A game where the best way to get better is by playing.

Discussion in 'Game Design' started by ratxt1, Apr 13, 2012.

  1. ratxt1

    ratxt1 Well-Known Member

    I've been thinking about what makes me want to try to get good at a game, and why I am so put off of trying to play most games well, even though I enjoy playing them casually and watching high level play(chess, SC2, SF are all good examples). And I think I have figured out why, and what would make me want to get better at a game. The answer is right in the title of this thread.

    A game where the best (or a good) way to get better is by playing seems simple enough, but I am hard pressed to find a good competitive game where this is true. I know friends who love the Age of Mythology series have played all the campaigns, and still haven't figured out that they should make more than 10 workers. And this happens in almost all competitive games, if you want to get good you have to actively pursue it, which brings me to my last thought.

    What would a game look like where the design goal is just that, to make an enjoyable competitive game where you can get good while playing. I have some of my own thoughts but wonder what other people think (the best thing I have so far is something along the lines of make the game super intuitive and at the same time super complex so that if you try to do the standard brute force way of getting good you are met with the complexity where as when you play you don't meet this because your intuition kicks in).
  2. rozencrantz

    rozencrantz Active Member

    I think that one problem you will run into is that people are information hoarders. Even if playing your game is a good way to get better, people will find things to learn and train when they aren't playing.

    While there are games that you can get better at just by reading about them (these are the games I like best) I think that the majority of games are as you describe, with one modification: playing consciously is the best way to improve. A lot of stuff I've read about while learning games has not made sense until I saw it happen a few times and figured it out myself. This was even true with classic "training-mode" things like combos.

    Training is of different values in different games, that I can't deny; I think that Arimaa is probably the closest you can get to what you are talking about, because it has the balance of an intuitive system with complex grounding that you described. If you don't play consciously though, if you aren't watching to see the consequences of your decisions, and trying to figure out what impact past decisions had on the current situation, then yes you can play all the campaigns and not learn much of anything.
  3. ratxt1

    ratxt1 Well-Known Member

    I would like Arimaa if it didn't take 3 hours. I do like abstracts in general more than most on this forums, and I know you are even more into them than me. My major problem with them is length. Spending tons of time on very cerebral activity is just exhausting and often takes the fun out of it.

    A game I had envisioned that I think would fit this requirement, was sort of a physical version of Toribash (rules would be simple but building it would be hard and above me)
  4. Logo

    Logo Well-Known Member

    Making a solid competitive game is hard. There have only been a few successful games to fit this criteria in all of history. Even more so if we restrict it to video games.

    Making such a game in a way that also has the spare time/budget to reinforce good habits is another level of difficulty on top of that.

    The only reason modern games don't fit the bill is because they don't offer performance feedback beyond win/loss rates.

    For example, A game like SC2 could immediately start the replay after the game ends (like Super Meat Boy does) to promote people to watch the replays and see what went wrong. That'd be one small step, but it'd go a long way.

    You could also do stat tracking and display on the home screen something like the top 5 stats (things you do well) and bottom 5 stats as compared to opponents you've played and a top player sampling. Examples would be things like "Avg # times using mouse when hotkey is available/game", "Average # of workers/game", "Average mineral float", and so on.
  5. Delha

    Delha Active Member

    I think a big part of the problem is that this is every bit as dependent on the player as the game itself. I'd go so far as to say the player is the more relevant piece. Some people will naturally analyze their own play and consider where they can improve, but many more will just go through the motions on autopilot. As rosencratz put it, they're not really playing consciously.

    Even for those who are actively trying, mistakes in valuation are practically inevitable. Without external input to break these misconceptions, one can very easily hit a performance plateau without ever understanding why. Even with external reference, I'd consider it more likely than not for a player to get stuck with at least a few of these bad habits.

    From there, the problem on the game end of things is that ease of skill acquisition is almost invariably going to be inversely proportional to depth. While you may figure out a good number of things on your own, a really deep game will -by definition- have a vast amount of potential to explore. That increases the likelihood of any one person overlooking a possibility (and thus never learning it from their own play), or developing one or more of the performance impacting misconceptions described above. The most obvious way to avoid that pitfall is precisely what you're trying to avoid here: pooling knowledge.
  6. rabid_schnauzer

    rabid_schnauzer Well-Known Member

    I call hogwash on the entire premise of this thread.

    Whatever *it* may be, you can get better both by doing *it* and by cross training in related activities..If you want to reach the top level of any competitive *it* you will have to both do *it* and cross training in related activities.

    You can't get to be a chess grandmaster without reading books, studying historical matches, doing problems and learning about chess computer AI - you can't get to the top just by playing chess. You're can't get to be an professional athlete without running laps, doing conditioning exercises, watching film of the competitors and learning from your coach - you can't get to the top just by playing your sport.,

    Heck, take running - something most healthy 3 year olds can do, and a sport where most of the training is just running itself. But if you want to actually compete in running events, you don't just run - you learn and apply things about hydration, breathing, stretching, diet, intervals in order to maximize the impact of your training and you need to have a strategy for when to stay with the pack and when to try to break away for the lead or use your kick to make it to the finish line.

    The benefit of cross-training is so fundamentally true of any competitive activity that trying to design to prevent improvement via cross-training is worse than futile. Whatever you invent, people will relate their other experiences to it, and if any of those experiences provide a competitive edge, they will become part of the training regime for top-level competitors.

