Project direction ideas moving forward

[doofus-01]

I don't think that's the problem. The thing that holds back the transition to another engine is our engine itself, with its complexity and with the lack of a suitable replacement (the Godot effort, sadly, didn't bring us Wesnoth 2).

I think the terrain graphics problem is that you want to figure out a way to use the existing assets for similar output with a new engine? If so, then figuring out what the existing macros/configurations do and what actually matters sounds significant.

Speaking of terrain macros, one of the many things I'm trying to do is getting rid of all the spurious warnings issued by wmlscope. I've been able to fix the warnings about cross-references and unresolved references, however I'm not really sure about how to proceed to fix the mismatched references warnings (which, more or less, are checks for incorrect data types in macros).
I guess I'll have to open another thread to discuss it in detail, however (since terrain macros throw a lot of these) has anyone ever used these mismatched reference warnings to fix something? Or should I just remove them entirely?

I'd say just remove them.

EDIT:

Let's face it, most gaming nowadays is done on consoles or smartphones. Which means that, to get somewhere, we should consider implementing a UI designed precisely for smartphones.

Similar to what Hejnewar asked, does it matter what most gaming is nowadays? It matters more who would most appreciate Wesnoth and who would be a better pool of potential contributors. Maybe that's smartphone users, but that's not the obvious answer (especially for iOS), so I'd want to see some evidence before steering Wesnoth into "appifcation" or even touch UI. It might be a chicken/egg thing, but from what Pentarctagon says, the mobile users shouldn't become our target demographic.

[inferno8]

I just stumbled upon this discussion and I'd like to share my thoughts on the topic.

First of all I strongly agree with nemaara about these two most important aspects requiring some serious attention: modernizing the UI and marketing the game to new players.

By looking at Steam stats it turns out BfW has been losing players since the release in 2018:
https://steamdb.info/app/599390/graphs/

The all-time peak was 4 years ago (1019 active players). 3 months later the number dropped to ~200 and since then it hasn't improved much. This means the game lost 80% of players in the first 3 months after the Steam release (!). Something definitely discouraged these people from playing. I think there were two reasons behind this: 1) the game having an old-looking not intuitive UI 2) a completely "unfair" RNG. Speaking of RNG, I can imagine players being extremely frustrated about their mage "missing all 70% cth strikes, while that damn enemy with 20% cth never missed that turn! How fair is that?!".

I've heard that many modern RNG-based commercial games have built-in mechanisms implemented specifically to prevent very bad RNG outcomes causing players to go berserk (and uninstalling a game for good). I think BfW should implement something like that at least for SP. Btw, I recently played "Inky's Quest" with that experimental "biased RNG" turned on and it was a surprisingly pleasant experience. My units were missing strikes, yes, but I was able to predict outcomes much easier than with the "true" RNG enabled. Maybe this is the way to go.

Speaking of UI changes, I think having a modern (and simple) UI is something extremely important here. In my opinion the new UI should resemble UIs from other games as much as possible. I am talking about functionality only. I am pretty sure new players will find it easier to master a UI that looks familiar than something which is very experimental or old-looking.

I agree with Elvish_Hunter - There should be a UI designed with mobiles in mind. Ideally for both mobiles and PCs. Why? Because since 2003 the gaming has changed a lot and mobiles are now primary devices for playing video games. Just see for yourself:

Spoiler:

Source: https://www.statista.com/chart/26697/mo ... in-the-us/

I know that mobile users were very much in the minority in the past, but I am pretty sure it was partly because of the bad UI experience. How people are supposed to play the game on smartphones if it is currently a painful experience? If Wesnoth wants to keep up with the gaming industry, it should at least try to implement a proper mobile port with an intuitive UI and touchscreen support at some point. No cursor emulation! I'd love to be able able to play BfW on my phone or tablet and enjoy it to its fullest. The current port for Android is not good enough in my opinion, it's not up-to-date and suffers from some scaling issues due to lack of dedicated UI.

Btw, I decided to post some of my old attempts at making a dedicated UI for smartphones. I am not a UI designer, but I had fun making these mockups. Maybe they will inspire some of you:

Some of my old UI ideas for mobiles

From the marketing point of view it would be great if BfW was available on new platforms like consoles too. I work for an ad company and one of our clients is a global corporation owning one of the most popular consoles in the world and they go to great lengths to promote this way of playing. Even if console support was to arrive in BfW 1.99 I'd definitely wait for it.

The point 7) about implementing ranked matches also caught my attention and I second that. I am not big MP fan myself but I've been an admin of one of BfW fan-sites for more than 13 years and we held dozens of tournaments there. So I do know how important the MP aspect of this game is. Having the ladder system integrated into the base game would be a great move and an excellent feature for all professional players looking for some action and fame.
Also it would be quite handy to have some sort of newsfeed visible somewhere in-game, where players could be notified on new tournaments and other global community events. Not everyone browses forums these days or uses Discord. But this is something of low priority. There are far more important matters like the UI rework.

