Auto-Play

Here’s an interesting game design development that I haven’t seen discussed here yet.

Auto-play. This is an increasingly common feature where you can turn on a mode where an AI plays for you. You become a spectator (at least, for the duration of that encounter/battle/level/whatever).

It sounds ridiculous, but apparently a lot of people like it. See this blog post from last year, Entering the era of Auto-Mode. And here’s a thread on Auto-RPGs at TouchArcade.

I guess this supports my thesis that fun is about a sense of accomplishment, and not about difficulty. In at least some of these games, it sounds like you can play pretty much the entire game in auto-mode; there is no difficulty whatsoever. But you still get to name your character, watch him level up, buy better gear with all the loots he gathers, etc. So you get a sense of accomplishment. Not fundamentally different, I suppose, from farming/gardening games where you just click to gather stuff, and then click to plant more stuff. (Except perhaps with less clicking.)

But man, this is generating some fairly heated debates in the player community. I don’t see anybody passionately in favor of it, but a lot of players saying basically, “You know what, it’s surprising, but I actually kinda like it.” And then I see people who hate it vehemently; they argue that if the game has an auto mode, that’s basically admitting that the game design is so shallow there are no interesting choices to be made. Not a game at all, but more like watching a movie, so they say.

But on the flip side, gamers these days are more casual and multi-tasky than ever before, and maybe a game that plays itself is just what they need. You can still feel like you accomplished something, without all that pesky having to pay attention and do stuff.

What do y’all think? How (if at all) should we be incorporating auto-mode into our games?

3 Likes

I imagine most of this is from the tradition of JRPG combat that isn’t engaging. Nine times out of ten no decision is being made, and you are just hammering confirm until everything is dead. In that sense, combat isn’t really important. The more important decisions are all made outside of combat.

The only real problem with it is that in a sizable portion of the gaming populous, combat has been conflated with being the meat of games. Hell, just about any game that doesn’t have combat these days usually has to deal with “this isn’t a game” arguments.

I feel like there was a game (about ten years ago, maybe flash) where the idea was that you didn’t actually do anything in battle, but the point was to figure out how to gear up the party right for the fight. I could be completely making this up though, because I don’t really remember it well.

2 Likes

I used to “play” Progress Quest a bit “with a friend” when I was in school. That game didn’t even have graphics, just text, numbers and progress bars I think. Between the two of us it basically was a competition who could manage to keep it running for longer throughout the days.

I can see the appeal, but I’d never ever buy a game that doesn’t have any decisions to make.

I think the concept can have a place in a game if it is thematically fitting within the context. E.g. a management game where you build a fleet of robot servants that are deployed into homes of rich people in some future society where they work for you and generate an income that you can use to upgrade and expand. There could be decisions to be made as much as you want (some of them just being cosmetics like designing your bots, company logo, your home etc.) and the game could be designed in a way that just never anything goes too wrong and if you idle away you basically just earn money (at a reduced rate compared to when you actively micromanage) to then do something with when you want to interact again. The challenge in the design would be to still make the game compelling to play and not feel pointless. If it’s straight up “this game has no interaction and plays itself” I doubt many would consider buying it. I’d imagine there could even be quite a lot of outspoken backlash.

1 Like

Auto-play is brilliant, because it allows people who don’t have time to play games to play games because of auto-play, simple as that. It allows your game to become more grindy and longer because it is auto-play which is keeping your players playing longer. It hides away the problems of “grinding” and “boring” stuff to the user because you can “auto-play” through it. But most importantly you don’t ever allow the AI of auto-play to be smarter than the person as you want auto-play to lose against bosses or any new areas as it is only designed to get rid of the grindy stage, but favour the person to “jump in” at any time to beat the boss. Yes the game fundamentals should be balanced around the auto-play, that you can turn on or off at any point of the battle this is key.

Without auto-play I would not have played the mobile games that I do, and same goes for my father or girlfriend because auto-play exists.

P.S. I just implemented it in my game and it is amazing to watch. :slight_smile:

1 Like

I was about to say that this is really about mobile games, but I remember trying Eve Online for a few days. I got a surprising number of kills on people who happened to be watching a movie or something while they ‘played’ and didn’t notice me shooting their ship. It was basically the only way I could blow up other ships, since I was a newbie.

But wasn’t this idea really explored and discussed in depth back when the WII first came out and everyone decided that nobody wanted games you could die in anymore until Dark Souls reassured us that some players still like challenge? This was one of the reasons that Dark Souls was such a big deal. I believe that people were discussing semi auto play features - where the game would like automatically beat whatever challenge you were having trouble with or something.

