Graphics improvements and Roadmap forward? A serious discussion.

Returning casual player here. Been wanting to play some RTS games so I've been looking at other games like BAR.

It surprises me just how far ahead BAR is in terms of graphics and effects. And that's not the only metric its ahead in, BAR has more players in a single match than Zero-K's average amount of steam players. https://skibidilolies.publit.io/file/image.webp (credit @WintergamingTV on youtube) https://skibidilolies.publit.io/file/image-h.webp

Which made me curious as that's also been an open source and free-to-play RTS. How did it get so popular while Zero-K remains an enigma to most people?

Over the days I've been looking at various feedback and grievances that I could find - A lot of the conclusions that I drew were that BAR had an audience of older TA players(or people looking for a TA remaster), that it had coverage from a lot of popular streamers and majorly that it has really good graphics and performance for a free RTS game.

It set me thinking, if Zero-K, which has several innovations like UnitAI, and equalized resources for buildpower(so working build priority), could have the graphical fidelity as well as improved (multicore?) performance as BAR, would Zero-K get just as popular if not more? I believe that Zero-K is superior in many metrics over BAR, its just that BAR happens to have
1 - Better graphics,
2 - Better marketing and
3 - A more attractive website.

Which is what led me to make this post:
1 - I am curious for the development path forward. This game does get updates and I have seen some engine updates and new units like the Odin. That said will this game really get to the point of BAR's graphics? I wish to see a discussion here about graphics and what makes BAR look so good - is it just higher resolution textures? Better directional lighting and reflections? More graphical effects?

2 - What sort of goalposts would be needed for this game to achieve in terms of development to catapult itself into the spotlight? I've been trying to get some of my friends to play and one even managed to join and play coop vs 2 AI. However most don't and some even said they prefer playing BAR than giving this a shot.

3 - The website could do with more creativity and polish. BAR actually has 3D models of their units on their website's unit wiki as well as interactive unit comparison graphs(which ironically would work much better with Zero-K's units with standardized unit costs).

I've also noted a few issues that imo this game may need to address -
1 - I've noticed this in multiplayer, both from watching replays youtube and in personal play, as well as from complaints in this forum - that resource sharing between players can be a major un-fun experience as it punishes good players but rewards bad players. Although there are advantages to this, i think an implementation that lets players at the start of the match(if not in real time during the match) adjust the ratio of resource sharing.

2 - Not really sure how true this stands today as there have been updates to the Zero-K engine, but apparently BAR handles multicore performance better than Zero-K. It could also be that Zero-K has a lot of UnitAI using the CPU, especially with 1000s of units. Still, i am curious if there are improvements to be made to this game's performance.

3 - Playable Chicken faction? Although this game uses factories as sub-factions, and while the choice of 4 commanders is available, this game doesn't have true factions. I know that factions is more a TA thing though the idea of having a playable chicken faction might be enticing to players(who want to play as "The bad guy" faction for example).

4 - Honestly I think a roadmap highlighting what's planned, what's needed, what may or may not be in the works would do this game a lot good. BAR's website does a good job highlighting their devlog, balance changes and roadmap - here's their Roadmap for getting the game onto Steam https://www.beyondallreason.info/development/steam-release

And with all of that if this game starts to get some serious steam I have no doubt donations and funding will kick in.

I think since the learning curve is a bit steep so it deters a lot of newbies, the graphics reason just feels like a cop out. Besides a lot of the time I usually play zoomed out anyways so the unit models never really enter my vision a lot. On the topic of the playercount issue, it's just luck. All we can do is keep rolling the dice by pinging people who make rts content in hopes that maybe we will get a lucky break and some video will shoot up in the staratosphere and attract new blood. What do you guys think?

Also i'm not really sure what a development roadmap will achieve since new players don't care about all of that and just want to jump right into a TAW match compeltely fresh faced.

I think there's a lot of room to improve the presentation of the game, but be aware that we're never going to match BAR in spectacle. Exponentially increasing scale is core to BAR's gameplay while ZK actively makes design decisions away from that direction in favor of being more interesting at small scale, and I think a consequence of that is we're always going to have less mass appeal than BAR simply because spectacle has mass appeal and the micro gameplay of ZK much less so.

quote: BAR has a current limit of 16 players per game, which is half of the current limit of ZK. So the above seems to me a misrepresentation of the status.

Heh, well i guess i could have said "...HAD more players in a match..."
Anyways, its based on this video released a few days ago.

quote: What is your personal opinion on the games?

Frankly I actually detest BAR for several reasons, having played Zero-K before. Its a good RTS for 2025, its smooth rather than janky and its competitive with tons of players. It also looks and feels really crisp, like I'm playing an Esports game. The music and sound effects are also some of the best. And umit has some creative, fun stuff of its own like being able to airlift enemy units with ur air transports, or being able to transport and drop turrets built within ur base. The graphics and effects are top notch, like i was seriously impressed by the EMP visual effects, such as the shockwaves that glaze off the units when EMP weapons hit.

