How to save Wesnoth

[Sir_Cryer]

@Chaosrider
U right man. I'm wrong and eveything i stated here is false and a lie. Wesnoth is not so popular anymore because players are growing and doing other stuff like u said. Wesnoth dont needs any stupid opinion like mine, a stupid common player that knows nothing about making addons and making some basic programming or lua or whatever. Sorry man, i will just shut up and stop disturbing you.

[Pentarctagon]

Alright people, calm down...

[name]

If the sole purpose of this thread is for random people to throw out their personal agendas for wesnoth under the assumption that this will somehow save it... then I want to play too. Project salvation can achieved by fixing two problems with the mainline campaigns:


Problem 1: Campaign scenarios plagued by turn limit failure conditions.

A byproduct of the gold/unit carryover system is that it must be paired always with a hard turn limit to preserve balance of future scenarios.
Thus almost every campaign scenario is reduced to trying to beat the clock, rather than a powerful or intelligent opponent.
Thus almost every campaign story line is reduced to heroes running away from battle, scenario after scenario, up to or including the finale.

Solution 1: More carryover pauses, resets and turn limit victory conditions.

Have more scenarios where the player leader is separated from its carryover-able army and purse, either temporarily or permanently. The leader could be washed overboard from the ship carrying the army and loot, reassigned a new force of soldiers and budget by a superior, forced to travel alone through the spirit world to reach a far away land, etc.

Turn limit victory conditions, or survival scenarios, also mix things up, but mainline campaigns have only a handful of these.

Solution 2: Replace carryover with more common bonus objective rewards.

Save the field marshal and be able to recruit his knights in the next scenario. Steal the dragon's 200 gold pieces and start the next scenario with said extra gold. Have a leader pick up an artefact and it has a new ability for the rest of the campaign.

All stuff done in campaigns already but put several of them at once into almost every scenario, and disable the carryover mechanic completely.

Problem 2: Campaign characters, places and events not well interwoven.

I would say wesnoth lacks the inter-campaign consistency and world building of other games like the original starcraft. It is rare that the same character occurs in even two campaigns and even then has little in the way of recognizable personality consistency. Places like Weldyn change dramatically in size, shape and surroundings from one campaign to the next. Major events witnessed or participated in during one campaign never or almost never occur in any other campaign directly.

Solution 1: Clump campaigns closer together on the world history timeline.

If more characters are contemporary you can develop the characters over more situations before the end of their natural lifespans. If more events are contemporary, they can have greater and more surprising influences on each other.

Solution 2: Reuse a scenario map across campaigns when it depicts the same location.

Solution 3: Have campaigns converge at major events.

For just one example, since multiple campaigns depict freedom fighters against the queen and her orcish mercenaries, have these folks at various times encounter and talk to each other or fight for or against each other, for at least a scenario or two.

[cephalo]

I have an another opinion on what challenges the future of Wesnoth. The game mechanics themselves are fine and WML is just an awesome vehicle for story telling that will never get old.

The graphical style, as good as it is, is just not how games are made anymore. Pixel art is cute, and some examples are definitely in the realm of high art, but you won't find a large pool of people who can improve upon what has already been done in Wesnoth. The kids are learning 3D these days, and there are many advantages to that approach, including the use of well established game engines that handle a lot of the added complexity for you, and the fact that 3D animation can be much more man-hour efficient than fluid 2D animation.

I don't know if any of you saw my dragons in Library of Kratemaqht, but I basically cheated on the pixel art requirement by making an animated 3D dragon and pixelizing each frame with an application I wrote. That was the only way such an animation, with 400+ frames, was possible I think, and in any case, I received some praise for it in spite of the fact it's not pixel art.

Young people get involved in game modding (basically what UMC is) because they want to learn skills that they might use in the future. Wesnoth is not a big draw in that regard right now.

If Wesnoth went with a fully 3D, fully textured game map with different hex altitudes and 3D units, I think it could possible become a spectacular thing over time and draw more people. Can I do this? No, I can't, but maybe some of you can.

