Training Times (Or Why the Fastest Click Wins)

[TheCJ]

4 hours ago, Thorfinn the Shallow Minded said:

because 0 AD is at least according to my understanding of the vision is not supposed to be a fast paced game

This I have to disagree with. I do not think a fast game necessarily means that "the fastest clicks win".

Especially because of the batch training mechanic, 0ad is very fast (in that you can get a lot of population in a short amount of time), but doesnt need a lot of clicks (one click every, 35-40 seconds? for groups of five for every barracks?).

 

The only other rts I played (AoE2) is very slow paced in comparison (booming in 0ad means you are full pop before 14:00. Booming in AoE2 starts at ~14:00 when you reach AoE2s phase 3), but needs a lot more clicks.

[chrstgtr]

3 hours ago, TheCJ said:

This I have to disagree with. I do not think a fast game necessarily means that "the fastest clicks win".

This. Everything that OP talked about in his last post isn't about clicks. The vision says 0AD shouldn't be "the most clicks wins." I don't think anyone reasonable wants that (see any thread on sniping). Players are limited in the number of their clicks in early game because they simply do not have a robust enough economy to do more than a few things. And, in the grand scheme of things, 0AD isn't a "the most clicks wins" type of game compared to other games within RTS and other genres. 

Again, if the game is moving too fast for anyone, I suggest to simply change the game speed to .75x speed or something lower. Changing the the actual train times messes with the underlying balance of the game and no one has put forward an argument on why that should change or how it would avoid a24's problems. 

[Thorfinn the Shallow Minded]

5 hours ago, TheCJ said:

This I have to disagree with. I do not think a fast game necessarily means that "the fastest clicks win".

Especially because of the batch training mechanic, 0ad is very fast (in that you can get a lot of population in a short amount of time), but doesnt need a lot of clicks (one click every, 35-40 seconds? for groups of five for every barracks?).

 

The only other rts I played (AoE2) is very slow paced in comparison (booming in 0ad means you are full pop before 14:00. Booming in AoE2 starts at ~14:00 when you reach AoE2s phase 3), but needs a lot more clicks.

Maybe Age of Empires needs a lot more clicks, but at least as far as I have seen, the EAPM of an average 2000 elo player in that game is around 60, which is significantly lower than Starcraft II's comparable elo, and much of Age of Empires 2's clicks in the early game may have more to do with things like quickwalling and deer pushing rather than aspects of the core gameplay.  

I assume that your talking about 3 APM is only in relation to training units.  That still does not account for other bits of APM of which it seems hard to find a metric for 0 AD.  Until that data can be found, I think relying on hearsay is unproductive.  

In watching some of ValihrAnt's gameplay, I saw that he was consistently getting housed while playing, meaning that even some of the better players struggle with the pace.  Raiding took place a bit past the 2 minute mark.  Another 1v1 had raiding happen during the 3.5 minute mark.  Granted, the game started marginally earlier than that.  If this small sample size is in any way indicative of the current meta, the early game is hardly boring.  Why then increase training times even if it does not impact the required APM too much as many of you have put forward?  

Even if it does not affect the APM too much, it does give a game with the economic complexity of an Age of Empires title a pacing akin to Starcraft.  If we compare Starcraft's resource system, it is worlds simpler.  When we look at the art and pillars of this game's design, it does not scream fast paced game, and slower paced games are better for drawing in casual play.  

3 hours ago, chrstgtr said:

Again, if the game is moving too fast for anyone, I suggest to simply change the game speed to .75x speed or something lower. Changing the the actual train times messes with the underlying balance of the game and no one has put forward an argument on why that should change or how it would avoid a24's problems. 

I hope that above I have explained properly to that effect.  Balance is always upset by decisions, and you consistently fail to explain why training times in Alpha 24 were so awful of a choice besides people hating it, thus I cannot address your concerns.  That said, I am sure that there are solutions to those that go beyond making the game into an experience in which you print out units.  

[chrstgtr]

29 minutes ago, Thorfinn the Shallow Minded said:

If this small sample size is in any way indicative of the current meta

Your small sample size is very unrepresentative of the current meta. Vali is so much better than the vast majority of people that his play style is very different compared to how most people play, including the very best players. So all your conclusions that follow aren't valid. 

34 minutes ago, Thorfinn the Shallow Minded said:

I hope that above I have explained properly to that effect.  Balance is always upset by decisions, and you consistently fail to explain why training times in Alpha 24 were so awful of a choice besides people hating it, thus I cannot address your concerns

To be honest you haven't really explained it. Instead, you've bounced around from idea to idea and now, for the first time, seem to be complaining about some unit spam meta instead of any of the other things you've previously mentioned. 

