VCMI source not released.
Free Heroes 2 source not released.
XcomApoc source not released.
OpenTomb source not released.
TwinEngine source not released.
Minetest source not released.
OpenMW source not released.
REGoth source not released.
OpenRA source not released.
Dune Legacy source not released.

and here is more.

I have finished polishing VCMI yesterday, it plays as good as original, although I sense weaker AI.
On Linux, native, 1920x1080, with random map generation support, localized.

Once a game is considered finished, its source often gets lost, so the only thing an IP owner remains interested in - is reselling it. In the rare chance of remake, it is usually vastly different to help cash on it.

Means, the only real alternative to have an old game still shining - is reimplementation in (at least) free engine. How far that progresses,.. depends solely on developers. But its still much better in the long run than patching an old closed bucket.

I was talking about your list of games.

You missed the working and faithful part!

Some examples: OpenApoc is definitely not working. OpenRA is a re-imagination – single player missions have to be recreated manually because it uses a different format and everything else (e.g. unit costs/strength) is changed so they have to be heavily rebalanced. Then there are all the Ultima recreations. Last time I ckecked Exult for Ultiima 7 was the most advanced one and it still had a fatal bug plaguing it's item handling.

Okay, I forgot OpenMW, but that does not change the complete picture that much. I've gone though that list on Wikipedia many times – most projects are dead and unfinished or they have very little in common with the original games if the game has any complexity.



Nice.


Obviously. It's too bad that many developers where so careless with the source code, the legal rights are a mess in some cases and that other developer's don't want to release the source code of very old case even if they've got it and there are no legal issues.

We are not talking about recent titles, all those titles are fairly old.
The mathematical model of the game can be copy-pasted. The data manipulation, network, 3d, audio, input stacks exist in are very mature open library forms today. What remains is decoding the data files (graphics, sounds, map resources, scripts) and glueing the stuff together. Boom - you have a from-scratch clone engine that runs on modern hardware/OSes.

I disagree with "untrue/re-imagination" argument because I am not looking to play it in 100%-way like on same hardware/OS stack - but rather have the title work efficiently in modern environment. This is why I don't take "source released, source not released" as an argument or excuse - the games leak very much data anyway, source is not that important. Ethusiasm and dedication are important.

That Wiki article is a pretty great example of my point that game engine recreations without source code is hard. Most of them are in terrible state.

No one's arguing that game engine recreations shouldn't be done btw.
If authenticity isn't important, why not just mod the game assets into another engine entirely rather than waste time recreating the engine of the game in question?

Its because that article tries to include all possible projects including duplicates, successful or not.

All classic console games are emulated, even in official re-releases, even Nintendo or Sega. So that amount of authencity is utopic, And frankly I dont miss my 14" CRT and 486sx. I also don't miss xcom bugs like skill overflows or huge monsters in closets, which openxcom fixed.

Reimplementing on different engine by modding is sure possible. Its all about how accurate it ends up.

Like game engine recreations, they practically never end up accurate enough. And no when I talk about authenticity I'm not talking about bugs not being replicated, I'm talking about gameplay elements being wrong entirely changing the game balance in small, or large ways. One of the mentioned ports, VCMI is a good example. Lots of people liked Heroes of Might and Magic III for its single player. Without an accurately recreated AI, it is not the same game those people liked, at all. Small or large things can be critical to the gameplay experience of many and pretty hard to reverse engineer.

I have no idea why you bring up emulation in this topic. Jazz Jackrabbit is perfectly emulated too.

Its just another bug, nothing more.

Which again is my point, game engine recreations are really hard almost all of them has bugs like that, small and large things that entirely break the intended experience.

No one should expect complex projects like that to be available and polished, they should be happy when they find them, not annoyed when they don't.

You can't easily fix bugs in closed source original engine and you can do this much much easier in reimplementation. In open engine, every problem is a bug. In closed engine, every problem is a problem. See the post #3. For example original Heroes 3 freezes in main menu and supports only 800x600. This is not going anywhere, this is why I use VCMI instead of Wine. This is why you should help OpenJazz instead of using emulation dead end. Even the most hardcore classic lover knows that old hardware will die some day and these old CRT are very unforgiving on eyes.
-- -- --
Also!

1. On OpenJazz website, the author has noticed the GOG release.

2. The lastest commit in OpenJazz is ten days old. This is really worth kicking and a very good chance to get things rolling (better). Please support the project!

I'm not telling people not to support OpenJazz, just to have more realistic expectations of such a project. In an ideal state, a reimplementation is a much better option, but in 99% of cases you're better off emulating the original.

Also, while not largely relevant to the conversation, a lot of closed engine games have been patched pretty extensively without source code. See Dark Souls 60 FPS patch mentioned earlier and Heroes3HD, a Russian fan patch that fixes both the issues you mentioned.

Not quite true. The Christian Whitehead ports of Sonic CD, Sonic 1, and Sonic 2 are recreations based on lots of research into disassemblies of the originals (work which then lead in to Sonic Mania, that's why it feels so much like the classics), and the few 3D Classics titles Nintendo released were similar. the Sonic community is rather exceptional in that there are quite good, well-commented disassemblies though.

I wish there was a place to see the status of OpenJazz. Bonus levels not working right is not a big deal, but bosses being broken is.

The only way I can play this game now is in Widescreen because I can't stand having the fast Jazz inch his way across the levels just in case there's an enemy out of sight. The closed-in FOV just doesn't suit the speed of the player character here.

First of all let it be considered as public domain already not just only open-source. Another thing let someone use Bermuda Triangle,Bridgewater Triangle,Bennington Triangle,Project Stargate,Project Montauk or even Philadelphia Experiment so I would or someone else would help finding lost prototype build of Sonic 1 for Yuji Naka. Maybe someone could use a Time Machine to even find for instance lost source code of Jazz Jackrabbit 2 in order to make Amiga PC-esque 2D cartoony fanmade,unofficial cutscenes that kind of standards in both Sega/Mega CD and Philipps CD-i.
I would use such time travel paranormal activities to travel backwards and forwards in time to the past,present and future literally,seriously maybe me I could do that or maybe even someone else. I would keep companies like Sega for instance to be still considered as first party nowadays rather than being third party…etc.