What about "no tutorial"

[Pumpkin]

In this post I will partially disagree with what I said myself in Tutorial's thread. I had a sort of revelation and I basically changed my mind. I will develop a new approach of the tutorial, or more accurately I will develop a new idea for introducing new players to the game without a tutorial.

I was looking at all the ideas about the AI War's tutorial and a new version for 2.0. They are all interesting, but they all revolve around one kind of solution for the "game's first experience" (which is, for AI War classic, suboptimal, to put it mildly): the tutorial. But it's not a problem by itself, it's a "suboptimal" solution of a problem: the introduction of a new player to the mechanisms of the game. The approach being currently debated doesn't satisfy me anymore. However, don't stop debating, it sure can make a good solution in the end; I'm just taking another path that I believe is promising, but I might be wrong in the end. (For the League of Legend players out there: "keep your lines, I go jungle".)

Enough introduction; let's get to it. My point is: a scenarized game, linear, guided, babysitted from one end to the other, doesn't feel AI War. (Once again, don't get me wrong: this is a conventional method that might work well. Keep developing that idea.) But I feel that doesn't fit in AI War. I don't even feel it's the right method in other RTS games, but even less in AI War. My idea would be something that I'll call a "zero feature first game".

Instead of a scenarized "campaign" the new player would instead be proposed a simplified game, but a normal game otherwise. It would only use something like a lobby script and some special journal entries. No handcrafted campaign, just banal loggy options and journals. That special game setup would be completely feasible with the normal game's lobby. The first game would just have "zero features": no waves, no reinforcement, no special forces, no threat, etc, all disabled with advanced lobby option (automatically disabled by that "lobby script", of course). The player would just be presented with a 10 planets "simple" galaxy map at sandbox difficulty and any possibly disable-able options disabled. Also a couple of also legitimate options (why not?): infinite resources (metal and energy) and no AIP. The player would be tasked with "build an fleet and destroy the two AI command stations on planet $A and $B." A journal entry would quickly explain (after "stop reading now if you're familiar with generic RTS controls") how to put a space dock on loop, select and move a fleet and cross a wormhole. That's it. So the player build its fleet, move from planet to planet, destroy the two central AIs and win. In the meantime, he experiences by himself the very basic of combat: move a fleet toward a guard post ("See? Ships automatically attack when at range and get automatically rebuilt up to a cap"). No need to detail that in a journal entry.

The player wins. 5 minutes for a RTS player, half an hour for my mother, the time to figure out the selection, the orders, etc. Then the player is proposed to begin a new game with more features. He's told he can enable them all or one by one, depending on its liking. These features are resources (metal and energy), knowledge, AIP, waves, etc, also bigger map, up to a full, 7/7 vanilla game (5/5 first, sure).


I was thinking about how to further simplify the game for a very first "tuto-like" game, and I went into rethinking things that might even be great improvement for macromanagement and clarity of the game. If it's good for a noob, it's good for me! Let's clean the stuff around the OCStations.

* integrate the energy collector into the command station
* integrate the knowledge gathering into the command station (Only one version of science lab, stealth and used for gathering on neutral planets and ARS scouting, maybe even merged with the hacker. Researches are done in the UI.)
* integrate the redirector post into the command station (or maybe make it a UI-only thing, not a unit in the game)
* integrate the beginning's producers (cryopods and cities) into the Home command station (or maybe let them out as they are for lore and AIP², why not)
* make remains rebuilders automated: they would be OCStation's drones, unbuildable and uncontrollable, only and automatically dispatched when needed.
* better build rate for space docks to let them work without a swarm of engineers (I would even say "remove engineers, make a healer fleetships (between scout drone and triangle) and make self-building at better rate").
Well, this one is a bit extreme, but I'm sure there is simplicity for both management and new players to be found here. But that might be achieved during the test times, when the AIW2's engine is earlybetatested.

Ideally, a planet would have only an OCStation and defensive stuff (FFields, turrets, etc), and eventually one constructor building per unit family on home (or production) planet (space dock, SShip constructor, mercenary, warhead silo (plus the metal harvesters and eventually some matter converters at home (speaking of which, metal to energy conversion may become an UI stuff))).