I've given you reasons why longer train times failed in a24 and you keep saying some variation of "well, go figure out a way to deal with it." No wonder why you haven't address any of my concerns.

Also, players hating it is a reason in itself. There are countless threads where players say that devs don't listen to players and failure is the reason why they are leaving the game, the game struggles gain a following, the game isn't better, etc. Almost every time devs come back and asks "what do you mean" or "can you give me an example, we never do that." Well, here we are again. 

[Dakara]

I don't feel like I've clicked a lot. At least I'm not bored in the economic phase.

And there are tools for those who don't want to click too much (automatic production).

I'm against extending production times. Maybe women by 8 to 10 seconds to indirectly nerf their economic boom and encourage unit military production. Increasing unit training times increases the snowball effect of the winners of the first fights, the power of champions or slows down the games if two players destroy each other's forces. Moreover, in these other RTS, dozens and dozens of military buildings have been made in the late game.

[borg-]

Well, I have played and still play some other competitive rts, and I can say, 0 a.d is extremely slow for me, and I often get bored during matches, especially against opponents 1800 and below.

The amount of actions I need to do in 0a.d compared to aom is ridiculous.

[Stockfish]

11 minutes ago, borg- said:

Well, I have played and still play some other competitive rts, and I can say, 0 a.d is extremely slow for me, and I often get bored during matches, especially against opponents 1800 and below.

The amount of actions I need to do in 0a.d compared to aom is ridiculous.

I agree to @borg- that 0 A.D. stands out for being a much less frenetic game compared to titles like Age of Empires II (aoe2) or Age of Mythology (aom). its more deliberate strategic focus where there are no constant micro battles that force players to perform hundreds of clicks per minute (except in high level games, but usually pro players like borg or vali doesnt overtask other players, tho they can do it perfectly) to manage individual units or special abilities. In 0 A.D., victory is built through macroeconomic decisions and largescale tactics, significantly reducing the number of actions needed to control standard matches. Micro plays are ofc very important, i usually like to do some 5 - 8inf minirushes to kill some wumen or counter rushes. 

I fully support the argument that the clicks required to play effectively are fewer here. While aom or aoe2 demand near surgical attention in every skirmish, 0 A.D. prioritizes planning and resource flow, making it more accessible to players who prefer strategy over reflexes. But ofc, as more reflexes and apm more advantage.

 Ipropose two key adjustments:

-Increase the base game speed to 1.15x, to streamline mid-game phases without sacrificing tactical depth.

-Reduce unit training times by 15 or 20%.

With this two things micro would be much important and the rest of the tactics remain the same.

Additionally, it’s important to highlight that 0 A.D. is designed for armies of 200–300 units, compared to the 100–150 of its competitors. This difference demands a less chaotic design but could also be leveraged to make large-scale battles more dynamic without overwhelming the player.

Anyway 0 A.D. doesn’t need to copy the frenetic pace of other RTS games, but it could optimize its rhythm to enhance its unique identity. I support @borg’s ideas and hope the developers consider these changes. Also i think 0ad needs a face-wash and make a better UI and PLEASE GO TO STEAM.

0 A.D. IS BETTER THAN MUCH OF ITS COMPETITORS, only needs few things and more players

[Stockfish]

That said, if you really know how to play, defenses in 0 A.D. can get incredibly demanding high APM, quick shortcuts, and precise micro. Building palisades, barracks, and towers midbattle while teleporting units to outmaneuver the enemy takes serious speed and skill. I love pulling off these kinds of defensive plays

And let’s not forget stopping rams on cc spamming palisades and houses to block their path. when you’re under pressure, 0 A.D. absolutely tests your micro and reflexes.

[real_tabasco_sauce]

2 hours ago, Thorfinn the Shallow Minded said:

slower paced games are better for drawing in casual play.  

I think casual players will play the game how they want to, irrespective of competitive players. A game with a steep learning curve and high apm can absolutely be played casually.

[Deicide4u]

28 minutes ago, Stockfish said:

-Increase the base game speed to 1.15x, to streamline mid-game phases without sacrificing tactical depth.

 

Yes, please. I don't want to wait forever for my infantry or rams to catch up to the rest of the army. And I surely don't want to wait forever for my buildings to complete.
 

29 minutes ago, Stockfish said:

-Reduce unit training times by 15 or 20%.

Not sure if I'd like this change, but I'm not a pro. I suck at this game, so I'll let better players argue with this.

