Modding Support - When?

[newageofpower]

IIRC, Styg stated that he was planning on supporting modding.

When will this be added? I understand that you may need a break after the hectic days of post-release patching.

[phobos2077]

I think he mentioned this in an interview or on forums. For some reason they don't want to release the tools they used to make the game.

As an experienced modder, I can say that ANY tools (especially those used to make the game in the first place) are always good thing for the community. I never understand what is wrong with releasing them "as is", even without any official manual/support, etc. People will figure everything out. By ignoring modders, you just force them to do it the hard way, reverse engineering file formats, game executable, etc. - which only slows things down for no good reason.

It is very important to release such tools as soon as possible, while the game is at it's peak of popularity (like first 3-6 months after release). inXile don't give a f**k about community so they just ignore mod tools (over a year passed since release - no tools, not much point in releasing them at this point - interest in the game is lost).

Please, Stygian, DO NOT make same mistake!

[player1]

The problem with releasing any kind of game creation tools that are not intended for community is that they are very unpolished.

Bad interface, implied functionality, bugs with known workarounds inside the team etc...

This is because they are not intended for general use, outside of project, so any polish is waste of time, when same time can be invested in game itself.  If there are any blocking issues, workarounds are provided, or quick code fixes.

If such tools are released in public, it would imply some sort of support, which as trade-off means less effort being done for bug-fixes, for example. And if no support is given, it would reflect to the game impression itself in community (damned if you do, damned if you don't).

Even more, since whole internal development team is Serbian, I would not be surprised if creation tools are in Serbian and not in English.

.

On the other hand, it would be great if any kind of tools that are used for asset packing or conversion become available to the public, and then let modders think about ways how to use that unpacked data to create content.

Still, even for this kinds of tools, there are limits. If any kind of licensing format is used, usually that means that such conversion/packing tools are also licensed and can not be given for free.

.

And sometimes, effort for releasing tools is not worth it. For example small minority advocates for modding tools, they get released after some investment by game team (and not working on something else) and then nobody does anything decent with those tools. These things are known to happen.

[phobos2077]

That again... I completely disagree. You are talking like some project manager and not a community modder. Have you ever tried making mods for games that have proprietary formats and no official tools? If you did, you would've said differently.

Just give us ANY tools, no need for any polish or support!!! Why is it so hard to understand?

Unpolished, bugged? - So what? We'll get around limitations. Still better than reverse-engineering everything and building tools from scratch when such tools already exist - hidden somewhere.
Tools in Serbian? - We'll translate them! We'll learn Serbian, for crying out loud...
Somebody will complain about quality of those tools? - Ignore them, at least the modding community will live and evolve.

I see no point in holding those off. No tools is always worse than bad tools.

[Grim]

Just a little additionnal insight :
Giving tools to the public can be seen as giving away your tech, too.
Styg made this engine from scratch, i'd understand if he doesn't want it to be stolen and used by others, for exemple.
By publishing tools, you don't provide them only to modders with good intentions, but to the "world", with all its good and bad.

[ShadowRun]

As a keen modder for Homeworld, back in the day, I have to agree with the tool advocate(s). It's what keeps the longevity. Once everyone has finished the core storyline and a few alternative endings, they hanker for more. If mods are available, they'll play them and then get inspired to make their own mod. If mod support is not available, they'll switch to another game, and uninstall Underrail six months later with fond memories. Sad but true.

The problem with releasing any kind of game creation tools that are not intended for community is that they are very unpolished.

...

Even more, since whole internal development team is Serbian, I would not be surprised if creation tools are in Serbian and not in English.

...

If any kind of licensing format is used, usually that means that such conversion/packing tools are also licensed and can not be given for free.

The thing is that security through obscurity does not work. If it's worthwhile persisting and reverse engineering, it will be done. We already know that the compression tools are free. As for the other objections, if Stygian put the development tools up on Sourceforge or Github, the modders themselves would discover, understand and fix bugs, polish the interface, provide necessary translations - anything to make our lives easier when we're tinkering and crafting our own dungeons and denizens.

In fact, modders and community support will even reduce dev effort by doing their work for them, for free when possible, just out of sheer appreciation of the game:

A lot of the dialogue and text in the game have some pretty serious grammatical and spelling errors. Things like "your" instead of "you're," or just some run-on or unclear sentences. (...) this type of thing does go a long way to break illusion and immersion.

(...) What I would suggest then, is that you open the dialog up at some point to a few members of the community that can help clean up some of the dialog after you've written it. And I'm sure you could find a few willing to help.

Indeed, many people have expressed concern over this and offered to help, but so far all we've got is a sticky in the bug forum to report typos & grammar errors. It works but is an annoyingly cumbersome method, considering that there are literally thousands of typos spread all over the place. Having access to plaintext dialog files would be so much better, even if they contain scripts and whatnot. But I guess that's not gonna happen.

If the dialogue files were also up in github, the community could simply submit commits that fix those and when a new build is about to go out, Styg could pull and review the changes to make sure they are all alright.

And sometimes, effort for releasing tools is not worth it. For example small minority advocates for modding tools, they get released after some investment by game team (and not working on something else) and then nobody does anything decent with those tools. These things are known to happen.

Just a little additionnal insight :
Giving tools to the public can be seen as giving away your tech, too.
Styg made this engine from scratch, i'd understand if he doesn't want it to be stolen and used by others,

As far as these arguments go, just consider what would have happened if Warcraft III modding had been deemed "not worth it" - there would be no DOTA mod developed. Look what that has done for the game engine and the community. Still thriving a decade after every other game of its generation bit the dust.