The greatest BfW feature to me personally is the ability to use this game as a framework to create custom content in form of add-ons and to play such content with others. From the marketing point of view this could be the ace in the hole. There are a lot of successful modern games revolving around creating content and sharing it with others (Skyrim mods, Minecraft levels, etc.). The availability of add-ons means it is hard to get bored with BfW once you give it a chance. And besides, I know of some people who thanks to this creative aspect of the game started a real-life career as pixel artists and programmers!
What I'd love to see one day is a special built-in editor for creating simple units, eras and scenarios. This would lower the barrier to entry for those who don't like to code or it is too complex for them (e.g. children). Making things more accessible should always be something to consider in the future.

does it matter what most gaming is nowadays? It matters more who would most appreciate Wesnoth and who would be a better pool of potential contributors. Maybe that's smartphone users, but that's not the obvious answer (especially for iOS), so I'd want to see some evidence before steering Wesnoth into "appifcation" or even touch UI.

This game really needs to find effective ways to attract more players. In my opinion this should be the major goal and no matter what you decide here, you should take that goal into account. I believe it's easier to catch fish in a pond full of fish than in a pond with little fish. The majority of players uses smartphones these days. It's a fact. If a proper "appification" (good UI, smooth controls) gave us 500 more active players, and even if 1% of them decided to be contributors or devs, this would mean 5 new people. For this kind of a game, that's a lot I guess. I am not saying that having a good smartphone app would magically solve all problems (it would probably create even more issues), but what other options are there? We could probably just pass time watching that Google Trends chart, which is a direct reflection of the current state of things (PC-only, no multi-plaform support, brutal RNG, old UI, problematic custom engine).

I've been part of this community since BfW 1.2 came out. I witnessed the golden age of this game and its lean years. What I know for sure is that in terms of gameplay Battle for Wesnoth is still a gem with an incredible potential. It deserves to be more popular. Whatever major goal you choose for this project, I'll support it

[Elvish_Hunter]

I'd say just remove them.

Thanks for your opinion That's what I was thinking to do as well, and I'll open a separate thread for the details.

The all-time peak was 4 years ago (1019 active players). 3 months later the number dropped to ~200 and since then it hasn't improved much. This means the game lost 80% of players in the first 3 months after the Steam release (!).

If we limit ourselves to Steam, yes. But if we look at the stats on SourceForge, we can see that in March 2008 Wesnoth 1.4 was downloaded 78'216 times; in January 2022, Wesnoth 1.16 was downloaded 12'737 times, which means we lost about 84% of downloads since then (but we didn't have Steam and Itch.io back then as alternative sources).

I've heard that many modern RNG-based commercial games have built-in mechanisms implemented specifically to prevent very bad RNG outcomes causing players to go berserk (and uninstalling a game for good).

If I remember correctly, there was even a mod in the add-ons server that disallowed our RNG and replaced it with damage reduction. Sure, this means that unit resistances and terrain defense become pretty much the same thing, with the defense acting as a generic resistance, but at least for some players it might be interesting to have this option.

I agree with Elvish_Hunter - There should be a UI designed with mobiles in mind. Ideally for both mobiles and PCs. Why? Because since 2003 the gaming has changed a lot and mobiles are now primary devices for playing video games.

I'll tell you a thing: a few months ago, I tried to install Wesnoth 1.14 on my Android smartphone. Would you believe me if I told you that the first thing I did after that was fetch a USB-OTG adapter and attach a mouse to the phone? Just to be clear, alep (BTW, has anyone heard of him recently?) did a wonderful job in porting Wesnoth to Android, but our user interface is designed for mouse and keyboard inputs, not for touchscreens and fingers.

This game really needs to find effective ways to attract more players.

That's what we all want

[nemaara]

@inferno8 Glad to see we're mostly on the same page.

I would like to throw in this thought as well. Our primary goal is (or imo should be) attracting more players and retaining them, as we said above. However, at present I don't think that offering easier ways to mod content or more modding features is the way to do this. While that is definitely important and a way to help already existing content creators, new players are not going to immediately jump to making their own mods just after opening the game. They'll mostly be players first, and then a certain portion of them are going to join the modding community. So I don't think anything in the backend is going to help us in this direction at the moment.

[Hejnewar]

If I remember correctly, there was even a mod in the add-ons server that disallowed our RNG and replaced it with damage reduction. Sure, this means that unit resistances and terrain defense become pretty much the same thing, with the defense acting as a generic resistance, but at least for some players it might be interesting to have this option.

