I hope I can do this topic justice. Every game faces the problem: making the game complex enough to be interesting to people with thousands of hours played (Hera estimates he has 20k hours played, for example), but still accessible and fun for new players.
Long-time players might not care very much about the experience of new players. This can be true even if the long-time players are the game’s developers, who might benefit financially a bit if the game is very successful. (Or a more popular game might just mean more profits for stockholders, who knows.) Most people just have fun playing with other people who already know the game. Hera’s 1v7 is a rather unusual event; most of the time, a top player like Hera has no reason to care if someone, whether they are a normal player or a popular streamer, starts to play AoE2, because it would be years before they became skilled enough to encounter Hera on the ladder or in a tournament, if ever.
But if we want AoE2 to grow, the new player experience is very important. And there is so much room for AoE2 to grow. We have gained many things over time, with DE: tooltips to explain attack bonuses. Auto-farm placement. But fundamentally, AoE2 is still a game which was made at a time (1998) when most people were still using 56 Kbps modems (about 10% of the bandwidth of 360p YouTube). Online play was not the main selling point. It was reasonable to expect that people would do single-player, including the tutorial missions that explained the concept of fog of war, before their first multiplayer match.
Now? People expect more. A game should be fun even during the first hour of play, even if someone decides to jump right into fighting other humans after installing the game. Sure, they’ll be bad at the game, but if there are thousands of other players online, why can’t a bad player be matched against other bad players and have fun?
That’s the introduction. I’m pasting what I wrote a month ago:
with early games, conflict between winning the game, and getting better at it. To win: do XYZ against other noobs, just spam knights. To get better: spend time reading tech tree, experimenting. So, too much pressure to win harms learning. Can happen with solo with disappointment from losing, or in team games from teammates getting upset at new players who literally do not know to continually create villagers for the first 5 minutes. Partly a matchmaking problem, but solution might require more. Most players playing campaigns and against AI does not mean playing AI is more fun, if most people don’t want to play AoE2. Many games are social. Complaints about Roblox age-based chat restrictions.
Maybe just letting players find their rating through FFA games, where there is less pressure.
Maybe: have placement games be FFA with at least one AI player, who attacks the strongest player. Use more metrics than just who wins to get a more accurate idea of rating. Combine with tracking quitters and afk players, so getting a low rating through placement system requires actually playing the game, not being afk. Allying disallowed, but setting to neutral allowed.
Allow players some control over when their placement period ends. A player who is afk reading the tech tree for half the game is not ready for 1v1 matches. After attaining a minimum competence score, based on the tracked metrics like TC idle time for early game or whether appropriate buildings are built, a button appears at the end of each placement match, to ‘graduate’ to ranked ladder.
Special AI behaviors when playing very weak players: military units pause and look threatening instead of attacking villagers. Could be enabled with a map or game mode flag, so implemented using AI script and advanced unit commands. Reduces stress for new players. Sort of the equivalent of AI player setting weak human players to neutral diplomatic stance, but without the safety assurance, and decrease in sense of danger, from actually doing that. The military units would not be retasked when the AI attacks another player, but would stay there looking threatening until enemy military shows up or they have been waiting to attack for a ridiculously long time.
If villagers evacuated the area, the military would not chase them, but might eventually attack buildings where the villagers were.
Lets people know when is an appropriate time to have military, better than the AI building military but just keeping the units in its base without ever attacking. That could train players to think that playing defensively is good, because the AI does it.
Compare to the AI immediately attacking buildings like a lumber camp, not villagers: player could think, “the AI is stupid”. Just standing and not attacking is clearly not something even AI would think is the best action, so it’s deliberate, not poorly programmed.
___
Evidence that new players would enjoy free-for-all matches more: battle royale games like Fortnite and PUBG, where a single player or team wins, and up to 99 other players are eliminated during the course of the match. In AoE2, if six out of seven human players in a FFA placement match (with one AI player) lose, then someone can be above average and still lose, so losing is not as painful. Most people are fine with just being “above average”, even if they’re not always the best at everything they do. And if people don’t get too upset at losing, then they can focus on the learning process that is necessary for really getting better at the game, even if it means performing worse in a particular match.