[iceiceice]

I have an another opinion on what challenges the future of Wesnoth. The game mechanics themselves are fine and WML is just an awesome vehicle for story telling that will never get old.

I fully one-hundred percent agree with this.

Young people get involved in game modding (basically what UMC is) because they want to learn skills that they might use in the future. Wesnoth is not a big draw in that regard right now.

So I guess my take on it is, the draw of wesnoth is supposed to be that it's way easier to get started modding it, it's just way more accessible than trying to use Unity or something like that. Wesnoth got big I think in the first place because it occupies this niche where, you can do very inventive and imaginative things with it, (with limits), it provides some powerful built-in stuff like the terrain system and the animation system, and it's still not too hard to learn to do all that. You get a lot of "bang for your buck" as a game modder and story teller if you decide to spend some time to learn WML.

I'm sure that many people will disagree with this, but at least in my mind, the "heart" of wesnoth is that it's a fully FOSS game engine that's got great game play, that's very accessible and very modder friendly. The pixel art aspect of it is just a means to an end -- pixel art is good because it's very accessible and easy to understand, even kids can figure out how they should make sprites and animation sequences, and they don't need to use extremely complex tools to make that stuff (if they don't want to). Jetrel, our lead art director, did not begin as an artist, he is a professional programmer, but he was able to pick up pixel art and with hard work refine his skills to the point where he's now a quite skilled pixel artist.

Making 3D art is quite difficult and represents a huge time investment. At least, the last time I checked -- my knowledge of how hard it is to make 3D art is probably like 5-10 years out of date, but somehow I doubt that it has changed that much. My expectation is that you need years of practice making 3D art even to make something passable.

If there's some way to simplify it or make it more accessible, that would be one thing, but that in my mind is the main argument against trying to redo wesnoth as a 3D game. This will also be controversial, but I don't actually see a strong reason against allowing users to have both sprite based art and 3D based art if they *want* to use 3D models for their units. But the current game engine is really not going to be capable of rendering 3D models.

[cephalo]

Making 3D art is quite difficult and represents a huge time investment. At least, the last time I checked -- my knowledge of how hard it is to make 3D art is probably like 5-10 years out of date, but somehow I doubt that it has changed that much. My expectation is that you need years of practice making 3D art even to make something passable.

I would say it's more like a trade off. It's much quicker to draw a single 2d base frame than creating and rigging a 3D character. However once your 3d char is all ready to go, animating proceeds extremely quickly and accurately, without having to worry about how to foreshorten that swinging polearm or whatever. In 2d, after your base frame is done, you still have to draw 50 more frames somehow!

My dragons were my very first 3d creations, while there was a lot of luck involved in the results since I'm not an artist, they only took about 4 months to complete. I used a program called Art of Illusion because I couldn't even figure out how to use Blender. That's 1200 frames worth of work from a complete novice!

Another factor that is important I think, is that learning this skill can lead to bigger things in games and movies for the talented artist, or even 3D manufacturing and printing for the not so talented artist. So even if it's hard to learn, there's a potential career benefit. Learning to make pixel art animated sprites for Wesnoth is kind of a dead end endeavor in 2015 I think the new kid stumbling upon Wesnoth for the first time knows that.

I realize 3D Wesnoth is kind of a 'pie in the sky' suggestion, but having that necromancers cave in the corner of the map be actually under the mountain soaring above is... Yes. I want that.

[GunChleoc]

The way graphics work in the Widelands game is that the artists create 3D art in Blender, but they then render individual frames as png files to be used by the game engine - pretty much like cephalo did, minus making it look like pixel art. This might also be a possibility at least for UMC for those who prefer to work in 3D?

[iceiceice]

I realize 3D Wesnoth is kind of a 'pie in the sky' suggestion, but having that necromancers cave in the corner of the map be actually under the mountain soaring above is... Yes. I want that.

