@JaggedAppliance
I think we'll just have to agree to disagree on some points.

I think it's fine to work on some feature without the intention to include it. It could just be an experiment. It could be because it doesn't work out the way you think it would for the user. Or because you're not happy with the implementation of it for the future maintainer. I do this type of experimentation a lot and it helps me with learning and exploring what is possible. Even if some of it never ends up in the product directly. it's not obvious in my mind that if I spent time on something that it also means that I want to see it become part of the product in question.

As an example, initially I had a focus on better graphics. I had a great proof of concept, but I was unable to make an implementation that did not have a terrible experience for everyone involved. I dropped it at the time. Years later with the hard work and effort of @BlackYps we now have the shaders. And the map generator and editor support for the new shaders too. It's a good user experience because of some automation he included into the map editor. Now it's a great time to include it. For those that are interested, see also his work on GitHub here and here.

@JaggedAppliance said in Another dumb idea from Dorset:

I'm not sure what the point was about the cybran nano. Just that that was game design happening? This is getting silly now but yes it's related to game design, but again it doesn't make anyone a game designer if they were involved.

I did not make this claim. It certainly was not the intention. I tried to make the claim that the decision is about game design - not that the people making the decision are game designers because of it. How people perceive a game is all about game design. And this made an impact on their perception of the game for some users. Interestingly enough, just like it did for you.

I find it interesting that you write that you do not understand why I brought it up. And you conclude the same paragraph with that it was a bad change that damaged the identity of Cybran. That is why I brought it up! And that identity is perceived, it is in my opinion related to the design of the game. Just like the majority of changes are except those that are purely technical (same user experience, but better code structure or better performing code).

I think we have a different definition of game design, and specifically where game design starts and ends. In my opinion, even some mod authors can be characterized as game designers. They introduce new mechanics, new rule sets, sometimes even entirely new genres (looking at you Dota). But just because you are a mod author does not imply that you are also a (good) game designer. Just like people who are good at the game also does not imply that they are a (good) game designer. But in my opinion we definitely need people that look at the game from the perspective of a game designer. Which is also why I think the teams should be merged - just like they were initially if my history is correct.

Statutes are whatever. Let's be honest, it was probably written in thirty seconds. It reads like that anyway.

I agree with you that it feels like these statutes can use some work. But I disagree with dismissing them. Without the intention to patronize you, but statutes exist to help the community understand what a team is doing here, and how the team is supposed to work. It originates from this proposal that comes with this document that was approved during a general meeting of the association. To quote the document at the end:

The detailed responsibilities of each team are intentionally not specified here so they can be changed without requiring a GM. Instead the teams have the duty to define their area of responsibilities and write them down somewhere public.

Which is what the statutes are. The team can update them as they see fit. Take for example the status of the DevOps, Game and Promotion teams. They're much more informative about how the team operates, even if some roles are vacant at the moment. Back to the statutes of the balance team - for now it's all we have to work with to understand the role of the balance team.

I appreciate you sharing your approach as balance team lead.

And I agree that describing the average mod author as both passionate and (to be) crazy (with their ideas) is probably a great description, in a good way .

I also agree that it's important to guard the game from ideas that just don't work. To come back to my earlier post - I'd like to express that even if something should not be in the game, it can still be interesting to just explore the idea with the mod author/contributor instead of just getting a 'no'.

The mod author is clearly passionate about FAForever in some fashion by spending so much time and effort. The conversation does not have to take hours, something as simple as just having a decent conversation (over voice) about the idea together can be sufficient. It can be meaningful for both parties, even if it is just about discussing the context of it and why it was declined.

Which brings me back to my first paragraph of this post. To me, these conversations can be more meaningful then the changes becoming part of the product. And that conversation can bring in a new team member in the future. It certainly worked that way for me. The reason I am here is because @Uveso spent some effort on my first mods and/or pull requests to help streamline them. His open and friendly attitude is what made this place feel accessible. Thanks for that .

I've quickly reviewed it for you - huge fan of #3 (on the right). I'll give some more details later this week.