I've heard that many modern RNG-based commercial games have built-in mechanisms implemented specifically to prevent very bad RNG outcomes causing players to go berserk (and uninstalling a game for good). I think BfW should implement something like that at least for SP. Btw, I recently played "Inky's Quest" with that experimental "biased RNG" turned on and it was a surprisingly pleasant experience. My units were missing strikes, yes, but I was able to predict outcomes much easier than with the "true" RNG enabled. Maybe this is the way to go.

Biased RNG is already pretty much that, it gives you predictable outcome and like it or not it makes things easier. Technically it sticks to a very close to the average, but the thing is that average is increasing by a lot amount of enemy units that your unit can take damage from without dying or risk of it and you should never play around the lowest damage you are going to possibly take, only around highest. (Dont play to wound, play to kill.) It also makes sure that you can almost always apply status conditions which is another buff for player, because player can use this way better than the ai. 

I would say that rng is good, because it allows players to discharge their frustration and come back. If there is no rng players will feel way worse because they will come to the realisation that they are just bad and this way they at least have something to blame for that. I say that with full knowledge that I wildly underestimate how bad players can actually be. (MP bias.)

I agree with Elvish_Hunter - There should be a UI designed with mobiles in mind. Ideally for both mobiles and PCs. Why? Because since 2003 the gaming has changed a lot and mobiles are now primary devices for playing video games. Just see for yourself:
Source: https://www.statista.com/chart/26697/mo ... in-the-us/

That for me is just a piece of data without proper context. How many of these smartphone or console players would play turn based strategy games (that usually take quite a bit of time) instead of candy crush (that they play in subway / metro)? Im afraid we will only find out after we try, but just looking at this and taking for granted can lead to a big overestimation of how things will actually play out. 

I agree that attracting new or even old players that are no longer here is important and I do have some plans of my own for how to maybe try and attract players, but at the same time I usually work in shadows and randomly come up with bigger stuff. I cant promise that it will happen but I can promise that I will try. 

[octalot]

By looking at Steam stats it turns out BfW has been loosing players since the release in 2018:
https://steamdb.info/app/599390/graphs/

The all-time peak was 4 years ago (1019 active players). 3 months later the number dropped to ~200 and since then it hasn't improved much. This means the game lost 80% of players in the first 3 months after the Steam release (!).

Is that unusual for Steam?

• https://steamdb.info/app/289070/graphs/ Civ 6: dropped sharply to under 20%, with some events getting peaks of 50% the original launch
• https://steamdb.info/app/1062090/graphs/ Timberborn: 30-40% of launch players
• https://steamdb.info/app/294100/graphs/ Rimworld: dropped to 50%, then peaked each expansion pack before dropping back to 50% of the latest peak
• https://steamdb.info/app/413150/graphs/ Stardew Valley: sharp drop, peaks when included in the Humble Bundles, then slowly building up

[Hejnewar]

Is that unusual for Steam?

• https://steamdb.info/app/289070/graphs/ Civ 6: dropped sharply to under 20%, with some events getting peaks of 50% the original launch
• https://steamdb.info/app/1062090/graphs/ Timberborn: 30-40% of launch players
• https://steamdb.info/app/294100/graphs/ Rimworld: dropped to 50%, then peaked each expansion pack before dropping back to 50% of the latest peak
• https://steamdb.info/app/413150/graphs/ Stardew Valley: sharp drop, peaks when included in the Humble Bundles, then slowly building up

That reminds me, TW: Warhammer 2 is considered very good game. It dropped within 2 weeks to about 50% from peak. Wesnoth peak is 1019 - around 300 of players that never dropped we have 719 players, after 2 weeks we still have 720 people - again 300 that never drop it is 420.
420/719 = ~58%, this retention rate is actually way better than most modern games have.

[LordBob]

I don't think that's the problem. The thing that holds back the transition to another engine is our engine itself, with its complexity and with the lack of a suitable replacement (the Godot effort, sadly, didn't bring us Wesnoth 2).

On that specific topic. The Godot effort at its highest was never more than 5 active people myself included, of which 1 has done 95% of the coding and another 95% of the art. We've ran into issues that a/ we don't cover every skillset, b/ the workload is high and c/ motivation is difficult to maintain under such conditions. Nevertheless, I do think that with a larger team, a coordinated effort and reasonable goals a Godot Wesnoth could bear fruit.

For the record, the current state of the prototype is this (units are placeholders) :

(yes, there's a day/night cycle implemented)

It obviously lacks most of the features of present-day Wesnoth, but I do think it's a robust start and could be made into something. Not 1.x because the road is long so we're sure to take detours and there will be plenty of issues if we were to port identically every aspect of 1.x. But there's definitely some good that can come out of it for Wesnoth.