So it depends on exactly what it is you are suggesting here. If what you want is like, fully 3D procedurally generated terrain graphics, something like this:

http://www.world-machine.com/

I mean sure we could do it but it would be starting over from scratch and it would be pretty hard. It would mean like, completely new art assets.

If you mean like "Iso 2D" graphics like was suggested in this thread:

http://forums.wesnoth.org/viewtopic.php ... 2d#p577265

It's more plausible but it still requires a large amount of the terrain graphics to be redone. The terrain art is one of the most significant bodies of work accumulated in the whole project. So you'd need the commitment of artists able to do it to the same level of quality, which is not a small task.

Also it will be controversial because it involves replacing a lot of art we have right now that is quite good.

It's also, a lot of work that doesn't really enhance game play. If you want to create panoramic views, you can already use story screens for that. But when you are actually trying to figure out how to invade the necromancers' cave, seeing the top of the mountain above him mostly just gets in your way I guess. We'd need to work out some significant UI aspects of this.

[cephalo]

It's such a drastic change that not only art would have to replaced, but each campaign would have to be retrofitted to the new rules that would likely be necessary. This would take decades.

For historical purposes maybe it would be best to support 2d as is, so that all the content continues to be presented as it was intended. Then develop 3D on the side for a good long time, with the intention that it would be for newer, post-2d content. The old content is still very enjoyable to new players, the issue is more how to encourage new content creation.

In fact, before any 3D graphics work is started, maybe take the game rules to 3D first with layers of 2D map levels, like Dwarf Fortress or Age of Wonders. Just to get a feel for what the consequences are for the mechanics, and to verify that such a transition is even possible while remaining the same game.

[Jetrel]

After the nice things iceiceice said above, I thought I'd chime in.

I don't know if any of you saw my dragons in Library of Kratemaqht, but I basically cheated on the pixel art requirement by making an animated 3D dragon and pixelizing each frame with an application I wrote. That was the only way such an animation, with 400+ frames, was possible I think, and in any case, I received some praise for it in spite of the fact it's not pixel art.

Well, that's a screamingly good argument in favor of using 2d sprites as the base style for the game. You didn't have any 2d animation skills, and yet your use of 3d made 2d sprites for your mod just fine.

One could argue against the use of pixel-art as the particular style we went for, but ... a game has to have some consistent style, and pixel art isn't any particularly harder to "match up with another's work on" than any other style (in fact quite a lot of folks argue it's easier). It's really hard to match someone else's style no matter what medium you use - 3d is absolutely no exception (it's often very obvious in less well-produced games when certain 3d models got done by different people). It's really hard to be consistent, period; we just picked a style that was relatively straightforward for people to be consistent in.

Especially when there are probably fairly decent "faux pixel art" shaders people have written at this point, if you're making UMC stuff and want it to fit in with the game.

Young people get involved in game modding (basically what UMC is) because they want to learn skills that they might use in the future. Wesnoth is not a big draw in that regard right now.

You know, this is an extreme projection of your own world-view onto others. Anecdotally, I can say this is false - certainly at least citing myself. Even when I very first began, the art style used by wesnoth was already no longer a good fit for commercial art creation skills, and I began nearly at the start of the project. I knew that the art I was learning to make was not a hirable skill.

Here's the crazy idea: the making itself is the primary end. Not making as a means to a future career, just making itself. It's a hobby.


In fact a lot of times people mod because they specifically want to improve the exact game they're playing, rather than get a professional game-making job, which would guarantee they'd be working on some totally different title.

It's such a drastic change that not only art would have to replaced, but each campaign would have to be retrofitted to the new rules that would likely be necessary. This would take decades.

For historical purposes maybe it would be best to support 2d as is, so that all the content continues to be presented as it was intended. Then develop 3D on the side for a good long time, with the intention that it would be for newer, post-2d content. The old content is still very enjoyable to new players, the issue is more how to encourage new content creation.