@Brutus5000 said in Another dumb idea from Dorset:

I don't think this is correct though. My feeling is there are lot of people who could implement any idea, but why on earth would they spent time on the ideas of other people instead of implementing their own ideas. Implementing own ideas is the reason why most people learned coding or modding on the first place. So they are not to blame for it.

Based on my own experience you can motivate contributors to work on ideas that were not (originally) their own. I don't think this applies to all area's of development - but it definitely worked out for me while I was game team lead.

The (Lua of the) game is a relatively limited project with limited complexity. You get to see people play with your changes. It can be very rewarding, regardless of where the idea originates from.

@JaggedAppliance
I don't feel I implied that having no bureaucracy is a great idea. Or that a dictatorship is a great idea. I also disagree that everyone who works on the game wants their stuff to get in. . As an example, take my history of things that did not work out:

• https://github.com/FAForever/fa/pulls?q=is%3Apr+is%3Aunmerged+author%3AGaranas

And that excludes the list of branches that are on my local machine. It's totally fine if things don't work out. What is not fine is that if you spent 20+ hours on something, but then are unable to have a normal, constructive discussion about it to try and see what it would take to make it work. That leaves a disgruntled feeling. This is what I tried to communicate. And that is the current status quo in my opinion the moment something touches balance.

I also disagree with your idea that the game is 'designed already'. I fundamentally disagree with that. In my opinion, all of the following is related to game design:

• Adjusting balance. After all, an unfair balance would not be as engaging.
• Adjusting path finding. We removed various annoyances to make path finding more predictable: 58f61afe and #4266
• Introducing new actions/commands. We introduced commands such as capping/ringing, spread attack that evolved intodistribute orders. Or the area commands, that we eventually not introduced because of game design issues.
• Introduction of new interactions/abilities, such as the Charge ability on the Loyalist and applying the (fixed) redirect ability to various Cybran naval units.

Even something as basic as the UI is related to game design, as the old UI was perceived as a window originally which would make the game less engaging. There are many more examples - almost all changes impact the feeling of the game. But what I find most interesting about your response is that back in your time your role was not to design the game. Maybe that was true back then. But if you take what I said here:

To make the game more fun then the original, instead of just more balanced for the top 50 players. To be less about eSports, and more about a engaging experience for the average player.

Make it a little less opinionated:

To make the game more fun, fair and engaging

Then you roughly end up with what's written in the team statutes of the balance team:

The main goal of the Balance team is the continuous balancing of FAF's gameplay to make it more fun, fair, and engaging.

Where balancing on its own is related to game design. As a few quick examples:

• Counter Play - Making Multiplayer Fun for the Opponent
• Perfect Imbalance - Why Unbalanced Design Creates Balanced Play

An interesting concrete example to me is the introduction of Nano Repair on the Cybran ACU. It traded some faction diversity (read: imbalance) for more balance. Was it good? No idea. Did it impact how people perceive the game? Yes! It created enough emotion for people to complain about it on the forums!

To circle back to the statutes, to make something fun and engaging are for sure a game design related task. That's what game design is all about. It's we all started playing this game in the first place. And it's why I think the balance team and game team should be combined. This separation makes things unnecessarily complicated in my opinion.

I do agree with you that it's not clear what the purpose and/or grand goal of FAForever is beyond just surviving forever. The statutes of the association state that it's purpose is the continued development of the game. But just as Yuval Noah Harari writes in his book about information networks that there's a many ways to interpret the written word of the Bible with many more consequences. So are there many ways to interpret the meaning of 'continued development' of the game !

@Saver said in SUGGESTION: AEON T2 Shield Generator Fix:

@Nomander I'm tinkering with this thought poetry right now. However, I don't know whether I can get the construction effect of the Aeon to the unit (Silver Sea). Unfortunately, I have already tried this in vain with another unit without success.

Reach out to us in Discord, it is easier to discuss the details there.

@Dorset said in Another dumb idea from Dorset:

I would hope we could get to the point where we narrow in on some of these changes and are able to hand over the work to the actual development team. Similar to the t2 Aeon Shield thread I would imagine the development team would just speak up when the time was right in whatever forum post was tracking such testing and changes and then it would potentially make its way into FAF Development soon after that.

