Why would you have left FAF?

Hi all, I'm not sure I'm legit engouh to post here: I really suck at sc, although I love it. I first played when vanilla was released, and a couple of years after FA. Took a long break, and now I'm back, and happy to be.

What is intimidating when returning is that you don't feel part of the community easily, although FAF and the team running it are quite amazing.
Of course, the learning curve is steep, this is a demanding but rewarding game. But when you play a custom game, more guys play to win, and forget about the fun. When you enter a game and say "hi", often enough no one answer... Too bad.
About the game itself, hard to know why 90% of newcomers leave... many of you have good ideas about what could make more stay !
Having played less than 100 games, I would like to have access to a more "structured" lobby: what is a "noob" now ?? is means nothing anymore ! Hard to find a game where your actual rating is representative of what's gonna happen ingame.
Anyway, thanks for the good work, and for keeping sc alive

@MazorNoob said in Why would you have left FAF?:

How about we make use of the achievement system and make a bunch of low-effort scenarios to beat?

Yes, and have the client keep track of user progression. Treat each simple scenario as a lesson, so when one is finished the client can suggest the next one. The client itself would encourage low-rated players to play these scenarios.

Some of them would have limited tech. So you might have a scenario where the only possible thing you can do to survive/win is to spam land factories and T1 tanks (because the other units/buildings are disabled). In this way, players would learn how to do a simple build order and how to spam T1.

So some scenarios would be about teaching people to build/use a single unit. The campaign already does that, but not as well as it could. For one thing, campaign missions tend to be too long. And they encourage turtle-style gameplay. A scenario that encouraged a faster start (encouraging a fast build order so they can deal with immediate threats) and then taught players to use a particular unit and only lasted 8 minutes would be better than a 30-minute campaign mission turtlefest.

Or just have a premade base to skip past the turtle phase. Mix it up between teaching build orders and skipping them entirely. Just giving people premade bases with adjacency bonuses in place will give them the idea to make bases like that for themselves. If noobs have a hard time even imagining making 12 land factories with mex adjacency and only 15 pgens (most noobs would get 50+ pgens before their 12th land factory....), then drop them into a lesson where they start with that and they will start to think that it's normal.

You could have a scenario where there are no mexes close to home, so the player would have to take reclaim to get started, fight there way to more reclaim, before they could fight their way to some mexes, and then have a normal game against an easy AI from there. So the player could learn how to use light artillery to break some civilian PD, with limited mass so they have to build the correct unit types. Maybe in that scenario they don't even get an ACU (no generated mass/power + you can't bulldoze through the PD with your ACU).

There would also be scenarios where the only goal is to grow your economy to a certain size. There might not even be enemies at all in these scenarios. That would help to teach players about how to grow their economy efficiently. And they would always have a few examples of replay files they could watch to see how other people did it. So a lot of the lessons would essentially be puzzles: "you can do this 8-minute puzzle to learn the game better."

Every scenario would be intended to teach a small amount of information in a memorable way. Some would be more open ("beat this AI, however you want to do it") and others would be much more rigid ("overwhelm 3 triads by making a lot of lobos and then the scenario ends immediately").

How about a scenario where there are no mexes, no ACUs, you start with 4 land facs, 20 tanks, 5 engineers, and 1 scout plane, the AI has like 70 tanks, and you have to scout out the map and pick off AI forces with "defeat in detail" in order to get enough reclaim to make new tanks to win. It would teach scouting, looking for mass-efficient fights, and taking reclaim. For some lessons, you completely remove eco management so they can focus on units. Other missions, completely remove combat stuff so they focus on eco. Instead of always throwing players into a situation where they can win by improving eco OR by being smarter with units (which doesn't specifically teach anything, it does push them to get better, but not in the same way as specifically teaching 1 skill/idea)

The idea would be, if a player completes all of the lessons, they should have enough skills to play at the low end of 1v1 ladder without feeling bad or to participate in low-end team games without feeling completely lost.

Rather than going for voice acting or lots of text in the scenario itself to explain things, the explanations should be kept brief and outside of the scenario, to make them easier to prepare. People who can't beat a scenario would watch replay files to learn how to do it. That way, there is no language barrier for anyone. The "mission briefing" in the client for each scenario (before you click on it to open it) could be a simple picture showing a good base layout along with arrows showing the order to make the buildings.

Having the lessons would be a way for people to get a sense of accomplishment, and an easy way to blow off steam if they're annoyed about losing. If they take some of that negative energy, and put it into completing lessons for 30 minutes, they would hopefully feel better about coming back to the competitive part of the game and trying again.

The worst thing is when people take that energy and spend it on playing pointless games against AI, which only teaches the players to turtle and not to learn new ways to play. "1 human + 3 AI vs 4 AI" is not a way to learn anything. Puzzles could provide more satisfaction, with not a lot more stress.

Of course we would also plant the message that once you finish all the lessons, you SIMPLY MUST try the ladder. Like: how can you complete all these lessons and NOT get on the ladder? You gotta show off your skills! We need you out there!