What would have happened if Halflife modding had been deemed "not worth it" - there would be no Counterstrike mod developed. Look what that has done for the game engine and the community. Still thriving fifteen years after every other game of its generation bit the dust.

Would it really be so bad if, for example, Styg's engine becomes famous as being the ideal fast and easy map-creation tool for tabletop roleplay and wargaming? If it got such a reputation for its free tools, then that is vast exposure to its target market:- how many of them do you think would be delighted to while away their late night hours crawling through the Underrail passageways that made all this possible, and checking out the vast library of community generated content?

Rather than asking if Stygian can afford the risk of exposing their dev and modding tools to the public, I think the question is more, whether they are able to afford not to.

[ShadowRun]

By the way, modders have already made some tools; general-purpose Underrail unpacker/packer and .udlg extractor/injector. Standard XNA tools can be used to modify all the graphical assets.

Aww man... - this is what I get for taking an hour to write my posts. As is often the case, epeli conveyed what I'd been thinking.

[player1]

You are talking like some project manager and not a community modder.

Exactly. That is the point.
You raised arguments from the modders side of view, and I those that are on the other side.

[Arclight]

And don't forget Jagged Alliance.... BearPaw and the other modders did wonderful work on a game 10+ years old, and still going strong.

[MichaelBurge]

The game files are mostly serialized versions of the .NET classes in the game executable. The class names change every time the obfuscator is run over it, but they have tags(attributes) that remain constant. I would estimate it's a week's worth of moderately intensive work to map out the meaning of the tags, parse in the game files, maybe do some simple manipulations(like changing dialogs), and writing out the changed game files. It would be limited to whatever the game engine makes easy.

It's mostly useful for dialogs and inventory: There are many many custom .NET classes embedded in the executable that implement behavior. For example, the NPC that tells you that Abram wants you has a custom class embedded in the executable for him. It would be significantly harder to change these or add new NPCs that require custom logic: It's not really a casual modding effort to disassemble the game with ildasm, add in a class for a custom NPC, and then reassemble with ilasm.  I did something similar when I wanted to change the color of my energy shield to orange. With the obfuscated source, you'd have to change the class you inherit from every time a new patch gets released, since I think the obfuscation process isn't deterministic. Or redo your changes against the new names if you edited an existing class. It's doable in a stable manner if you had the unobfuscated binary, but I can see them equating that with giving away the engine source.

I can see that if there isn't a clear separation between the user-facing dialogs, inventories, NPCs, etc.  and the game engine, that Styg might not want to release the tools because any realistic mod that someone would want to make would require custom .NET code that is annoying to support injecting into his obfuscated engine. Any new items or equipment would require this, as an example.

I don't know though: Even without developer intervention, modders could probably distribute mods as a zip containing patch files for the main game and support files for the dialogs and such. You(or a program on your behalf) would disassemble the game, use the associated tags to map the class names in the disassembly to stable equivalents, apply a patch to load in any additional dialogs or other support files, apply the patch files for new classes(such as items), and then reassemble. The patch files could be created by writing a C# class(the language doesn't matter), compiling to an assembly file, and then disassembling the assembly file. This scheme should be resistant to small bug-fix patches breaking the mod. I don't know enough about the engine to know whether it would be resistant to larger changes, such as if some new area were released or something.

[ShadowRun]

We can be confident the GUI will be using patterns such as MVC because Spaghetti code is not cost-effective. It would be enough to just expose that API and keep everything behind it a black box. That way compliant modules could plug into that API such as a class that dumps rendered maps to file rather than the current one which displays them on screen.

Likewise there will be a bunch of resource files (likely in XML) that must conform to a format (a schema). They get loaded in across another API. Plenty of games have utilised an "override" path allowing a customised datafile subset to be overlaid ontop of the official set. As long as this API were released (essentially all the XML schemata for these datafiles) the rest of the game engine could remain a black box. Modders could add an entry to NPC.xml with a custom GUID; some entries to dialogue.xml with what (s)he says; etc. etc. and not (need to) know how the engine's secret sauce operates - it is enough to know merely how to alter its ingredients list without breaking the mixing bowl.

[Sanger]

I've used MichaelBurge's unpacker to take a look at a few of the loose game files out of curiosity and they're hard to make sense of. Do NPCs even have levels?

Edit: nevermind, I got it. Incidentally, Carnifex is level 20 and has 16 strength and constitution, 12 dexterity and agility, 10 perception and intelligence, and 14 will.

In fact almost all of the NPCs in this game have crazy high stats, geez!

[newageofpower]

I've used MichaelBurge's unpacker to take a look at a few of the loose game files out of curiosity and they're hard to make sense of. Do NPCs even have levels?

Edit: nevermind, I got it. Incidentally, Carnifex is level 20 and has 16 strength and constitution, 12 dexterity and agility, 10 perception and intelligence, and 14 will.

In fact almost all of the NPCs in this game have crazy high stats, geez!

Tanner....

[Sanger]

Gorsky is pretty much a tiny god with 71 total ability points. Bisson has 83 total ability points! The hell.

[Ninjaxenomorph]

I'd love to help, but my programming skills are zilch, and translation skills limited to "What American accent would you like to convey?". Just a player that would love to get his hands on mods. I can write, though, and I could probably make a few maps, and I'm full of ideas.