[Pentarctagon]

My impression was that Haldric has more or less turned into its own thing, which while being Wesnoth-like in some ways, also no longer had the goal of being "Wesnoth 2.0"?

[LordBob]

My impression was that Haldric has more or less turned into its own thing, which while being Wesnoth-like in some ways, also no longer had the goal of being "Wesnoth 2.0"?

Yes and no.

Haldric did turn away from being "Wesnoth 1.x running in Godot" because it was a daunting task for 2-3 people alone to adapt 10+ years of content and struggle with the constraints caused by pre-existing choices (I mean that on the technical side, e.g. terrain transitions and the like), while also battling with the inevitable challenges of adopting a new game engine.

Which is why we opted for building a prototype from the ground up, and eventually had to make choices that may or may not correspond to how 1.x works.
Nevertheless at its core it is still a Wesnoth game, with the same terrains and factions, to the point where future campaign ideas we've discussed revolve around revamping the Rise of Wesnoth. Yes we did imagine a few different mechanisms, but it's nothing that cannot be changed.

At the moment it is not Wesnoth 2.0 and if I'm being honest also a bit in limbo. But the prototype is there, Bitron has done most of the heavy lifting to create a viable skeleton of a game, its repo is accessible in Github and from what I've seen, Godot is a flexible engine with modern-day features and an active dev team.
If it's only the 2 of us it will likely continue to be its own thing, or be nothing. But if the wesnoth devs as a team want to get to grips with Godot and use Haldric as a base for their own 2.0? Hell, I'm in!

It would be nowhere near simple and content would likely have to be produced in baby steps - for example, first release with a core set of terrains, 2-3 factions and a single campaign, then expand in ulterior releases. But it's definitely doable. The prototype in its current state is essentially the work of 2 people, so I have no doubt that a larger team with a more diverse skillset could successfully make a Wesnoth 2.0 out of it.

[Pentarctagon]

I'd be open to it at least, but it would need to be something that the current team as a whole agrees we should do (also the github repo looks like it's pretty outdated?). The elephant in the room of course was, and remains, that all current content (mainline and UMC) as well as nearly all art assets in 1.x would wind up being dropped in the current Haldric. Just switching out Wesnoth 1.x's GUI with an external library would be a much less invasive change, for better and for worse.

I'm curious what others' opinions are on this though. Early on there was also discussion around being able to reuse more of 1.x's code/content, so maybe that could be revisited as well.

[octalot]

Godot 4 is currently in alpha, and its main scripting language is sufficiently different to Godot 3's that it's probably worth starting with Godot 4 if we decide to move as a full team to that engine.
https://godotengine.org/article/gdscrip ... now-merged

I'll probably take a wesbreak once 1.18 is released, so will only be a minor player in the 1.20 or Haldric effort.

[nemaara]

From name:

It would open up quite a lot of game play opportunities if support for real ranged attacks gets implemented over the next development cycle.
The feature was being worked on, but progress seems to have stalled: https://github.com/wesnoth/wesnoth/pull/4859

And as a general comment, I think so too but this is a bit more of a backend thing so less of a "big goal".

[LordBob]

I'd be open to it at least, but it would need to be something that the current team as a whole agrees we should do (also the github repo looks like it's pretty outdated?).

It's possible Bitron created a fresh repo when we went 3D and you were looking at the old version? The one we've been working on until recently is here : https://github.com/Byteron/haldric-next

On the topic of the asset elephant... Well, unfortunately, if you do stick to 3D this is very likely. Initially we had started by adapting the game as it is now, 2D tiles and everything, but recreating the terrain transitions and macros proved to be too much of a challenge, hence why we eventually built something from the ground up.
3D is not compulsory, however. If someone can reproduce the existing terrain system in Godot and port WML or whatever it's called nowadays in a way that doesn't break existing scenarios - or at least keeps an open door for a reasonable conversion method - then there may not be a need to ditch existing assets and content.

Personnally I would be up for it, because revisions of something are often timid and/or conservative as opposed to starting from scratch and allowing ourselves to make unconstrained decisions. But I reckon it would be a couple years at least before the game can offer to the type of content it currently has, so I can definitely see the issue.

[Elvish_Hunter]

From name:

It would open up quite a lot of game play opportunities if support for real ranged attacks gets implemented over the next development cycle.
The feature was being worked on, but progress seems to have stalled: https://github.com/wesnoth/wesnoth/pull/4859

And as a general comment, I think so too but this is a bit more of a backend thing so less of a "big goal".

I'd like too to have real ranged attacks implemented. While it might be true that it's a backend thing, it's also true that its frontend effects would be very evident, at least in UMCs.