In fact, before any 3D graphics work is started, maybe take the game rules to 3D first with layers of 2D map levels, like Dwarf Fortress or Age of Wonders. Just to get a feel for what the consequences are for the mechanics, and to verify that such a transition is even possible while remaining the same game.

Yeah, that's the kind of thing where, if we suddenly got a bee in our bonnet and decided that's exactly what we wanted to do to wesnoth, as responsible programmers, we'd still keep any of the code for that strongly forked away from the existing, working code. We'd continue to keep and maintain the existing codebase, only mothballing it as interest dwindled.

Even if we, ourselves, did it, it would for all functional intents and purposes be a separate project with the same name. At which point one can really just open the doors to anyone doing it.

[Jetrel]

I actually agree with this almost entirely; it's actually much harder to pull off than you're suggesting, but unlike most other old-hat type people, I'm not the kind of person who "dismisses ideas like this out of hand".

As you say, right now, almost all scenarios have turn limits, because the current design, without alteration, strictly requires it. By being something we absolutely have to have, on (almost) all levels, it shoehorns all of our scenarios into one formula - which is really a shame because there are lots and lots of other great mission objectives in strategy games, and we don't get access to them. Our turn limits are a fine "rare treat", but being served at every meal is a bit much. It's one reason I've always preferred skirmish maps or MP over single-player campaigns.

As with any such suggestions, I'd say that if one were to change it, the best way to prototype such a change would be to make a new campaign (or a rehash of a current one) which fully puts an alternate playstyle through the paces.



What's the point of the current design:
Our current holistic design more or less requires the turn limit to keep the player from grinding indefinitely (not merely to grind carryover gold, but also to farm up a roster of L2/L3 units). The kneejerk response is that getting too 'fat' by grinding would unbalance the game, but ... hey - why is that bad? Why is grinding into god-mode stats a bad thing?

One principle of game design I've taken a long time to suss out is that there are a lot of actions players can take inside a game that ... really aren't very fun for the player, and players are actually fairly bad at giving themselves a good time by only doing the most fun stuff. It plays into a lot of weird OCD-ish stuff in the human mind, but suffice it to say, as a game creator, you want to carefully steer players away from 'self-destructive' behavior that's going to make the game much less fun for them. Essentially if you're not careful, a significant subset of players will find any hooks in your game that can turn it into a skinner box, and sink into a long boring grind just because they've got a weird quirk that draws them to do that (I know I do, to a decent degree).

Those of a more purely libertarian "I can do whatever I want with my own sandbox" mentality ... may want to try to think of this not so much in terms of being a restriction of rights, so much as being more like a gaming sommelier. De-incentivizing destructive behavior without strictly forbidding it (for example, by allowing grindable stuff to eventually exhaust itself, or something, so you don't just find the first dungeon in an rpg, grind your characters to max level, and have a deeply unsatisfying bulldozer run through the rest of the game). I've done that at least once as a kid, and although I learned not to, I still really appreciate games that willfully avoid hitting that button of mine. It's like not tempting a recovering alcoholic with beer. We're inherently dealing in a medium that is addicting, but the addictive qualities of games are only very loosely coupled with what makes them fun, if at all. I feel that working in such a medium requires a substantial amount of responsibility on the artist's part.


So, I think a design triumph behind the current system that would have to be preserved is the strong anti-grinding aspect the current design achieves. But I don't think "turn limits" are the only way you can do that. I'm not going to toss around suggestions for how (there are plenty of ideas), I just wanted to assert its value. I do think the angles you were playing to are good leads on a solution.

Problem 1: Campaign scenarios plagued by turn limit failure conditions.

A byproduct of the gold/unit carryover system is that it must be paired always with a hard turn limit to preserve balance of future scenarios.
Thus almost every campaign scenario is reduced to trying to beat the clock, rather than a powerful or intelligent opponent.
Thus almost every campaign story line is reduced to heroes running away from battle, scenario after scenario, up to or including the finale.

