially ported to what - 15+ platforms?

Yes, I've read the FAQ, yes I know that engine support is entirely at the whim of devs, and yes, I know that this is probably a dead horse. It just seems silly to keep dodging the issue. SCI1-2 are the most major engines that ScummVM doesn't support. The fact that this topic keeps coming up repeatedly means that there's a devoted fanbase of people who want to see better SCI support on modern systems, and IMO FreeSCI is not the answer to that, and never will be.

[clone2727]

You could always code it yourself?

[Illuminerdi]

I might just do that if it doesn't happen eventually, alas my coding skills right now are...limited at best. I can write you some wicked bucket sort algorithms!

If ScummVM doesn't have SCI support in 3-ish years when I finish my degree (and hopefully have better coding skills since...well...that's what I'm paying the school for), then yeah, I might just do that.

[criezy]

I found little actual discussion on the topic of how slow FreeSCI's progress was, or the idea of merging the code

There is this thread in which the question is raised and Lars Skovlund, a (the?) FreeSCI developer gives a reason why he is not favorable to a merge with ScummVM.

[Collector]

As I have said in other threads, I was disheartened with how slow FreeSCI development is and that there is an almost chauvinistic disregard for any OS but Linux, but these are things best taken up with the FreeSCI people. I am no longer terribly concerned about the fate of FreeSCI since DOSBox has improved so much. There are very few SCI games that don't run easily in DOSBox and most of those are Windows only games.

I don't mean this as a slam against the FreeSCI devs, but that is the state of things and Sierra fans have many other viable options. I would certainly welcome a ScummVM/FreeSCI merger or faster, more portable development of FreeSCI, but it is becoming increasingly moot and the time and effort of the ScummVM team are probably better spent on other things.

In the mean time, just use DOSBox for SCI games or any DOS game not supported by ScummVM. It has become more compatible with a greater range of games and easier to use than a real DOS machine.

[Noelemahc]

In the mean time, just use DOSBox for SCI games or any DOS game not supported by ScummVM. It has become more compatible with a greater range of games and easier to use than a real DOS machine.

Us Symbian freaks would die to get Quest for Glory in a mobile form. I know I would, since coding isn't something I'm capable of. Alas, ScummVM is one of the few, if not the only one, actually, genuinely cross-platform emulators/interpreters out there to posess an officially and centrally maintained repository of non-PC ports.

This is to say, I would love to see SCI support in ScummVM, but if the devs say 'no', then no it shall be. Us non-coders have to learn to suck it up every once in a while because all the useful stuff that we can contribute are thanks and bug reports.

[sev]

no. as for the 'goal' argument, it simply doesn't apply since scummvm's initial goal was to support lucas art games, recently this has changed, more because i feel the devs run out of things to do.

they've simply done an excellent job, and fast.

FYI. First non-SCUMM engine was added to ScummVM in 2002, with ScummVM 0.2.0. That was AGOS engine used for Simon the Sorcerer 1, 2 and other games. So it's not recently changed.

Also we still not supporting every SCUMM game, i.e. there are several notable ports like Apple II or PC Engline which we were unable to support as of today. So it's not that we run out of things to do either.


Eugene

[sev]

I would certainly welcome a ScummVM/FreeSCI merger or faster, more portable development of FreeSCI, but it is becoming increasingly moot and the time and effort of the ScummVM team are probably better spent on other things.

I am sure that lots of people (me included) will definitely welcome such merge, but.

• Who will do the merge? FreeSCI devs are not up to it
• It will not help automagically SCI development. Basically same people will work on it, so no speedup
• Codebase is too different, so FreeSCI devs have to be convinced that the efforts on studying new codebase will bring them benefits

So, any points on those questions?


Eugene