Unlock the engine to allow more than 8 players (+ new colours)

Hello,

Some campaigns have been pushing the potential of having up to 8 players (7 AI players around you) to the limit, which has been the limit since ye olden days of the CD version. But AOE2 has gone bigger over the years - more civs, bigger maps… as the higher processing power allows - so why not increase this cap too, allowing for more than 8 players in a single game ?

Imagine custom scenarios with 12, 16 or even more players, how grand it could be.

I’m mainly thinking about custom campaigns as it would be very experimental for multiplayer, but if it’s stable why not also allow to go crazy in unranked ?

Much of the engine would need being tweaked as I assume the 8 slots are currently hard-coded, and this experimental version would surely be more CPU-intensive by the mere virtue of bigger bones, let alone the higher amount of units on a large map, so it would probably need being a separate build from the regular AOE2 (that would retain the 8-players cap) for a while at least.

As a bonus, it would also mean having more player colour choices to make each player still recognisable. Black white brown burgundy navy pink… I’m sure scenario makers would make great use of the new choices.

What do you think about this idea ?

16 Likes

I would absolutely love this.

5 Likes

It would be really cool, but a change in the source code of that scope would be extremely tiresome to do

5 Likes

Have you tried a recent scenario called wogastisburg? Blink blink
There I try to somehow surpass that limit artificially using the change civ name trigger and through apposite triggers to modify units stats but yes the colours are just 8.

1 Like

AoM is the only game in the series that as 12 player support, which is kinda ironic since it’s the only game that doesn’t even have 12 civilisations.

I know it would probably be very hard to do since the number 8 is likely hardcoded in many many parts of the code but it would be so nice to have 12 or even more players.

AoE2DE supports such massively huge maps that 8 players just feel empty.
There is so many scenarios that can’t really be made since you want more then 8 individual actors.
I would personally not mind if we were still limited to 8 human players as long as we get more AIs to make bigger scenarios.

I was working on a fall of Rome scenario recently but I had to cut out a lot of important civilisations making many parts of the map kinda empty:

  • Slavs
  • Saxons
  • Berbers
  • Persians
  • Armenians

Because of the 8 player limit.

Same happened to me in RoR too when I wanted to make an Alexander the Great scenario. Had to cut out the Phoenicians.

The AI is also pretty bad at handling multiple independent bases. It can practically only have one base.
So it doesn’t work well if you want to have one civilisations with 2+ cities distributed over the map. You kinda need to make each of them their own player. But then you can only have 8 cities on the whole map.
Sometimes you need 1-2 additional players with special roles like bandits too.

A higher player limit is like the one thing I want the most in AoE1/2/3 (I never made scenarios for AoE4 yet).

Performance is often not even really an issue. If I want to fill the map with different cities I’d likely want to limit them to a low population anyway so they don’t become too powerful.

5 Likes

MP lagfest someones gonna red clock starting in darkage lmao

hopefully in another 5 years and all PC generation moved forward a bit we’ll see this feature implemented. technically speaking its probably already in the game, genie editor shows up to 16 color so it might have been at point designed but scraped due to performance issue

3 Likes

yes and?
If we want that we can get that.

If we can now play on the largest map with 500 population and 8 players, why can’t we play on a smaller map with 200 population and 12 players?

I don’t even care if we get more human players, just allow me to add more AI.

3 Likes

That’s why I proposed it as an altetnative build, don’t force the heavier bones on weaker PCs. Ideally the heavy mode would be backward-compatible.

Why a separate build?
If you can’t handle 8v8 then just don’t play it.

Many PCs can’t handle 4v4 yet either.

yes and?
I am talking about MP games lagging

and if you didn’t know, that also affect single player game

3 Likes

Yeah but even if not everyone can use this feature why shouldn’t it be implemented for the people that have powerful enough hardware?

Because you’d need a heavier coding to accomodate for the higher player cap, if it’s indeed hard coded. The game would become more CPU intensive than it is now, for games 8 players or fewer.

But so long that both versions can still open the same files (up to 8 players) and play them in multiplayers it should be of little concern. Got a weak PC, keep the regular version. Got a powerful one, pick the heavy version which has access to bigger games too.

Though maybe the increase would be negligible indeed, in that case there is no need to bother with 2 versions.

i didnt say anything about it shouldn’t be implemented? all I said is lagfest and its the truth, u think I dont want to have 12 or 16 players? well I do but then we all gonna lag hard

there are players already lagging in dark age or feudal with very minimal units, and then there are games that are like 300-400 pop for 8 players so imagine what would happen for those going for 1k pop 16 players. prob not even the best PC can handle it since the ai pathing of this game is single threaded

heres an example what it feels like for just 1vs1 200 vs 200 and its all from pathing. now times that by 8 for 16 players 200 pop.

heres the vid reported 3yrs ago

Heavier coding?

How should the code for more players reduce performance?
Like why should it take any longer if those player numbers are not being used?

1 Like

If it’s hard coded with lines of codes forming solid slots, then you’d run that heavier code, even if most would be suppressed by conditions enabled if the player exists.

But I’m no software engineer and I don’t know what the source code looks like, so the difference may be negligible. Or it will not, I have no idea. Anyway it would be easy to test in practice, run the same 4v4 AI only game N times on both versions and compare average performances.

1 Like

Guess what?
I am.

Me neither and to be honest I rather not want to know how that spaghetti code looks like.

I think you could, if you wanted to, come up with a code that runs slower when there is a higher possible player number. But that would require purposely writing the code as inefficient as possible.

We currently have 9 players (Gaia = 0 and then 1 to 8) that means the smallest we can save that number is 4 bit, which would allow up to 15 players plus Gaia.
But that doesn’t really matter the game probably runs in 32 bit anyway.

The bigger issue would be to recode some of the stuff that is manually set up for 8 players.
For example some of the “resources” store per player values and because all the resources are numbered all the way though they would have to either have a gap between player 8 and 9 or rearrange the resources.
But guess what. They just rearranged the resources in the last patch, probably for very different reasons though. But that shows that it doesn’t seem to be that hard to do for them.

3 Likes

Yes it’s the stuff written for 8 players that I had in mind. It feels tempting to “bruteforce” its construction given the small amount (write for 1 player and copypaste 8x with small changes), instead of making it streamlined and easy to expand.

To avoid changing the interface design too much : slide bars while only keeping the room for 8 at once.

Why they used the same 22 year old engine instrad of trying to use a newer more open sourcable and less hard code oroented one sounds smarter

1 Like

Some random map scripts would break with more than 8 players as they are right now. This includes any map that uses direct placement (which is required for non-circular or any other non-standard start locations - though the map nomad would be fine since it technically has player lands, even if it doesn’t use them).

I do think that the recently added additional map sizes (under a launch parameter) and the change to drop one player (rather than all) on a desync are necessary steps towards changing the player cap. There may be other changes that would need to be made (beyond what’s already been made and what’s already been mentioned).