I don't think your idea is dumb. But, this is sadly not how it works in practice. Not only are mods unrated, which makes a ten fold less popular. There's a more deeper problem in my opinion. And apologies for the long post.

Historically, there was one game team. But at some point the game team was split into a balance team and a game team. Based on my experience as game team lead between 2021-2024 there's some significant friction the moment something touches the balance area. This not to jab at the balance team - it's just a fact. Some (previous) balance team members even stated this publicly in the past. And this split into two teams is a mistake in my opinion. I don't want to write up a post to ask for approval. I want to constructively discuss it, toy around with it and see it (not) work based on observations made possible by a proof of concept.

Pulsar example

As an example, there were talks in various places about giving the Cybran a unit similar to the Absolver. Because of a lack of ability to implement it by the balance team, it did not quite get beyond just talking. As a game team lead, I thought this was a great idea. The first prototype was the made in January 2024. I paid @Balthazar to make the initial prototype according to the specs of the Absolver. See also #5869. The Pulsar was born. Now, the Pulsar is by no means perfect. But it was a great start and it allowed us to test, tweak and tune it. However, now more then a year later the unit still did not see the day of light. And at the moment the balance team is talking about how the unit will never see the day of light. It was never given a chance to flourish.

Some may respond: you should (or still can) just turn it into a mod! But that's not the point. The point is that in order to implement some ideas it can take many hours or even days of work. Take the Pulsar, there's a time lapse. And that does not include the work by @Nomander and myself to tweak it after the initial delivery by @Balthazar .

Painting feature example

Another example, take the painting feature (#6725 by @Ctrl-K , alternative approach #6726 by me) took many hours to make from several contributors. Whether that is tinkering on binary patches (#111 and #112 by @Ctrl-K and reviewed by @Hdt80bro ), tinkering on the Lua implementation and/or reviewing and discussing the features. The same applies to making a unit (modelling, animations, setup the blueprints and weapons, scripting when necessary, iterations on all of the previous). It applies to almost any relatively significant change.

Why is the painting feature more interesting and engaging to work on for me? Simply because there's less bureaucracy. There's less talking to talk. It's more about the feature, it's more about figuring out how to make it tick. How to make it fun and engaging. How to make it work for other contributing teams. How can we make it so that moderators have an easy time moderating them? And how can we make it expressive enough for casters and trainers to use it with ease? And at the same time, how can we make it so that players can paint conveniently yet at the same time have the ability to mute users that take it a bit too far. So far, the discussions surrounding the painting feature has been super constructive and progressive. The fact that there are two competing implementations is/was also interesting and beneficial to the end product, regardless of which one is chosen in the end.

Mods

About the idea of turning things into mods. Yes, turning things into a mod initially is a great idea. But it's also a bit of an escape. As an example, Equilibrium was a (large) mod about various balance ideas. See also its changelog. A lot of these changes were made to improve the experience of the game. The changelog is also written from that perspective. I wasn't around back then, but apparently this entire mod was created out of necessity because the bureaucracy at the time would just prevent it from even being taken serious. Now, years later, a lot of the features that are described there have become part of the standard game mode. Yet, nobody who worked on Equilibrium is even around anymore. With maybe the exception of @speed2 .

Mods are a great tool to toy around with features and create a proof of concept. But then an similar amount of time investment should be expected from those that make the final decision about whether the proof of concept works and should be implemented into the standard FAForever experience. That's not the case at the moment, it's totally out of sync.

Which also brings me to this point:

@BlackYps said in Another dumb idea from Dorset:

I agree with the others. Making a mod achieves what you want. The fact that we don't see many suggestions implemented for testing is not because there is no possibility to test things, but because nobody actually wants to implement it.

In my experience this is not true. There are numerous people willing to implement things. @Saver is a great example. So was @MadMax before that. And there's numerous other people that I can't possibly all list. Also look at the massive mod packs being made by various people, still now after more than a decade. As an example, take @CDRMV. Or just in general the amount of mod and map work being done by reviewing the vault.