Also ideally, capturing a planet would be just a matter of sending a colony ship, choosing a kind of OCStation, waiting (defending) while it finishes building and start placing turrets, FFields, and stuff (actual strategy, no "don't forget energy, don't forget science, don't forget redirector, ..." checklist.)

²: when cryopods and cities under the home FField dies and increase AIP, loosing is near. I don't think it's an important thing they are out there and have AIP on death.


Again, a good ol' tutorial might be the way to go. I'm just exploring an off-path potential solution.

[Draco18s]

This is right up the ally of what I meant by several linked, compartmentalized, "campaign maps."  The first one is stripped down and the player gets told how to build a fleet and go crush something.  Then they get told about waves, one happens, they defend.  And so on.

Small, easily accomplished tasks with obtainable goals.

[Pumpkin]

This is right up the ally of what I meant by several linked, compartmentalized, "campaign maps."  The first one is stripped down and the player gets told how to build a fleet and go crush something.  Then they get told about waves, one happens, they defend.  And so on.

Small, easily accomplished tasks with obtainable goals.

Sure. But the linear, handcrafted facet of it made me consider another option. But yeah, the goal is the same.

[Tridus]

Do the Civ games have tutorials? Might be worth looking around to other complicated games for inspiration on that front.

[PokerChen]

 Civilisation games have advisors, which are capable to turning the first ~100 rounds into a tutorial. IIRC, Civ 5 has a recommended tutorial game in which you are placed in a 2-player continent where the AI does very little, I think, until you finish. However, these advisors do have faces and voices - that's a lot more engaging than a text box (even if its mildly-annoying, the former is better).

[chemical_art]

I am still strongly against any compartmentalized tutorials, for it would fail at conveying what the rest of the game is about. I feel like a directionless tutorial, no matter how simple, will work because new players will either be overwhelmed or bored and that is a failure of a tutorial. A tutorial should teach, but also motivate a player to actually play it.

What I am for is the automation that is presented for colonizing planets as (the default) option that of course could be disabled. Automating as much as possible is desirable in my eyes.

[PokerChen]

This topic just reminded me of EVE University. A game that's so deep you need people to teach it, who can teach using multiple ways. It starts to be influenced by how different people learn differently....

*palms face* did you people learn AD&D type tabletops?

[Misery]

One problem with creating an ultra-simplified first game as a teaching tool is that players tend to get more than a little dumb about these things.

Like what happened, over and over and over, with Bionic Dues:  New players would start the game, play on Normal (which is WAY too easy to be called "normal" difficulty), kill basically EVERYTHING in a single hit with no tactics or preparation, declare the game to be "shallow and easy", and never touch it again because of how bored they got.   That it might have depth that they didn't see because they didn't take time to go further never occurs to them.  I watched even dedicated strategy-gamers do this.  THEY, logically, should be the ones to know better, but I still watched them do this.  They don't think "There are harder modes.  Maybe that changes things up and reveals the depth". 

Give the player a super simple first run, and that's likely to happen again.   The current tutorial is simple enough, but it still gives that feeling of complexity that can help to draw in the players that are the likely sorts to want to play the game.  Where it fails is explaining the really ADVANCED stuff; it leaves that stuff so confusing that it ends up pushing out the players anyway.


.....also, players are bloody lazy.  Learn by experimenting with stuff and trying things for themselves?  Hah!  Nobody does that these days, which is sad.  Which actually is part of the problem even with the current game:  Players hit the end of the tutorial and have no clue where to go next, so they don't try to go anywhere at all.  They're used to games telling them to press A to jump.  Someone in another thread even pointed out some percentages related to achievements, showing that the number of players that even completed a single game (or even lesser milestones) is frightfully low.  Sure as hell is a lot lower than I ever would have thought.  The tutorial that's there may teach the basics, but when that's over and the time comes to learn further on their own, circuits get blown.


That's not to say an idea like this CANT work.  Mostly what I mean here (and what I may be phrasing entirely wrong) is that player psychology interferes with something like this.  Just like it interferes in other ways with most other current ideas.