[Emacz]

you can also increase unit movement speed slightly, which increases eco a little, to build a little faster and produce sooner.... we have messed around with various ideas on historical and open to trying others if you want!

[TheCJ]

9 hours ago, Deicide4u said:

Not sure if I'd like this change, but I'm not a pro. I suck at this game, so I'll let better players argue with this.

Well, we're having these kinds of discussions precisely because we want everybody to have fun playing this game and not just a small set of extremely good players 

12 hours ago, Thorfinn the Shallow Minded said:

the early game is hardly boring

This I would agree with. There are a lot of tactics one can do (or atleast try) in the early game if "normal booming" seems too boring.

But normal booming really doesnt take high APM; Just looking at some replays on the replay pallàs, normal APM values are between 20 and 40. For starcraft, an APM of 80 is considered low (afaik, I'm no sc expert though).

12 hours ago, Thorfinn the Shallow Minded said:

and slower paced games are better for drawing in casual play

I don't believe this to be generally true. The game can't be too slow, or it will be boring, even for a casual player. Of course, it can't be too fast either, and the most enjoyable pace for a casual player will be slower than the most enjoyable pace for a very competitive player, that much is true.

But I don't think 0ad should be much slower than it already is, even for casual players.

12 hours ago, Thorfinn the Shallow Minded said:

you consistently fail to explain why training times

But that's... precisely the problem, isn't it?

Of course it is theoretically possible to lenghten the training times to your suggested values while still keeping 0ad fun and unique.

But in order to do that, we would have to know exactly which other changes are necessary to achieve such a goal. Just changing the train time doesn't work, the players don't like it, it isnt fun. Then we would have to find out why its no fun; do the units die too quickly, so you can't reach high pop, do the buildings take too long, do they not deal enough damage... who knows. 

And if we did manage to find out what we would have to change aswell, we would still need to tweak the balance. (as an example, lets say we find out long train times dont work because units die too quickly, then we still need to figure out how to make all units die slower without destroying the balance between the units.)

 

Tl;dr:

@chrstgtr basically wanted to tell you that it isnt that simple, just changing the train time value will not work, finding out how to make it work is not that easy.

 

Lastly, in addition to such a change being a lot of work, the arguments supporting such a decision are not very conclusive.

[Thorfinn the Shallow Minded]

18 hours ago, TheCJ said:

I don't believe this to be generally true. The game can't be too slow, or it will be boring, even for a casual player. Of course, it can't be too fast either, and the most enjoyable pace for a casual player will be slower than the most enjoyable pace for a very competitive player, that much is true.

But I don't think 0ad should be much slower than it already is, even for casual players.

Agreed.  The point I made was that 0 AD has training times significantly faster than many of its fellow RTS games, and the numbers I offered, while placeholder of course, would be conservative changes relative to other games.  Some have suggested just changing the speed, which slows everything.  I for one do not want that to be watching my soldiers moving that slowly.  

18 hours ago, TheCJ said:

But that's... precisely the problem, isn't it?

Of course it is theoretically possible to lenghten the training times to your suggested values while still keeping 0ad fun and unique.

But in order to do that, we would have to know exactly which other changes are necessary to achieve such a goal. Just changing the train time doesn't work, the players don't like it, it isnt fun. Then we would have to find out why its no fun; do the units die too quickly, so you can't reach high pop, do the buildings take too long, do they not deal enough damage... who knows. 

And if we did manage to find out what we would have to change aswell, we would still need to tweak the balance. (as an example, lets say we find out long train times dont work because units die too quickly, then we still need to figure out how to make all units die slower without destroying the balance between the units.)

I would never claim that a single change would be without side effects, but the doom and gloom that some would suggest I think is a wild exaggeration.  Supposing that the early game feels far too sluggish.  Perhaps that's more the fault of things like not properly incentivising players to scout from the get go like most RTSs (Which I think is a problem in 0 AD regardless of whether training times are changed.).  From my perspective at least, temporarily ruining a product is a perfectly normal part of prototyping in the interest of creating the most robust iteration possible, and I do not think the risks are as massive with this comparatively modest suggestion.  

By the way, I searched the forums for topics with the keyword 'alpha 24,' and the in the one topic that criticised the alpha, I found that only christgtr pushed for training times to be reduced.  I could not find a single other person pressing for that change.  I'm sure there were lobby conversations about it, but the outrage regarding training times seems a bit overblown.  Granted, there were many things people didn't like about Alpha 24 (It was a long topic.), and maybe that was a lesser priority.