I think it's just that the quality of the evaluation is not in sync with the quantity spent on something. It just sucks to get a 'no' without concrete feedback or a direction to improve and/or get accepted. And if you're lucky then the people in question even took the mod/changes for a spin, instead of the disapproval being based on hypotheticals. Meanwhile, you spent 20+ hours on it. That's just extremely discouraging.

User experience, bureaucracy and spread sheets

While we're throwing in what we do in real life - I've studied and taught game design as a student at the university. Game design is about creating an engaging experience for a specific audience. This contrasts with the average discussions here on FAF that's about spread sheets, statistics and hypothetical scenario's about extensive micro that only about 50 players can actually do in practice. Which brings me back to my first paragraph about the game team and balance team being two separate teams. The current approach and direction of the two teams is, in my point of view, fundamentally different.

As an example of the bureaucracy and the view being fundamentally different: mobile factories that actually work were already thought of and implemented in Equilibrium. It took years for the same feature to reach the standard game mode with #5227. See also all the other related work. Now all mobile factories have this feature... except for one. The Megalith still has the old build mode. You know why? Because the balance team thought it was unique. It already works different then the other mobile factories did before. And yes, it may be unique. But it's about the experience of the end user. Make it work exactly the same way as all other mobile factories for the end user. Wouldn't it be a ten fold better (user) experience if the Megalith would just poop out the eggs like the game team wanted to do? And - of course - it would still have its own smell... flavor and therefore still be unique !

Eventually this was done by mods. And they even made the pooping animation, which ironically is already implemented to some degree by @Saver and @Evildrew . You can find it in the vault by searching by author.

tldr: in my point of view the gap between the game team and the balance team make it impossible for these things to happen in a streamlined fashion. I already tried in the past and I still think that the teams should be combined. Just like it was in the original development team of the game. And that all members do not necessarily need to be good at the game, but instead have a good understanding of game design and the capabilities of the engine. To make the game more fun then the original, instead of just more balanced for the top 50 players. To be less about eSports, and more about a engaging experience for the average player.

While writing all of this, it reminds me of The Next Major RTS Will Fail. This Is Why. and Dear Developers, Stop Listening to Pros. And also why I am so excited about games like Tempest Rising. To me, that game is more about the average player and the experience of it then it is about eSports. Just like the original Supreme Commander: Forged Alliance .

And unrelated: I can highly recommend the earlier referenced interview with Chris Taylor by the University College Dublin as a whole. It does tend to jump around a little, but it's (almost) all interesting to hear to me.

And for those who read this and are wondering about the painting feature: we're actively looking for feedback on Discord. If you want a live demo of both implementations, just ask!

edit: writing a text is hard

@Deribus said in SUGGESTION: AEON T2 Shield Generator Fix:

Personally I don't like the 4 "petal" T2 shield generator model. Would it be possible to have an additional set of petals come out of the first?

Other structures (looking at you, factories) hide 'additional' mesh information for the upgrade underground. You can't do that here for things that are up in the sky.

The current shield generator:

The new shield generator, next to its upgraded counterpart.

I feel like it having two 'petals' at the top is important for the silhouette (looking at hq rework discussions, I learned). But based on how animations work, you can't scale them into existence or something like that. All you have is rotation and translation.

Top petals

Direction 1

About the top petals. Move the top petals on top of each other. Then hide the bones Arm_03 and Arm_01 recursively. That way it looks like the original model:

Then the first thing you do when you start upgrading is unhide the petals and move them to the right position. This way the time it 'overlaps' is minimal. And the original base model looks the same.

Now, this type of upgrade animation where we 'duplicate' the mesh is unusual. I don't think any other structure does this. It may feel out of place because of it.

Direction 2

This is based on how the animation of the Seraphim shields work. We could create a 'puddle' at the core of the base of the shield to create the new shield. Somehow, seraphim units manage to apply the build shader to only the upgraded parts. I'm not sure how that happens, but if we can reproduce that then we can re-use the build shader for the Aeon unit to scale the new 'top part' from small to large. Just like we do with regular structures.

This direction requires additional investigation.