The FFA format also often results in lower aggression early on, as seen with T90’s community games, although the ability to ally with everyone is also a factor with these. Might be nice to introduce a lobby checkbox, similar to the Diplomacy checkbox, to allow setting other players to neutral but not allying, as I suggested for these placement games. I really think community FFA games would be more interesting with this setting.
Using FFA for placement matches would have the purpose of making the game more fun for new players. It would not make new players better at the game. I wouldn’t be surprised if the average person who has just started the game is worse than T90’s 0 Elo rating uncle (haven’t watched any of the videos). So fixing problems in the matchmaking system, including ‘smurfs’ and people who deliberately tank their rating by quitting multiple games in a row at the very start, is still important. Allowing actual new players to do FFA placement matches and then start their first 1v1 ladder games at 0 rating also allows ‘smurfs’ to do this.
I think that playing against someone who quits a couple seconds into the match is a worse play experience than playing against someone who is 200 rating stronger due to an inaccurate rating or bad matchmaking. See: Make early quitters and Elo-manipulators play against similar players
After doing several placement matches, it would be easy to compare metrics of player performance, like idle TC time, with the typical performance of players at different rating, to see what rating a new player is most similar to. This basically means starting new players at sub-1k rating, instead of 1k rating, and as players improve it can lead to ratings deflation, but if that’s a problem: Ratings deflation and Ladder rating improvements for fun and profit
SUMMARY: FFA ladder placement matches with seven humans and one extreme AI that attacks the strongest player. The AI’s units threaten but don’t immediately attack players who are very weak.
- Make it happen!
- Bad idea
- The new player experience is not important
- I don’t care
Uh, this is a bit messy but a comment: https://www.youtube.com/watch?v=i3H3wYjpatE&lc=Ugxf-MnS6-6Xj0THTPh4AaABAg
I’m not militantly against the new change to the boars but if the devs want to make it easier to play online they need to sort out the starting elo. Requiring new players to lose 20 times before they have a balanced game is a much bigger hinderance than any game play change.
172 upvotes.
And I thought of this a month ago:
anti-smurfing matchmaking mechanic: have ‘points rating’ separate from ‘matchmaking rating’. For a mature account, matchmaking rating falls more slowly from losses; player’s visible rating will fall, but will still be matched against opponents close to old rating. Removes reward of ‘play weak opponents’ from losing repeatedly, but only makes sense if deliberately losing is punished, otherwise players will continue to throw games and cause a bad experience for other players until ‘matchmaking rating’ finally decreases.
___
And now for something completely different: limited co-op mode.
I might have posted about this before, but I don’t see the topic, so I will just describe it again.
Normal AoE2 is about controlling an empire. In the original campaigns, this was 75 units. Now it’s 200 units. This is all a bit overwhelming for new players. Fundamentally, people just want to have fun, and it’s hard to have fun when you don’t know what to do, and can’t perform at the level that you want to even if you do know roughly what you should be doing. (It’s one thing to know that having idle villagers is bad, and another to consistently have zero idle villagers in late game.)
MMOs often have the problem that a player with a lot of playtime cannot meaningfully interact or play with a new player, because the first player has a much more powerful character. In AoE2, the difference is purely in skill, but it still makes interaction difficult. The solution is to control a limited number of units within a larger empire: limited unit co-op mode. Then a new player can focus on things like raids that just involve finding the right places to do damage to eco, or they can learn tricks like quick-walling that, with enough practice, can be done fast enough to remain useful to someone who also has to manage an empire.
Letting a new player feel like they are doing something useful in a game with strong players is one purpose of this mode. The other purpose, which is just as important, is making the game more accessible for new viewers.