But that's about it. I almost hate everything else about that game - the lack of UnitAI is what drives me nuts, and imo this is purely due to me being conditioned by Zero-K from having using it so much. Asides from that I really dislike the proportional difference in the way BAR handles resourcses, like why isn't metal, energy and buildSpees(in seconds) at a 1:1:1 proportional ratio?
There's also the lack of starting plays, realistically just 2 factories you can open with and thus just the same few openings in every game. Plus tiers makes it so that you're really locked with just raiders and man do they suck. Its all about micro in the early game and imo this takes away from the idea of the game being massive scale.

Also not impressed in the late game units and superweapons either. Titan, Juggernaut, Behemoth and Sol Invictus are mostly just duel weapon units that do nothing interesting, they just walk around with their heavy weapons firing at the nearest targets. They're not even... burtsy? Like Paladin in ZK has clear support capabilities with its long range laser, but also burst fires its disarming rockets. Scorpion is invisible, can stun lock other heavies and its a really cool looking unit. And yeah the Detriment can jump, has railguns with scattering projectiles, has a D-gun as well has anti air. And chad Ultimatium to one shot most of these.

And even despite having miles better graphics than Zero-K, gawd the designs of BAR's units seem really gaudy, more like plastic toys for kids. The yellow-red coloring doesn't help either, nor do the awkward sizes of units and structures. A lot of the times I can't tell expensive units from cheaper ones.

I guess I'm sorta "angry man yells at clouds" over the fact that its BAR and not Zero-K that got a grpahics overhaul and massive playerbase.

I think the most important improvements for getting and retaining new players would be to the web site and the strategic content on the wiki, not improvements to the game itself.

I think presentation is unfortunately very important for mass appeal. Something like 3D models on the webpage or army composition calculators would help here. Can Zero-K maybe just copy things like that from Bar?

quote: Despite this it has around 15K active players.

Could you please share where you get this information from? And what time period.

Based on what I computed (please check if I made it wrong), in the last 3 months (arbitrary, sure, but you are free to check more), ZK had 8495 different accounts playing as of today (see below for reproducing).

[Spoiler]

quote: Steel_Blue BAR and Zero-k share the underlying engine, the Recoil Engine. There's a lot of history as this engine has grown and forked over more than 20 years, and the BAR team has been responsible for most of the recent enhancements. The nature of BAR having a larger playerbase has increased the number of open source contributors as well, I believe. Graphics are not purely a model/texture quality problem. The Lightbringer Update shows BAR's attention to some of the other parts of graphics, such as all the lighting, and BAR has enhanced old maps as well as created a massive amount of new maps that all adhere to the same modern aesthetic. As BAR was being developed they worked on aspects like the UI and shaders for selection, the cute beautiful way units get highlighted when you click on them. It's easy to say that BAR has more dedicated creative-types who have contributed to its graphics in the last 5 years. Here's a call to action to improve zero-k's lighting if anyone is looking to do it.

Thanks for this information. I went through those links and I am impressed by what BAR has been able to do.

Some of the graphical details that standout as a Zero-K player that BAR has:
1 - Perhaps the first visual improvement i noticed was the expanded mirrored area outside the map. it makes the game feel like its playing on a large open planet than in a little box. It adds size and scale, and a sense of realism. Its more immersive. It doesn't seem like it would be much graphical work to implement and perhaps there's already some mods/custom widgets that do this.

2 - Ground textures. Perhaps with a higher resolution texture pack, maps can feel more visually updated. Even a few maps at a time could do the game some good. the terrain on BAR just looks fantastic. I wonder if it would be possible to even import BAR maps and play as a mod in Zero K.

3 - BAR nailed surface lighting and reflections. Somehow managed to have RTX but within screen-space only as an implementation to save performance. Dark areas are dark, while reflections are bright, shadows are sharp and dynamic. Not a fan of the over-the-top addition of lighting on all the units tho, way too gaudy and distracting. I'm no expert at graphics so I do wonder if it would be easier to implement just better reflective lighting and shadows onto units than say revamping the textures, 3D models and resolution of the units.

4 - Effects such as the EMP weapons have nice shockwaves in BAR rather than the electric shock animated sprite effect in Zero K. Also not a fan of the rainbow upgrade animation on commanders and geos.

I really wish there was some sort of info by the devs if there are any such plans to bring better graphics to this game.

quote: 1 - Better graphics, 2 - Better marketing and 3 - A more attractive website.

Fundamentally, people with time, energy, and some level of skill have to step up and take charge of one or more of these roles. This list reads like the primary skill of three of the BAR core developers, at least as far as I can tell from the outside. The doomy outlook is that open source games will only ever excel at the strengths of their founding developers, which would mean that ZK will never have marketing and that BAR will never be well designed. But I hold out hope that it is possible to add new skills to the core team.