Bottom petals

Move the bottom petals underground. During the animation you can make them move up from the ground. And as they do that, they start pushing the structure 'up' as they get into position. Hiding mesh below the ground is applied by a lot of animations of structures. The benefit is that the base of the model is then essentially unchanged to the original model, until you start upgrading of course.

With all of that said, thank you for your time on this. You took the extra step. Not only did you make the animation, you also took the time and effort to make it compatible with AIs.

@Caliber said in Add the Tombstones from the 2025 April Fool's Update as a Permanent Feature:

Its just a UI feature so people can choose if they want them or not in an in game setting

This is not a UI feature.

@CM_Nicholas said in Another Novax conversation:

Maybe something could be done about the damage, but if your nerfing damage you have to give it something else, someone mentioned more aoe, but then would that give it shield spill at that point which would arguably be even better than it is now, or then you have to think "oh shit, they can hit two pgens at the same time" which will kill your grid faster, makes targeting engies much more worth, etc, etc...

Shield spill is applied to all damage against a shield, of any kind. The only exception that I can think of is the tech 3 mobile shield buster of Aeon - I'm not sure if that spills.

@CM_Nicholas said in Another Novax conversation:

I think novaxes are relatively fine as they are, Maybe a slight balance change, but you have to remember a novax is 36k mass iirc, 2 of them costing about the same price of a t3 arty +/- 2k mass, but the damage output is nowhere near t3 arty, yeh arty doesn't have the precision of a novax, but a novax doesn't have the damage of an arty, it takes 2-3 artys of novaxes to be able to penetrate anything heavily shielded, and at that point you could have a Game Ender up.

In my point of view, the current 'goal' of a novax is not to bust a shield. It's either to grab all the unprotected but high value targets. Or to assist a Mavor to destroy the shields when they are depleted.

It also highly depends on what map you play. A Novax has little to no value on a map such as Dual Gap. It's much stronger on a map like Seton's Clutch where resources are more scattered.

@Dorset There's a lot possible in this game. Of course, not everything. But if as long as you don't draw outside the lines then all of this wouldn't even be much work.

@Dorset said in Another Novax conversation:

I agree with you something dramatic needs to be changed not just cost changes.

I've discussed this idea last year over voice, and I'll post it here. It has three parts. The goal is to actually utilize the satellite layer. This layer exists in the game, but there's not much going on there. Ideally these satellites become part of armies, like any other mobile unit. To support the Fatboy - from above!

Not a ground station, but a factory

At the moment you need an additional ground station for each satellite. This is a little bit cumbersome. Both from a programmers perspective (the satellite is not shared when your defeated - it is rebuilt at the station). As I think it thematically lacks a bit.

Proposal: turn it into a factory. Allow it to pump satellites, like a regular factory.

Multiple type of satellites

Instead of just having one satellite, there will be three:

• 1. A scout satellite. Relatively cheap and provides various types of intel. Enables the UEF to have Omni intel on-site to counter stealth and cloak, for example.
• 2. A missile-based satellite. This would be the 'default' satellite, at the same costs of the current satellite. It fires a salvo of missiles. And those missiles can of course be intercepted by TMD. The missiles do not track, just like regular missiles. They will have some AOE though. And of course, you don't want to use this against Cybran missile deflectors . By being missile-based, the satellite is easier to counter:
• • It makes the satellite worse against mobile armies. Just like mobile missile launchers.
• • It makes the satellite unusable against naval armies, as they are usually packed with mobile TMD.
• • You don't need to built shields everywhere, likely just 2 TMDs at each location.
• 3. A laser-based satellite. This would be similar to the current satellite, except that it's designed to bust shields. The beam is continuous. For its costs, it will deal relatively low damage to non-shielded units (think 1/4th). And it will deal about half the amount of damage against shields as the current Novax does. But then of course continuous.

All satellites will have an energy upkeep. The rational of @Nomander that upkeep is unusual for mobile units is correct. But given that these are the only units on the satellite layer, they can be their own niche. They are essentially flying buildings anyway. All weapons of the satellites also have upkeep to fire, just like for example stationary artillery and the Ravager do.