Brilliant suggestions @arma473

I'd be happy to help with training, how do I volunteer?

I'm working on a 5 minute build order for Archsimkat's Valley for my friend. It's nothing fancy but it would be useful for any new player I think so I can use that as a basis for training people.

I like armas approach. Also I think jips rainmaker survival is already a great look at what tutorials missions for ladder could look like. Also I think the league system offers a great opportunity for rewarting people who finished their tutorial missions. They could just receive a certain rank level for completing them. And probably will also be able to play at that level when they finished the missions.

My low efford suggestion to making good tutorial missions would be to remove pd and stationary aa from coop missions. Not sure if that will help with player retention. But definetly better ladder newbies.

If you ever had to introduce new people to Supcom or FAF you would know exactly why. Me and a few friends have been trying to get our other group of friends into supreme commander. We play with each other almost 3 times a week (about 4-8 people) but only two of us (me and another guy) has played FAF multiplayer.

• The campaign teaches you everything you shouldn't do in multiplayer. Almost everyone in the campaign just turtles up and techs. Someone could've have played Vanilla supcom for 200 hours and have no experience fighting with t1 units. The biggest hurdle in FAF is unlearning everything singleplayer taught you.

• Vanilla Supcom FA doesn't have a basic tutorial since it's an expansion. The first mission just drops you into the action with 25+ units & buildings to work with. If you never played Supcom 1 you get nothing.

• The FAF tutorials are NOT HELPFUL AT ALL for onboarding new players. The first ones have you memorize map specific build orders, if someone doesn't even know the basics then memorizing a 15 step build order is intimidating.

• The learning curve is there, but it's exacerbated by the fact the knowledge is best found through outside guides rather then vanilla gameplay. Only someone already invested into Supcom will binge YouTube Supcom tutorials or read a 3 page guide. If you don't have a friend to play with, then you are stuck with the public lobbies. That's pretty intimidating for a new FAF player.

• New players don't know how big the game is supposed to be (partly because they play it like other RTS's). They need to learn it's ok to build 20 factories and to expand outward.

@biass

I think some sort of "casual matchmaking" could go a long way. It could also have an option for game modes other than 1v1.

@FtXCommando A lot of people in the thread mentioned how they were put off by multiplayer in one way or another. Getting kicked because no games, or getting wrecked repeatedly in ladder, or joining a high level game and getting raged at, or landing in a thermo / gap / astro loop. The point of adding singleplayer / coop scenarios is to give them something else that's less frustrating while steering them towards proper game style. This does lead to retention, because when you get frustrated by multiplayer you have more choices than either trying multiplayer again or playing another game.

More importantly, this is something that's obviously doable. An absolutely minimal implementation is adding a bunch of missions to the coop tab, that requires little more than designing, say, 20+ simple scenarios. A lot of effort, but straightforward. More stuff can be added to it - achievements, running a singleplayer game, a dedicated tab, matchmaking, etc.

Your proposal is more ambitious, but requires more resources and constant upkeep.

• Create cultural platform -> get someone (or multiple people) to maintain it, all the time.
• Find people with time / energy /cultural capital -> see above.

It's better, sure, but not relative to required effort.

@FtXCommando said in Why would you have left FAF?:

Can someone explain why playing AI training games retains players? It feels like you guys are conflating “trying to make players not lose” with “retaining players” because the latter is infinitely more difficult and basically impossible to artificially do.

80% of Supreme Commander (vanilla) at our 4-computer student association was 4v4 with humans vs. AI.
Exploring the game mechanics in Team vs. AI is both social and not too stressful.

Community is worth nothing If the only game experience available to newcomers is shit. How many people comment under youtube casts that they've watched a lot of casts, but are anxious about playing multiplayer? We should give them an alternative that's not shit for their first twenty games.

Nobody said that it has to be strictly "training" material either. Give them a bunch of tough scenarios that will take many attempts to beat as well.

Getting kicked because no games

No amount of training missions will fix that problem. It seems to me that many people are getting put of, because they feel they don't even have the chance to prove themselves in a real game.

A lot of the suggestions seem to revolve around the problem "it is hard to get good" but this is different to retention. We first have to make the player want to get good at this game, before he will even notice a lack of training material. (In fact it is probably better to have low rating, because then you get lots of players to play with.) Also you don't have to be very good to enjoy the game. Yes, you need to understand how you can control the game, but you don't have to be 1k to have fun.

I agree that we need to encourage building communities more and the clans are an obvious tool for this. I have been in this community for some years but I don't know what I would do to join a clan. Message random people with a clan tag I guess?
There is no accessible list of clans that states their purpose. What clans are open to noobs? are there clans with the goal to reach 2k together? are there clans for shitposting? I don't know. And we should change that.

Think of scenarios that teach a particular skill as tutorials and it makes sense. In modern games it's absolutely expected that the first few missions will be this kind of tutorial.

Then you stitch together all the skills you've learnt and suddenly you're halfway competent. Also, each step along the way gives a dopamine hit because the player can recognise that they now understand something new.