Graphics

Graphics has a low, but positive, chance of becoming excellent with the current approach. Half the work is technical, which I can do a lot of, and people like ThornEel and garfild888 contribute graphical improvements here and there. I think ThornEel's plasma cannon trails look better than the trails in BAR, and the technology on display is 20 years old. Most of the more recent tech you see in BAR is actually in Zero-K, or at least will be soon, and anything that is highlighted won't be too hard to add. We have their fancy lighting system, but it is only used on Solar and Singu because I implemented it, then asked for someone else to configure it.

Here is an example of the work required: BAR has a 39,000 line configuration file just for the lights attached to units. I thought it had to be automatically generated, but the 350+ commits over the three years of its existence suggests that each of those lines was entered by hand. Creating this file required near-zero technical skill, in the sense that it doesn't require modelling software, programming knowledge, or anything that would traditionally be considered fine art. All it takes is someone with enough taste, energy, and time working away in notepad until they achieve an artistic vision.

A solid majority of the commits to the lighting file are by founding core BAR developers. I don't have the patience for that. I'll tweak unit AI and terraform design forever, but when it comes to graphics, I like it and advances can be exciting, but I'm generally not that fussed. The most important thing to me is that the graphics clearly show what is going on, and that they are appropriate to the underlying mechanics and use of the unit or effect. At least at this point. We had worse graphical problems in the past, such as terrible explosion ground scars, which I upgraded to passable as soon as the engine technology allowed.

In short, what we need is someone sufficiently obsessed with graphics to write a 39,000 line configuration file, or at least to oversee the writing. They would need an artistic vision that they want to realise, although to complicate matters, it would have to roughly match my ill-defined one (I too think a lot of BAR looks sort of bad). Someone has to lead it, to know how good a map has to look to be featured, to know how to apply and balance lighting effects. I keep adding BAR tech and either tone it down (like the mouseover highlight) or leave it unused, because I don't really have a vision for what to do with it.

There a lot of areas of graphical improvement that take more perseverance than upfront skill. Map lighting could be fixed, and the unsalvageable maps could be identified and culled. Even just cataloguing the BAR advances would be helpful. But beyond notepad editing, mapping and modelling are skills that can be learned.

Just an aside, programming skill can help too, at least if you are too lazy (like me) for copy paste. Compare the 25 lines to define Solar lights in ZK to the 160 in BAR:

• https://github.com/ZeroK-RTS/Zero-K/blob/master/LuaUI/Configs/UnitLights/economy.lua#L16-L40
• https://github.com/beyond-all-reason/Beyond-All-Reason/blob/master/luaui/configs/DeferredLightsGL4config.lua#L30307-L30462

Marketing

Marketing is in some sense easier than graphics, and in another it is much harder. I have an interest in graphics; cool visuals are nice and the technical side can be interesting. I personally have no interest in doing marketing or being marketed to. I don't like boasting or oversimplifying. 64_Bit_Dragon is doing some marketing, but I'm not sure how well it is going or how anyone could help. The upside of marketing is that it happens outside the game, so it doesn't have the same potential conflicts of something like graphics.

The hidden issue with marketing is that it takes a lot of work. From what I can tell, a lot of the time of a founding core BAR contributor is spent on marketing and community events. Just "messaging some streamers" appears to be insufficient. They have to be courted and managed, and you probably need to start with small streamers and build up rather than go for the big ones. This takes time and research. Polished videos and update posts help too. There are also Youtube channels that churn out game news, 64_Bit_Dragon got Zero-K into one for something. Having a bunch of videos hanging around to be used like this is another way to go.

Just look at the text at the bottom of this page.

quote: Zero-K is a free real time strategy (RTS), that aims to be the best open source multi-platform strategy game available :-)

I didn't write that, but it encapsulates the Zero-K approach to marketing. BAR is "The ultimate real-time strategy game" and "Real-Time Strategy Redefined". I don't have it in me to write such statements.

The Cold Takes sit at the intersection of the need for marketing and things I actually like creating. But the marketing aspect is pretty slim. I could probably create update highlight videos, but there is no way I have the time for it. Or maybe someone could repackage Cold Takes into some sort of design hype video.

Website

The website is tricky. It does a lot for the community by hosting forums, profiles, and being integrated with replays, maps and mods. This is due to Licho being a core developer with web expertise, although these days he is quite low on time. But it isn't flashy and doesn't give prospective players much to go on.

People try to improve matters from time to time. I'm not sure what happen to the attempts. Some try to overhaul the whole thing, and find that there is too much functionality to preserve. There was one, I think by therxyy, that focused on the landing page. I think a fancy landing page would get us most of the way there. Fancy news could be a big improvement too.

As for the steam page, if anyone wants to organise a steam page update, feel free. The content is mostly from 2018 still. Again, this require time and effort, but a lot less than the website. Send us improved pictures, videos, and text.

Your screenshot point is taken, although there is also a difference between staged shots with the UI disabled, and shots of the game that can be played. Eg is closer to staged.

Also, Discord is not an image hosting platform. Your links are dying.