When a player runs out of power the satellites simply stop firing because of the weapon upkeep. When a player loses the last Novax center all satellites become uncontrollable and unelectable. They start losing health (and visually: orbit). When their health drops to 0, they crash to the ground.

Allow SMDs to intercept satellites (manually)

This idea is not new. It's used in LOUD. I like it thematically. It provides the player with an option to immediately counter a satellite if the player thinks it's necessary. For example, countering the laser-based satellite would be a no brainer if you are also being shelled by artillery.

Now, I don't talk about costs here because that's all too soon to discuss. It's the idea - the theme and the experience of it, is what this proposal is about. And of course the positioning of the units in the roster, what are their purpose and how do they remain balanced mechanically?

From a different perspective: what if interceptors are easier to produce, perhaps faster (yes, strategic bomber I am on to you) and higher fuel capacity to allow them to be more capable against tech 2 units?

The opponent can of course do the same and make additional interceptors too.

It is replaced with the Context Based Template feature. Primarily because of performance reasons, but secondary it is also more feature rich and less 'accident' prone. It's by default assigned to the tab key. Try hovering over a deposit and hit tab, then try again with a tech 2 engineer and hit tab twice!

@Phoenix_NZ What is the spec, both CPU and GPU of the user with issues?

Do you run the campaign with any mods enabled? If so, try turning those off. In other cases, please try and see if you can make a recording of the behavior. Also observe your task manager. Is the resource usage spiking, as an example?

Unless it does not work, you should be able to unpause any pause you initiated immediately. If you miss clicked, you can correct it as usual.

if a guy needs to afk for 30s+ a player may still unpause after 10s, being annoying, and wasting a minimum of 2-3 timeouts or more

Maybe I'm sleepy, but I don't get it? Wasn't this more of an issue before?

Can you define what makes it bad in your experience?

@goblinsly2 said in CPU performance tests:

Do you have any data on performance ? I am on 5600x and get severe lagging when playing with 11 M28's on 20x20 maps. After about an hour, depending on unit limitations, it will go down to -5. Mainly im interested on effect of x3d, to figure out if 5700x3d would be a worthy upgrade for me

This game is not made for 11 AIs of any kind. The Lua thread that AIs run on is single threaded. No matter how good your CPU is, you'll never be able to run 11 AIs.

My reference was to typical pvp games, campaign maps and/or games with 1 - 3 AIs.

I am not sure.

I managed to implement a rejoin-like functionality. It allows a player to attempt to rejoin the lobby after X seconds when the lobby did not manage to connect to all peers. It is similar to a player rejoining a custom lobby. See also #6479.

However, I noticed that the lobby implementation is brittle. The rejoin tends to fail by... crashing the game. And I'm not doing anything special there - when I exit the lobby it occasionally crashes. Maybe that is because I run multiple instances on the same machine, I do not know. At the moment it is disabled.

If we do want to take this road then I feel we need much more statistics. Grafana (a tool to see how many lobbies there are, etc) recently got back up with thanks to Sheikah and Brutus. We'd need to expand that to track other lobby-related statistics . One of them is the local status of a player. And when a player attempts to rejoin. And whether that succeeded.

All co-op maps are interesting of course! But beyond that, you could try out Rainmakers Survival, see also:

• https://gitlab.com/supreme-commander-forged-alliance/maps/rainmakers

It is on the vault. It's a map I made, so do take this me promoting my own things .

@BlackRedDead said in Suggesting rule change: make immediately unpausing after a player asked for a pause against the rules.:

Tho in that case, it needs an exception procedure for players being dropped out in the meantime! (to not await their unpause signals anymore!)

And hence, my case that everything gets complicated.

I appreciate your input, but it appears to me that you do not understand the code base and its limitations. We do not have access to how it works internally - there's two globals exposed to the user side of Lua that allows a user to (un)pause the game. We can add additional logic on top of that, such as a small delay. But we can not mess with how the signals are sent over the interwebs. That's all in the binary and it is definitely not something we should mess with.

I also do not believe that people will 'abuse' this. It feels really dark and pessimistic. A small delay in a pause exists in other games too. Besides, any form of abuse can be reported. And if individuals really want to stop or break the game out of spite then there's easier ways to do that .