There could even be some kind of badge that shows up in lobbies so that other players know that this new player has achieved enough training goals that they will have a basic idea of what to do, which would mitigate being kicked for zero games.

In my opinion, some will stay others will leave and will be members who will play, leave, play again. We all must have in mind that this game is quite old but has something interesting that made me come back.
About the changes or improvements

• they must be countable
• as a rule is good to have targets, aims
• Also they must be done in a reasonable periods of time ex: 1 in 6 months and understand what is the achievement
• What are our target client? age? sex? location? all this count
• etc...
  It is very good that questions like this arise and they are discussed with community

Same as it was 7 years ago:

You join a game with 0-rated you get booted, ask "Why?", told to join "all welcome" games.

You play "all welcome" games to find beautiful -8 gameplay full of hysterical, messed up people who just shit on you for not knowing how to "play the game" on extremely strategic maps like dual gap and astro crater.

TMM was pushed hard by every councilor across the board (including myself) so there is a variety of games and at least some introduction to content beyond the aforementioned trash experience.

Community is a very large reason why many people stick around, if you turn up with a group, you're much more likely to stay, there's natural competition between members and the experience becomes more chilled. Problem is that a large majority of people who join are going to be coming in on their own, I'm with Ftx on this as my experience was similar to his. Mainly played ladder up to 2016 where I found the GB lot.

People don't usually want to improve to then commit to a game, they will improve because they're already committed, and that's why training material won't resolve this issue. Again more material is good, just doesn't help with this issue. The factors bringing people in are doing extremely well, Yuri, Brnk, Gyle and many others have put some serious hours into the community and to that they have my thanks.

Retention is a completely separate issue, you don't leave a game because it's "too hard" or because one person called me a smurf, you leave a game because either the community is the issue or the game isn't fun.

Now unfortunately we can't dictate what is fun to some people so naturally people will leave when the game becomes less entertaining to play, now, usually that doesn't happen for newer players, but since the point is players leaving before 2 years, it cannot be completely disregarded. This will hopefully be helped with the addition of tmm due to less games being dictated during the lobbysim, but the results will not be seen for a while, to see over 500 names on the 2v2 leader board gives me hope.

As for the community, it's not bad. It's leagues friendlier than league, CSGO and other high player count games but that doesn't quite cut it. It can be a little discouraging being a 0 0 to find nobody wants you, but hey that's where ladder and tmm help by removing hosts. Maybe a better clan system would help, possibly categorizing them into relaxed, beginner friendly, competitive, coop groups etc.

What you need to take from this is that no, training isn't the answer to retain players, as most people play the game to enjoy it regardless of how good they are, if any more evidence is needed, refer yourself to the last ladder league where lower brackets had the players with the highest time spent in ladder. What will probably help is a more helpful way to introduce new players, I know #Newbie channel works for people below 50 games but maybe having a newbie channel per major language (Russian, English, German etc) might help people find others in a similar situation.

@nine2

I feel like the most likely reason people leave is because they have other games to play. Some people want to try RTS but inevitably decide they prefer fps or mmorpg, etc. Every game has a learning curve and every game has new players, good players, and great players.

I don't really feel that FAF has alot of abrasive players; most people who play a game as old as this one do so because it's great and have an appropriate sense of ettiquette as a result.

I do have some suggestions based on my time here that I think would greatly improve player retention both for new players and experienced players:

1. Ladder matching i think would work better if instead of matching closely related ladder ranking, there are also matched based on number of games played. Maybe create a new players bracket where they'll only be matched with other players in the same bracket, maybe 0-99 games or 0-199 games. This will give new players to play against other new players, learn from each other, add people to their friends list, etc.

2. Instead of having ladder games launching every 3 mintues, I recommend having them launch after X number of players are queued up to play. The matching system can then more closely match rated players. Possibly it would be nice for the client to show how many people are queued to play(EDIT: I see in 1.4 this has already been added) / how many people are required for the games to launch.

3. Upon matching, I think it would be absolutely awesome if each pair received a popup of 3 ladder maps and given the OPTION to veto one of them. The system could pull a 5k, 10k, and 20k map from the pool, each player can then accept all 3 maps or veto one of them not to play. if both players choose different maps to veto, the 3 map gets played. If 2 or more maps are left unvetoed, the system can then randomly pick. This will allow players at both low and high levels to play a wider variety of maps. This will also allow people who don't have time to play a large 20k map to veto it to play a quicker game and vice versa. This would be a big boon for people to enjoy ladder more. I personally didn't play any ladder in december because I really didn't like most of the pool. (BONUS: devs can also use the information received about map veto vs draws to determine what maps people like vs don't like, thereby allowing everyone to essentially have a vote as to which maps to remove from the pool everytime it changes)

4. As for custom games, Morax has a very good point about people getting kicked from all welcome games. I might suggest required hosters to retype the game title each time instead of the client saving the name from the last game, although I can see some players saying that would be annoying. Sadly I have no other idea on who to improve custom games simply because of the nature of custom games.