22 hours ago, TheCJ said:

Lastly, in addition to such a change being a lot of work, the arguments supporting such a decision are not very conclusive.

Fair, I don't think that it would be good if I tried to make them such.  Nothing substitutes empirical data, and while I did want to test alpha 24 to contrast its slower training times, I unfortunately could not get it running.  On the flip-side of what I've argued for, there might be good reason to even see how the community reacts to faster training times as many individuals have pressed for.  While I remain skeptical of the merits of such a change, if the community supports it after rigorous testing, who am I to challenge that? 

On 3/4/2025 at 10:35 AM, Stockfish said:

Additionally, it’s important to highlight that 0 A.D. is designed for armies of 200–300 units, compared to the 100–150 of its competitors. This difference demands a less chaotic design but could also be leveraged to make large-scale battles more dynamic without overwhelming the player.

While I think you make fair points, if we really wish to depict such large battles, shouldn't we be playing with units on a squad or battalion level rather than force players to manage hundreds of peons?  That would definitely make the design less chaotic and allow for meaningful micro that the game has hoped to introduce since this project began.  That, however, is a completely different can of worms.

I think that the conversation has stagnated, and I rest my case.  Even if 0 AD is less frenetic compared to Age of Empires 2 or Age of Mythology (Of which the evidence for, while important, is only anecdotal), as I said in my opening post, the factory-like way units pop out is odd compared to other titles, and considering bringing the numbers down is nothing radical with that perspective.  0 AD does not have to make itself unique by being one of the fastest paced titles in its genre.  

[Seleucids]

@Thorfinn the Shallow Minded if you could make a mod that changes the train times to the values you think are appropriate, I would be willing to test it with you in 1v1

[chrstgtr]

15 hours ago, Thorfinn the Shallow Minded said:

By the way, I searched the forums for topics with the keyword 'alpha 24,' and the in the one topic that criticised the alpha, I found that only christgtr pushed for training times to be reduced.  I could not find a single other person pressing for that change.  I'm sure there were lobby conversations about it, but the outrage regarding training times seems a bit overblown.  Granted, there were many things people didn't like about Alpha 24 (It was a long topic.), and maybe that was a lesser priority.

Look harder.  I found this in under a minute using a targeted search. 

On 02/05/2021 at 8:26 AM, ValihrAnt said:

Returning to the old train times is something that has been requested by quite a few people

The thread doesn't discuss why most players want current speed times (probably because it was widely accepted that current train times are desirable). 

Actually, only one person pushes back on shorter train times. It was you. You don't give any reason for wanting longer train times other than noting that other games have longer train times. 

Notably, your latest reason for wanting longer train times--to reduce spam--is directly refuted by...yourself: when you say "The problem with changing training times is that it does nothing to fix the fundamental issue [with spam]."

16 hours ago, Thorfinn the Shallow Minded said:

From my perspective at least, temporarily ruining a product is a perfectly normal part of prototyping in the interest of creating the most robust iteration possible, and I do not think the risks are as massive with this comparatively modest suggestion.  

This is quick way to lose the player base. After greatly increasing the player count during COVID, the player base almost immediately decreased with a24's release and it hasn't recovered. In the face of criticism, some used your exact logic to justify a24's changes. It's obviously a failed strategy. Purposely "ruining" 0ad would be a disaster

[Seleucids]

If the alphas can be released very frequently, e.g. once per week, then I think most players would have the patience. But if we are stuck on the same bugs for 2 years, players will loose patience. A notable example is the Han farming upgrade bug. It could have been resolved within minutes but the player base had to wait 2 years for the fix to be applied to them. 

This is exactly why I propose Thorfinn to make his own mod then play it with me and a few interested people to get a feel for it, before pushing anything final to the alphas. 

[Thorfinn the Shallow Minded]

5 hours ago, Seleucids said:

If the alphas can be released very frequently, e.g. once per week, then I think most players would have the patience. But if we are stuck on the same bugs for 2 years, players will loose patience. A notable example is the Han farming upgrade bug. It could have been resolved within minutes but the player base had to wait 2 years for the fix to be applied to them. 

This is exactly why I propose Thorfinn to make his own mod then play it with me and a few interested people to get a feel for it, before pushing anything final to the alphas. 

I agree that this would be the best way to resolve the current impasse.  Unfortunately though, I have never modded the game and only in recent days have been active on the forums due to a hiatus at work.  Once it begins anew, I doubt that I will have the time to properly commit to such a project, and given the nuances required, I doubt this would be a simple one and done matter.  If I do find the time and energy to do that though, I would be happy to reach out to you.

