No game changes prior to tornaments

Thanks for the clarification katastrophe. I appreciate it a lot more than scattered posts on the topic. I think I basically agree with where you are coming from, however, I put slightly different weighs on the trade offs that need to be made, and I think I have more information about the nature of those tradeoffs.

There is an idea of what ZK is aiming for, but it may not be as specific as you like. Also the goals can change slightly as we wander around design-space and follow whatever direction seems like the most fun. The fundamentals don't change. This includes things such as:

• Try to make as many land factories as possible ploppable most of the time (depending on their level of move type specialisation).
• Avoiding factory RPS.
• Making all units useful in some situation.
• Making Planes and Gunships both be reasonably viable as support/switch factories, and have them represent a level of escalation.
• Have early game interactions not be so swingy that they regularly end games early, but also not to remove cheese entirely.
• Be able to play a lot of game sizes on a lot of different types of maps.
• Avoid removing weird physics interaction when possible. Making the interactions consistent enough to be usable is fine.
• Avoid having games invariably turn into porc wars. Avoid the lategame strategy bottleneck (eg spam Funnel and Bertha).
• Have commanders be decent but don't let them dominate the game.
• Keep everything interesting ie Quant's Rule.

I think you may not see the underlying goals because many patches don't actually take Zero-K directly closer to the goal. Correct me if I'm wrong, but you may be expecting steady progress towards the goals, whereas I have come to learn that attempting to progress directly towards a goal is often a bad way to go about balance and design. I think this could be what fundamentally seems to set our stances at odds.

You might think that since we're close to a lot of these goals that improving on them would be easy, but it is the other way around. Improvements in one area wobble the game, pushing other parts of the design loose. Sometimes, instead of trying to eek out improvements with minor tweaks, it is better to back up and look at the underlying systems to try to find a way around a blockage in the design (at this point I'm basically describing simulated annealing, but I'll push on anyway). Add this on to the fact that players will have their own, slightly different, goals for the game, and patches end up including things that various people like or dislike. This is inevitable for a game that isn't just locked in a holding pattern.

The mex cost nerf was in response to a lot of changes, built up over years, that caused the early game to be a bit too much of a mad land-grab. If the best way to play is for everyone to move out ASAP, with minimal defense, then the economy advantage can quickly snowball into a prematurely determined game. Changing the mex so drastically certainly destabilised balance in the short term, but I think we've reached somewhere better because of it (if not yet, then eventually). The retreat-advantage change is similar - raider interactions were quite swingy and it became apparent that a 'perfect' balance would be knife-edge and may not even be much fun. Another recent example is Bolas. The addition of Bolas did not make Hovers more balanced, however, I expect that giving the factory two raiders will make Hovers a lot more balanceable in the future. Dagger has been a problem to balance for many years, I don't back up and try a new route lightly.

On the topic of urgency, there is not much point honing the balance of aspect X if there is a deeper issue with aspect Y that will invalidate the work done to balance X. How should I address factory RPS before addressing more fundamental issues with how raiders behave? For example, Spiders don't have the units to do well in the type of game where mad-dash expansion is near-mandatory. If I want to work on Spiders before working on the pacing of the early game do I:

• Give Spiders ridiculous buffs to counteract their economic disadvantage?
• Move Spiders towards the average by making them faster and Flea more like a raider?

The former solution will make Spiders OP when the mad-dash expansion issue is addressed while the latter makes Spiders generic and boring. I would much rather have a slightly underpowered factory that is waiting for the right time to be fixed than render a factory boring in the service of quickly making it viable.

Urgency is asymmetric in that something being under powered is much better for Zero-K as a whole than something being overpowered. Some things languish in the niche basket because it is tricky to devise a way to improve the unit without making ZK worse as a whole (see Skuttle). Many other issues have competing voices so it takes a while to sort out the nature of the issue and what sort of change would solve the underlying problem. People will have different ideas of what is urgent depending on how they like to play.

On introducing something that is OP at first to get feedback, introducing a balanced unit isn't an option. It is too small of a target to hit. Even if it is somehow actually balanced people will either:

• Not know how to respond to it, and think it is OP.
• Not know how to use it, and think it is UP.

The best that can be done is to quickly respond to how the new thing is behaving and tweak it before it gets too monotonous. If I seem to have a habit of choosing OP over UP (and the choice is hard to make deliberately), perhaps it could be due to OP things showing up in games and being memorable, it would take some digging to find out. I don't think the game has been hit by such situations all that often over the last four years, but our experiences will differ based on activity.

quote: i am completely against making tourneys close to patches. but if i recall, the last few tourneys ALL were directly after patches? or is my memory tricking me?

I don't want to dismiss you out of hand so I checked the tournaments from the past two years:
[Spoiler]
It looks like I only failed to not make a balance patch within a week of a tournament on one other occasion, and just for a Dominatrix nerf. I've always thought a week was plenty of time so I wasn't attempting to attain a larger gap. This thread shows that I should at least pay more attention to the one-week buffer.

quote: do you actually enjoy simply PLAYING your game or is it pure work for you? I do not know you very well, but from my own projects i learned that one can get pretty ignorant of the fact that there is a huge gap between ones own perception of a project and the perception others have of it.

Yes, but I enjoy it more when I can play a game without being haunted by bugs and issues that I have already solved.

quote: i prefer (PREFER) getting the variation from my opponents, not the game.

I like this ideal but in practise some mix of the two is required. Even in games with much larger playerbases people play in patterns and the meta solidifies. At some point with Zero-K I was hesitant to make balance changes because 'good players' were not particularly active at the time. I thought that any changes I made would risk undoing whatever progress we had made towards balance back when the game was being more thoroughly tested. In the end I decided to lean more towards balancing the game for the people who were actually playing the game, rather than those who could in theory. If the meta had decided that something was OP then I was a bit quicker to make a change that would allow more diversity.

Sprung
exactly not. :D Thats when i think change is needed desperately.
as i said: "so, after those patches, everyone plays the new op-thing, wich is as boring as playing the exact same thing for years and doesn`t feel satisfying in any way."

GoogleFrog
Thank you for your answer. finally i feel like an instance of our communication has not failed. even if we have our differences regarding the game, you are the dev and i am not, i can understand your points and live with them. this relieved like 1 or 2 years build-up frustration in me. you made me a happy man (until the next patch i guess...)

NorthChileanG
but if you imagine an ideal scenario where you could plop every factory on every map and could start with whatever you want because it`s not autolose, wouldn`t that also bring you to a maximum of playstyle-diversity and suprise?
(I know this is impossible to achieve, specifically because of maps/terrain.)

GoogleFrog
how do we factor in maps? How should i adress map-selection when testing?
i would also like to give my detailed thoughts about retreat-range-advantage, but i don`t know if you want to spend your time with even more text i spill out...

that is extremely funny because exactly what you describe was what i got to as well when i thought about a "balancing" map. I wanted to get into mapmaking for this specific purpose, but i was unsure if changes made to units in the future wouldn´t render all the thought put into the map useless in the long term.

quote: I share the ideal, but i'm pretty sure Shifting Sands is Amph.

I've had mixed results with amph on SS. They might do better now though I'm not sure what their current power level is. But yeah, as my alterations suggest, I'd reduce the water to more limited, localized areas. Perhaps you could have a mex area with shallows, and one with plateaus, one with flats etc.

Getting a bit off topic now so I'd suggest we shelf this chat within this thread, but anyone please feel free to take it further.