‘limited unit coop’ mode could be combined with limited APM. Maybe best way to do a tournament, as it allows top players to control the empire without putting all the focus on them. APM limit set per player, similar to handicap. 20 APM (as actual commands, or eAPM) for the empire players, no limit for the ‘limited unit coop’ players.
Better for a tournament with a high prize pool, as instead of ‘a random player’, or ‘a player selected from a pool with an arbitrary limit on ladder rating’, can select the best player possible, just with limited APM.
I think I wrote a whole thing about how limited unit co-op is good for viewers. Without copying it, the general idea is that for a new viewer, following a game (like the 4v4 Rage Forest tournaments, with over 1500 units in play) is hard. They can’t judge whether what players are doing is good or bad. But if a player can only control a few units, then all of those units are important. The audience is allowed to care what happens to those units.
In a normal match, a caster can choose to follow a certain unit. But the player won’t be treating that unit as special, so the audience would inevitably get disappointed, so casters almost never do this. They look at the entire game, as this is what a typical AoE2 player or viewer cares about, but this is just too confusing for a new viewer.
Any viewer is more likely to care about a tournament with a high prizepool than a random game (unless it’s unusual, like Hera’s 1v4 against big streamers or his 1v7 against DOTA players). And what a new viewer is watching will probably be a caster, since that’s the whole point of casters. If the caster mainly follows a few units, it’s easier to understand for the new player, but the caster won’t do that unless those units are actually special.
So, limited unit co-op is the way of making specific units special. The rules:
New lobby settings, usually hidden just like handicap: APM limiter (actual commands to units, not keypresses or changes in selected unit), and unit control limit in co-op mode.
Example, 2v2:
Two players select player number 1. One chooses 20 APM limit. The other enables limited unit control.
Two players select player number 2, and do the same thing.
In a game just for fun, no need for an APM limit; it’s mainly something that would be for tournaments, to reduce the performance difference between players who could potentially control the empire (see: TheViper’s LimitedViper series), in order to justify any part of the tournament prizepool going to the limited unit players and therefore the audience caring about those units.
Limited units: can control a number of units equal to the empire age number. Dark Age = 1 unit, Imperial = 4 units. Control of the unit lasts until the unit dies or is converted; the unit cannot be executed/deleted. The only button that other units show for the player is a button to take control of the unit, which is grayed out if the unit limit is already reached.
Buildings, like palisade walls, that the limited co-op player’s units have placed can be deleted, like if palisade walls are used to delay boars when luring on Black Forest. The ‘empire’ player can also delete these buildings. The limited co-op player can’t issue commands to buildings, but can view their activity queues. They can garrison their units like normal, and can set ungarrison locations for a building which are active only for that player (like for efficiently setting a villager to a new task and ungarrisoning in that direction). The same mechanic could be used for allied buildings, with the ungarrison location depending on the player who initiated the ungarrison command. If all ‘empire’ players resigned or dropped, a limited co-op player would be given the option of taking control of the entire empire, or also resigning.
Examples of what a 2k+ rated player could choose to do if they could only control four units, in an empire with 150+ units:
-
Alcatraz-level quickwall micro (or alternatively, this)
-
Four monks, or three monks and a scout, or two monks and a scout and a villager
-
Four fully-upgraded mangudai going raiding, able to kill 40 villagers per minute (35 real-life seconds)
-
Finding ways to use a runner in front of a ranged army with ballistics
-
Just finding the right location to sneak some production buildings where opponent doesn’t expect them
-
Villager in a siege tower (now 0.96 speed without any infrantry garrisoned) to sneak past castles, or just dodging all the arrows with micro without the tower
-
Roleplaying a small party of people on a quest to do a thing that could not be accomplished by a larger force
Would new players and new viewers like this mode?
- New players would like it
- New viewers would like it
- Experienced players would tolerate it
- Experienced viewers would tolerate it
- Casters would like it
- None of the above
The purpose of the poll is to give the developers some feedback that isn’t just my opinion, but while I can see casters hating the mode for being something different and new, TheViper has made videos before about hero villagers in his games, most recently this one.