[Dunedan]

11 hours ago, chrstgtr said:

This is quick way to lose the player base. After greatly increasing the player count during COVID, the player base almost immediately decreased with a24's release and it hasn't recovered.

Most of the players who started playing 0ad during COVID, stopped doing so before a24 got released. There wasn't an immediate decrease with the release of a24, however during a24 the number of players playing online continuously decreased, until it reached a point where it was roughly equal to the number of players before COVID. I'm not sure if we can attribute the loss of players during a24 entirely to a24 or to what percentage it was just due to lifted COVID restrictions or people naturally loosing interest for a game they played for a while.

That said, I personally stopped playing 0 A.D. after a24 got released as it felt just dull and boring compared to earlier versions.

9 hours ago, Seleucids said:

If the alphas can be released very frequently, e.g. once per week, then I think most players would have the patience.

Even if we would want to, it'd probably be a bad idea, as only players with the same version of 0ad can play together and the constant need to update 0ad would annoy a significant portion of the player base. Many players aren't as invested into 0ad and aren't as tech savy that they'd want to update 0ad once a week.

In my opinion the perfect release cycle is somewhere between once a week and once every two years, however the number of players playing a26 online had been pretty stable over the course of its life time.

10 hours ago, Seleucids said:

A notable example is the Han farming upgrade bug. It could have been resolved within minutes but the player base had to wait 2 years for the fix to be applied to them.

For such issues, players can use mods. If mods are limited to a small number of changes/fixes it also becomes easier to get them onto mod.io and to convince other players to install them.

[chrstgtr]

1 hour ago, Dunedan said:

Most of the players who started playing 0ad during COVID, stopped doing so before a24 got released. There wasn't an immediate decrease with the release of a24, however during a24 the number of players playing online continuously decreased, until it reached a point where it was roughly equal to the number of players before COVID. I'm not sure if we can attribute the loss of players during a24 entirely to a24 or to what percentage it was just due to lifted COVID restrictions or people naturally loosing interest for a game they played for a while.

That said, I personally stopped playing 0 A.D. after a24 got released as it felt just dull and boring compared to earlier versions.

That isn't what the data says. The data shows that the number of players basically fell off a cliff with a24. 

 

[Dunedan]

22 minutes ago, chrstgtr said:

That isn't what the data says. The data shows that the number of players basically fell off a cliff with a24. 

What I'm referring to are the number of players playing online using the multiplayer lobby. The data you quote is from players having feedback enabled and from players using the Linux snap packages. Both of that isn't representative.

[chrstgtr]

14 minutes ago, Dunedan said:

What I'm referring to are the number of players playing online using the multiplayer lobby. The data you quote is from players having feedback enabled and from players using the Linux snap packages. Both of that isn't representative.

I don't have MP stats. Anecdotally, it is pretty clear the MP community hated a24. There were a ton of old players that stopped playing. There was an effort to stay in a23. Complaints about the alpha were endlessly expressed. Etc. 

Enabled feedback is a datapoint. There isn't a reason to think that either SPs or MPs are more likely to enable feedback than the other. To the extent that the feedback stats are more representative of the SP community a24 was also a massive failure there. It's also certain that the MP community couldn't drive that entire drop off. So there were definitely uninstalls. 

All this is to say that the "let's break it before we maybe fix it" is one that losses players.

[Deicide4u]

1 hour ago, chrstgtr said:

That isn't what the data says. The data shows that the number of players basically fell off a cliff with a24. 

 

Snap is universally hated in the Linux world, so your data is a bit...inadequate. I'd think most players will either use their distro's package, or just download and play on their Windows PC. Also, Debian has backports and Ubuntu has PPA's if new versions are wanted.

Either way, if the increased training times would cause another rift within the "pro" community, then you'd better not do that. The game needs good players to guide others.

EDIT: Increased training times, not decreased.

Edited 11 hours ago by Deicide4u
Inverse of what I've said before

[Stan`]

Just fyi we are now higher than we were for A23b with A26 and now a27

[chrstgtr]

3 minutes ago, Stan` said:

Just fyi we are now higher than we were for A23b with A26 and now a27

Great. I always find these stats interesting.

It would be amazing if we can consistently retain these users. 

[Seleucids]

5 hours ago, Dunedan said:

update 0ad would annoy a significant portion of the player base.

We can do 2 types of updates: micro updates where only public is changed, then large updates where the engine is changed.

For the micro updates, we can do an auto-updater: click a button and pull from the git repo.