Sports games have had this sort of feature for at least a couple of decades. It’s very entertaining to watch, and although the Madden/Live predictions aren’t always accurate to the real-world events, it also provides a way to stress test your AI by pitting it against itself.

Isn’t there also a setting in Quake/Unreal that just throws a bunch of bots in an arena? Was also great spectating.

1 Like

I like to approach JRPG auto-battling from a different angle - I think auto-play in JRPGs is actually a symptom of a design problem: shallow combat systems that offer few interesting choices.

I like to hold up Vivi in Final Fantasy IX, actually, as codified in Rule #147: Vivi’s Spellbook Principle which reads as follows:

People consider JRPGs boring, because you make nearly no meaningful choices, ever, the game just uses a Skinner Box mechanism to keep you playing (gotta gain another level! I might learn another useless spell!)

JRPGs are the unholy hybrid* of the Visual Novel and the tabletop RPG (for reference, Dragon Warrior and Final Fantasy, the first two JRPGs ever.) The problem is, the combat system is the place where you could theoretically make meaningful decisions…except, that’s a rarity that often comes up in gimmick battles.

Complicating this, is that we’ve had super-easy Mook battles in JRPGs for so long, that a JRPG like Lost Odyssey gets interpreted as being too hard, when in reality it’s difficulty curve is merely ‘above average’ like the NES RPGs. Alternatively, if you were to use interesting mechanics for every mook, you’d soon have nothing left for the boss characters (I recently read a thread about boss fights on this subforum, so that’s another topic I won’t dip into.)

While I disagree with the prevailing train of thought in the modern games industry that JRPGs are a dead genre in the “they won’t sell anymore sense”, I will concede that they’re dead in the “we can do new things with the format” sense - JRPGs are a well-explored space that have strongly established conventions that serious efforts at breaking will cause serious backlash - some of this is why Final Fantasy XIII failed, actually. That and the god-awful level design.

I think to avoid this, you have to have more clever mechanics in Mook battles, but more to the point, it’s a better idea to de-emphasize combat, like later versions of Dungeons & Dragons did with skill check scenarios that allow for XP gain. Battle-heavy JRPGs are so well-explored now, you’re more or less not going to make an interesting battle system that offers clever choices, but you can give the player a feeling of agency if you let them do something that feels clever and masterful.

TL;DR - If you’re implementing auto-battle, maybe step away from battles, because they’re uninteresting, and add something that is.

JRPGs are actually my favorite genre, and I truly love them. That doesn’t mean I won’t winge about their shortcomings, however.

1 Like

Is auto play just the next step from twitch? Or is it for people that can’t actually stream twitch?

“New AUTOPLAYYYYYYYyyyy. Why watch others play when you can watch your own game being played for you!!”

…I absolutely love the idea. Just throw a Cruise Control button on your game and say, “Are you sure you want to enable Auto Play? Can’t make any guarantees that you won’t die…” It won’t work for all aspects of gameplay. But the truth is, there are a hell of a lot of programs out there that bot games already. From a programming standpoint… what an interesting little challenge. Endlessly interesting. We already have an idea of how our games should be played… but to actually dictate the play? Granted, the game can still be resumed manually. I am currently making a platforming game and a resource/building game and I can’t see how you autoplay a platformer BUT I can see how you could autoplay the little war game and watch it all unfold with a can of soda. You could even fast-forward if it was all strictly autoplay! How weird would that be?

Make your settings, fast forward to see what happens, rewind, change some settings, run again.

Make it an experiment or something so that you can only rewind 3 times. When you rewind you press start & stop for the first one. For the others you press start & stop but it will automatically stop at half the rewind period of the preceding one. That way you can’t keep rewinding to the start.

I guess you could argue that one of my favorite mini-games, Cooper’s Little Adventure, is an auto-play game, since you don’t have to jump or direct your flying for the coins, especially once you get a few upgrades. It’s not long before you’re mostly watching it, just as with any idle clicker game.

But just as with those games, the real game isn’t the jumping for coins — it’s the upgrades. You have real, meaningful decisions to make about which upgrades to buy and in what order. And of course you have the whole massive development curve that produces a sense of accomplishment. But I think that accomplishment comes only if you feel you have some agency in making it happen; thus it’s important that you chose those upgrades. If you took Cooper or Cookie Clicker and had an auto-mode for the upgrades too, so you literally had nothing to do but watch, I don’t think it would produce a sense of accomplishment at all.