Overall, this sort of thing is just going to be very difficult to deal with.  How can anyone make it a streamlined experience to learn a game that is essentially the Dwarf Fortress of strategy games?  That's going to be quite the task.   That being said, any ideas at all that anyone might have, let's definitely keep them coming.  Maybe what's needed is a combination of some ideas like this?  I don't know....

[Kahuna]

it fails is explaining the really ADVANCED stuff; it leaves that stuff so confusing that it ends up pushing out the players anyway.

I'd like to know what exactly are these advanced and complex things that people talk about. What's considered advanced and complex and hard to teach to new players? If we figure out what exactly is the obstacle it would probably help with designing the tutorial.

[Misery]

it fails is explaining the really ADVANCED stuff; it leaves that stuff so confusing that it ends up pushing out the players anyway.

I'd like to know what exactly are these advanced and complex things that people talk about. What's considered advanced and complex and hard to teach to new players? If we figure out what exactly is the obstacle it would probably help with designing the tutorial.

Hmm, interesting question.  Probably going to require responses from a bunch of people as I suspect it differs from one to the next.

For me, one of the biggest issues was getting an understanding not so much of what everything DOES, but what everything is FOR.  There's TONS of different units and structures in this game.  So many.  In other RTS games the relationships between them tend to be relatively simple.  The whole rock-paper-scissors idea, and usually not all that much else, particularly as those games are so much about micro more than anything.  AI War has that R/P/S aspect with it's most basic units, but then dumps about a squillion other things on top, each of which has it's own use/purpose.   Actually getting a clue as to what some of those purposes are though is something I had a ton of trouble with (and still do).  Getting a unit/whatever I've not used much before, reading the info on it, and going "Okay, so this does some funky things... and I can use this for what, exactly?  What strategic benefit is generally associated with this thing?"   There's all sorts of different things, so very many, and lots of them are quirky or do a variety of strange things (which other things then may have immunity to, which needs to be considered in figuring out that thing's "role"), and so on.  Just learning how to really use all these things adds to the big confusion in the game.

Secondly, advanced "tactics".  The sorts of complicated moves that not only can be a bit confusing, but that the player may not even know exists.  Or if they do know about them, maybe they know too little to get even an idea of how to get started with that tactic.  Something like beachhead attacks, for instance.  Confused the hell out of me for awhile.  I knew OF it, but getting an idea of how to even go about it, or what it was best used for (without having to resort to guides of any sort, which is important here), was bloody baffling.   And that's just one example; there are others. 


That's not to say that the game needs to explain every tiny little detail of every tiny little thing... of course not, that really couldn't be done.  But it doesn't even remotely help guide the player into a position where they can even start working towards an understanding of any of this.  And pure experimentation can be very frustrating/irritating in this game, and often doesn't really lead to info by itself (at least, it didn't for me, in many cases).

[kasnavada]

it fails is explaining the really ADVANCED stuff; it leaves that stuff so confusing that it ends up pushing out the players anyway.

I'd like to know what exactly are these advanced and complex things that people talk about. What's considered advanced and complex and hard to teach to new players? If we figure out what exactly is the obstacle it would probably help with designing the tutorial.

I don't think there is any "advanced" stuff in AI war. However, there is a lot of different mechanics, which combine and make monsters out of simple problems.

One problem with AI war is the "this game has a high learning curve fallacy", which in game forum generally translates as "3l4t3" talk "there is no tutorial", and "the game is designed so it requires you to know the entire game before starting it". Does not mean that the game is more complex than the next. The fallacy is that not explaining how something works does not make it more complex. It just adds frustration time to figure what should be told or written somewhere.

For example, how fun would be chess if you weren't told of how the pieces move, and then we add a rule that says "if wrong move, game over" ?

What AI war really requires is to introduce and explain more gradually mechanics as the game unfolds - and that's enough to win "easy to normal" difficulty games. Then, for harder game, you'll have to discover how to combine the simple mechanics you've learned into something much more efficient. In game concepts simpler to understand, what I mean is :
- "you need to expand to reach the AI core" => taught by the game when learning game objective.
- "you can abandon worlds if they're too dangerous for you" => taught when capturing a science station in a "out of the way" place.
- "hey player, you need to defend your worlds" => taught by the game.
- "if I build a chokepoint..." => advanced concept, because in combines all 3 before with the meta concepts of exploring and planning your expansion.