Solution 1: More carryover pauses, resets and turn limit victory conditions.

Have more scenarios where the player leader is separated from its carryover-able army and purse, either temporarily or permanently. The leader could be washed overboard from the ship carrying the army and loot, reassigned a new force of soldiers and budget by a superior, forced to travel alone through the spirit world to reach a far away land, etc.

Turn limit victory conditions, or survival scenarios, also mix things up, but mainline campaigns have only a handful of these.

Solution 2: Replace carryover with more common bonus objective rewards.

Save the field marshal and be able to recruit his knights in the next scenario. Steal the dragon's 200 gold pieces and start the next scenario with said extra gold. Have a leader pick up an artefact and it has a new ability for the rest of the campaign.

All stuff done in campaigns already but put several of them at once into almost every scenario, and disable the carryover mechanic completely.

[name]

As with any such suggestions, I'd say that if one were to change it, the best way to prototype such a change would be to make a new campaign (or a rehash of a current one) which fully puts an alternate playstyle through the paces.

I feel tempted to try such a thing; specifically the rehash of an existing mainline campaign.

So, I think a design triumph behind the current system that would have to be preserved is the strong anti-grinding aspect the current design achieves. But I don't think "turn limits" are the only way you can do that. I'm not going to toss around suggestions for how (there are plenty of ideas), I just wanted to assert its value. I do think the angles you were playing to are good leads on a solution.

It is hard to imagine a system that more so rewards grinding by curtailing or removing gold carryover and unit recalling. This kind of thing (campaign-long persistent assets) is a black hole for obsessive player tendencies. When you lose a high level unit that could have lasted you the next 20 scenarios, your hand doesn't need any feedback from your brain to start reaching for the save scum load menu. But if that unit could only last you to the end of this scenario at best anyway, it's much easier to just let go of it and play the game as it was meant to be played.

On that note, what if, at least for some entire mainline campaigns (or even all of them) gold carryover and unit recalling were simply removed completely? Leader and story critical "loyal"-traited characters could still persist with their experience and leveling throughout a campaign. But units the player recruited and gold they acquired would be erased every scenario. Would not the benefits outweigh the costs? Do developers or players truly love recalling units more so than they hate turn limits always being the primary challenge of every scenario?

[tekelili]

I considere myself quite authorized in recall and turn limit mechanics, after invest several years thinking how balance a multiplayer campaign I love to play. One thing I learned is that recall mechanic becomes "exhausted" very soon. Imo is impossible control its destructive influence on gameplay further than X scenario, and mainline campaigns greatly surpass that "scenario limit". Imo they should be shorter or totally hack recall mechanic. BfW is not designed to work well when players can recall almost full army: Things like faction balance, healing, upkeep, and AI behavior just becomed too disrupted.

Edit: I dont think developers have to fix recall mechanics, in fact I think root of problem is trying to create a standard. How players acumulate xp and carry that power to next scenarios should be tuned case by case by campaign designers. So my proposal is "break standandarization, it is not helping game."

[name]

Imo is impossible control its destructive influence on gameplay further than X scenario, and mainline campaigns greatly surpass that "scenario limit". Imo they should be shorter or totally hack recall mechanic.

By the way, did you ever find the way to access and delete units from the recall_list using a scenario's wml?

Or can you only set the "allow_recall" attribute to "no" every scenario and pretend the recall list doesn't exist?

[James_The_Invisible]

By the way, did you ever find the way to access and delete units from the recall_list using a scenario's wml?

WML can access units on recall list just the same way as units on the map (but in some cases you need to use search_recall_list=yes). Units on recall list have x,y=recall. So, to delete units from recall list you just use tag [kill] with some filter (unit's id, race, type, etc.).

Or can you only set the "allow_recall" attribute to "no" every scenario and pretend the recall list doesn't exist?

The attribute is called disallow_recall and you would have to set it to yes but see above.