Perhaps this is a key insight… a well-designed game with an auto-mode simply isn’t really about whatever you’re automating; the real game is the other stuff you have to decide.

@Azmar , as the person in this thread who appears to have most experience with it, does this fit with your view of auto-play, from a game-design standpoint?

I don’t believe this game would fit with auto-play, because it literally plays the entire game for you and in this case the auto-play would probably be better than the user. It’s only suppose to support and abstract away the boring parts of your game to keep ppl playing, not literally be the game. As the key part is to always allow the option to turn it off as it would never be as “smart” as the user when it comes to any real actions in the game. I don’t think this concept would fit in any platform game.

@AndrewGrayGames Yes the combat system naturally only offers a few choices for the user and those choices are generally properly made to support the system. But a few choices in a strategy system does not make the game shallow, that is like saying League of Legends or most moba games are shallow because you only have 4 skills, heck even darkest dungeon! Also more choices does not give the game more depth either. The brilliant part with a few choices (4 skills) is that it’s easy to understand for the user, and this allows you to focus on a battle system that actually holds depth to keep them interested (like darkest dungeon), I would argue you can make a more balance able system this way by having constraints. It’s weird you may lose more people or people may become less interested easier because you have so many skills in your system and its not very “new player friendly”.

The system I am modelling even even went to the point of having only 6 simple stats in the game, and they even removed “dodges” in the game. Literally they removed so many complicated parts and made it so simple, and yet the game has more depth than most MMO’s I’ve played in the last 8 years. It left me in shock for a long time when I realized this, and changed my whole view of making games.

I really disagree, although it certainly depends on specific games as JRPGs have a huge range in combat systems, there is actually far more tactical depth and decision making in most JRPGs compared to their western counterparts. The reason for auto combat is the raw volume of those combats and their lopsided nature: that you should easily win most if not all of them.

After doing massive amounts of analysis of many games, I think the conventional wisdom on how JRPG mechanics are weak, stale or boring is not remotely correct. There is way, way more innovation, experimentation and risk taking in JRPG combat design than western RPG. If you really want good or unique ideas on turn based combat, JRPGs are the place to look.

The problem is that JRPG combat tends to be more of a grind, not a lack of depth, but a lack of challenge.

1 Like

Cooper’s Little Adventure seems to be what you would call an “Idle Game”. You can get more coins by jumping, or you can simply accumulate coins passively while idle. These games are also Resource Management games.

Exactly so you accumulate passively (this is the auto-play), and creating better auto-play mode with jumping etc and now the user never needs to play or even exist.

Well, certainly the battles themselves are simply battles of numbers and not so much of actual skill. You could determine the winner in milliseconds, even run a best 2/3 with randomized combat styles and decision making within. An auto-battle system could free the player up to actually do other things while the battle is taking place, or simply allow them to skip battles entirely so they never have to leave the Overworld map. You just walk into the enemy and you could play the sound of swords clashing (short ting) and then if you win they get tossed backwards and fall to the ground and you can walk right over their dead bodies. You can still monitor your party’s HP and MP, heal and use potions and stuff. All the fun stuff without the grind.

Granted, you might be able to battle more effectively than the auto system. This seems a way we can have jrpgs in the 21st century.

Just another note on the JRPG auto battle thing. I think people are getting this example wrong.

A lot of games that have a strict transition to a tactical mode from a strategic mode, especially with very frequent combat have historically always offered an ‘auto resolve’ mode. This can be seen in classic games like Jagged Alliance 2, through modern day 4x like Age of Wonders 3.

In games where combats are frequent and often pit imbalanced combatants with a strict mode switch for combat, you will see auto battle options. This is very different from something like clicker games or progress quest, which are designed to play themselves, instead of offering the option to the player to ‘skip a boring, repetitive, meaningless’ battle.

I won’t get into a massive debate on wither JRPG or Western RPG is more guilty of this, what I will say is that JRPGs have an order magnitude more variation in their systems than their western counterparts.

Western RPGs are generally built around the idea of character customization, self expression and almost always root in one of two very clear combat modes that have existed since the dawn of gaming. There is little change in western turn based combat since early Ultima games in the 80s.

1 Like

Yeah, I’m “playing” while replying to this thread in another tab.

Neither Jagged Alliance or Age of Wonders are JPRGs?

That’s the point. JA2 and AoW offer the same auto combat option, with the same goal as often offered in JRPG: to skip boring or monotonous fights.

This is not about a game playing itself, it’s about skipping over boring or overly repetitive content.