Basically, like chess. A good teacher will at least tell you how the pieces move, and a few opening, and a few mistakes to avoid during game start. Then, as you play, he'll introduce other concepts. Last, you'll have to figure really advanced tactics on your side.

I watched even dedicated strategy-gamers do this.  THEY, logically, should be the ones to know better, but I still watched them do this.  They don't think "There are harder modes.  Maybe that changes things up and reveals the depth". 

.....also, players are bloody lazy.  Learn by experimenting with stuff and trying things for themselves?  Hah!  Nobody does that these days, which is sad.

Well, a lot of people don't find learning nor trying fun. And, a shitload of games / mechanics are so badly designed that I completely understand how they feel. I don't count the number of games where I've run myself into a corner because, when learning the game, I made bad choices that ended up being critical mistakes in making the first game fun, and nothing, absolutely nothing in the game pointed to that being an idiotic choice. So... yeah, when I see a game like that, I don't usually scrape the bottom of the s*** that was the first playthrough to find whatever minuscule nugget of depth that's there.

Actually, having a choice in a game, and finding out that it's a bad one - it's ok - if somehow in the meantime you can recover from it. Finding out that because you did something 10 hours ago, it's now IMPOSSIBLE to have fun, and you have to restart ? That repulses me.

In games like AI war it's even worse, because when I lose 15 minutes in a roguelike and die because of a stupid mistake, I lost 15 minutes, and it's expected. In a 20 hour long game in normal difficulty ? I expect to win (yes, I've been playing games for 25 years now. I can beat something on the first try on normal difficulty...). If I lose because I took a wrong turn that nothing warned me about, and with consequences very far down the road that I can't adapt to ? That means I spent the better part of a week getting there, and I have to restart because I made a mistake 15 game hours ago ? Worse, it's because the game dev couldn't be arsed to make a tutorial to cover what he put in his game ? How can that be called fun ?

So, no. I disagree with the "players are lazy" mantra - some are, but mostly... A lot of game devs are lazy in designing their games, and especially lazy in making a decent tutorial. And they hide behind the "it's complex to learn fallacy" to justify what they've not done.

[Coppermantis]

One problem with creating an ultra-simplified first game as a teaching tool is that players tend to get more than a little dumb about these things.

Like what happened, over and over and over, with Bionic Dues:  New players would start the game, play on Normal (which is WAY too easy to be called "normal" difficulty), kill basically EVERYTHING in a single hit with no tactics or preparation, declare the game to be "shallow and easy", and never touch it again because of how bored they got.   That it might have depth that they didn't see because they didn't take time to go further never occurs to them.  I watched even dedicated strategy-gamers do this.  THEY, logically, should be the ones to know better, but I still watched them do this.  They don't think "There are harder modes.  Maybe that changes things up and reveals the depth". 

Give the player a super simple first run, and that's likely to happen again.   The current tutorial is simple enough, but it still gives that feeling of complexity that can help to draw in the players that are the likely sorts to want to play the game.  Where it fails is explaining the really ADVANCED stuff; it leaves that stuff so confusing that it ends up pushing out the players anyway.

I think this can also be helped by a good trailer. In Crusader Kings II, for instance, there are obviously a great many different mechanics, many of which aren't particularly complicated but there are a lot of them, and you're going to spend some time learning about what everything does. It has a compartmentalized tutorial with some of the disadvantages chemical_art mentioned - it has a bunch of basic/advanced/expert level tutorials for various functions, which explain what certain buttons do, but fails to really tell me how I'd use these tools to execute the grand strategy one would expect from the genre.

However, the trailer is, I think, a very good one. It actually shows a scenario playing out in a narrative fashion,without any technical details that a new player wouldn't be familiar with. This made more eager to actually explore the mechanics, since I knew from the trailer that it was, in fact, possible to do some really cool stuff with the tools the tutorial teaches you about, and that motivated me to spend more time delving deeper in hopes of experiencing that kind of story for myself.

That's of course not an excuse to have a lacking tutorial. AI War's was decent, I think. Walking the player through a mini-campaign is probably the best way to teach them, I think. On a smaller map, you don't get as much of the advanced strategies like making choke points through gate-raiding and the like, and the core philosophy of taking as few planets as possible isn't as important, especially on a map as tiny as the tutorial. So, while a compartmentalized tutorial has disadvantages, I'd think at least two would be ideal- a "mechanics" tutorial that just teaches the tools in a sandbox setting and a "strategy" tutorial that walks the player through a simulated first campaign as if they were really trying to win with a standard setup.

In particular, I think that starting the players in a regular campaign with so little guidance and especially with something as important as AIP disabled is a bad idea. You risk letting them think they can play this like a traditional RTS and teach them bad habits, which can lead to the "I lose twenty hours in because of a bad opening move" problem that Kasnavada talked about. Even with it enabled, on a tiny map, they won't be able to fully understand its significance.

[Misery]

In games like AI war it's even worse, because when I lose 15 minutes in a roguelike and die because of a stupid mistake, I lost 15 minutes, and it's expected. In a 20 hour long game in normal difficulty ? I expect to win (yes, I've been playing games for 25 years now. I can beat something on the first try on normal difficulty...). If I lose because I took a wrong turn that nothing warned me about, and with consequences very far down the road that I can't adapt to ? That means I spent the better part of a week getting there, and I have to restart because I made a mistake 15 game hours ago ? Worse, it's because the game dev couldn't be arsed to make a tutorial to cover what he put in his game ? How can that be called fun ?

Aye, this is a pretty good point here.  Nobody likes having that happen.  No matter what game it is, it ends up feeling like a waste of time.   I mean, it's one thing if it's something like you're doing fine, but the enemy starts fighting harder, and you're slowly overrun, just because you're simply not keeping up with them very well strategically.   But if it's because you made ONE mistake in a game that presents you with 10 squillion choice-making moments over the course of 20 hours?   Yyyyeeeaaahhhh.... that's a problem.   I've seen that happen in this game, definitely.   Hell, one time I lost a game, I don't remember exactly WHAT it was, but the game threw something at me that I hadn't ever seen it do before.  Some lightning-fast thing that plowed straight through to my core planet without bothering to do much to the other worlds.  And my reaction is like "WTF?!?  What the heck is that?  What was I supposed to do about that?  I've never seen that before!" and then.... BAMF, it's over.   I don't remember how far into that particular campaign I was at the time, but it was far enough that it was REALLY irritating.  I didn't know about this thingamajig (whatever it was, this was awhile ago, details very fuzzy) so there's no way I could have possibly known how to prevent it;  I had plenty of defenses and such, but obviously not the RIGHT ones for that specific situation.   But the game never explained anything like that to me.... what was I supposed to do?

Which I suppose is what I mean by "advanced" stuff.   The sorts of more quirky units/objects in a game.  It's not that these elements are all that much more complicated/hard than anything else, but maybe they're a bit obscure, maybe they don't show up all that often, or maybe they're confusing somehow... one way or another, they should be explained somehow, is my thinking.  In something like a fighting game, for example, animation cancelling is considered an "advanced" mechanic.  It's not something that you need to know to play the game in general; it's not part of the absolute basics.  If you're playing in a more casual sense, or are just still new to it, playing on a low difficulty, you don't need to know it yet.  But if you want to play at a high level?  THAT is when you need to know it, but it's something that beginners are also going to have a downright hard time with; just flinging it at them at the very start isn't going to help them learn it.   Fighting games though pretty much ALWAYS explain these things nowadays, in split tutorials.

 But how do you do that in a strategy game with a zillion different units?  Where players feel overwhelmed by complexity from the very start?

[Tridus]

This topic just reminded me of EVE University. A game that's so deep you need people to teach it, who can teach using multiple ways. It starts to be influenced by how different people learn differently....

*palms face* did you people learn AD&D type tabletops?

Yes. I do not miss THAC0. 

[Draco18s]

I got into D&D post-THAC0.  Well, I did read those rules, but I never played a game until 3.5