    And that's why we have contradictory design goals before the end of the first post, "make the game super intuitive and at the same time super complex"
    Xom, Jobber, Leartes and 1 other person like this.
  7. Waterd103

    Waterd103 Well-Known Member

    Well, He ask if it's possible to create a game where getting better in an activity that is not playing, is not possible.
    I have the answer: No.
    zem likes this.
  8. ratxt1

    ratxt1 Well-Known Member

    eh I will refrase me statement a little since I see it was extreme. My point was that to get better at a game it seems, the best way is almost never to play the game but instead to read about the game to memorize certain aspects of the game to practice excecution of the game. Heck all i want is a game where I don't have to do all that (or not as much at least) to get good.
  9. Waterd103

    Waterd103 Well-Known Member

    The main way to get good games is to ....play them. So you confuse me.
  10. Kai-

    Kai- Well-Known Member

    The best way is to play the game. But you need to be focused in your playing such that you're playing specifically to improve - improvement beyond a mediocre level doesn't happen automatically.

    As to why using outside resources (replays, strategy discussion, etc) is necessary is that it's just not feasible for one person to be able to figure out everything they need to know about the game to be successful.

    Nothing stops you from getting good to a point on your own of course, but using outside resources just makes you much better much faster.

    So I guess the answer is to just do as much as you're comfortable with and otherwise play the game.
  11. Kayin

    Kayin Well-Known Member

    A competitive game that one person can learn and analyze and master with no outside information sounds like a damn boring game. There is also the fact that in most games, the best way to improve IS to play. Study and training mode or whatever need to be applied in-game for it to really take hold. The more you play and understand a game, the more you can apply outside resources.

    Also this whole concept kinda pisses me off. If you don't want to put any work in, why do you deserve to be good? Why should you be rewarded? There is a whole conversation to be had about what things are worth having players learn and what not and the hardest to learn game is often not the best game, but when it comes down to it, getting good at ANYTHING requires work. And there is a very good reason you can't get good at games by 'just playing them' and that reason is that other people are better and more motivated than you. If no one consulted outside sources, you'd do fine learning a game on your own, but that's not it. People are willing to work hard to be good at things they love. You can ease the learning curve, streamline things, make things as accessible as possible but at the end of the day, you won't be good at anything unless you're willing to get a little dirty. Don't say "I'm willing to work hard! I'll play the game a lot! That's hard!", because that's not how that works. Learning to be great at something does not involve only doing the parts of it you love or are directly fun.

    These are competitive games. You're SUPPOSED to lose for not studying and practicing hard enough.
  12. deluks917

    deluks917 Yomi League 1 Champion

    Yomi is a good choice imo!

    I think you basically want a game that has not had much work put into it already and no execution barriers. Even if you wanted to read a detailed work on how to play Yomi (I certainly would) you cannot - no such book exists. friiik is clearly the best or almost the best and he says he does not really watch replays (idk if he watched them when he learning the game).

    note: I don't agree that "playing is the best way to improve" without extra qualifiers. Playing more is definately a terrible way to improve at Chess or Street Fighter. Also I have never seen a game where the best players were the people who played the most (or even remotely close to the most).
  13. Kayin

    Kayin Well-Known Member

    That doesn't matter. If someone thinks the people who play a game the most should be the best, they should go play an MMO or something. Skilled players can play less because they know how to maximize the use of their time. They get the most bang for their buck, while many lesser players who might actually 'play more' squander their play time.
    Obscura likes this.
  14. deluks917

    deluks917 Yomi League 1 Champion

    ^I'm not saying its bad that grinding does not work. I'm just saying grinding does not work in most games.
    KayinN likes this.
  15. Gon

    Gon Active Member

    Related here is having features in the game that record and track your progress so that your improvement is both noticeable and quantifiable.
  16. MrEnzyme

    MrEnzyme Member

    Then you haven't seen most (if not all) competitive games. The top people in pretty much anything are the ones who practice the most, even though hours spent is not 100% of what makes them the best. There have even been studies that people who spend the most time doing something are the ones who achieve mastery in it (look up the 10,000 hour study).

    With a lot of games you have to practice a long time to get the mechanics (combos in fighting games, macro in RTS games), and even after that you have to practice more once you have those down. So as long a you have some kind of focused practice, playing more does make you better.
  17. deluks917

    deluks917 Yomi League 1 Champion

    ^I said playing the game. I did not say practicing the game. There is a difference.

    I will give an example. When I was in middle school I used to play Smash Bros Melee for hours every Saturday with my friends (maybe an average of 4 hours). I did this for over two years. I literally learned almost nothing about smash Bros. In fact one day I learned about the concept of SHFFL. Over one weekend I spent about 20 hours practicing short hops and l-canceling (I could not fully shffl). In those 20 hours I became 20x better at smash. In 400 hours of playing with pals I became maybe 1.1x as good at smash.

    Their have been studies that most people who play chess every Saturday gain roughly 0 elo points per year.
  18. rozencrantz

    rozencrantz Active Member

    Again, conscious play is the key difference. Playing casually, obviously you won't improve past a certain point, but it's just a lie to say that playing isn't the best way to get better at Chess. You have to play people 100 Elo points above you, and you have to PAY ATTENTION. That is the #1 reason that casual players don't improve: they aren't watching themselves play, they aren't trying to crib new tricks off the opponent, and they aren't remembering the outcomes of any experiments they are doing.

    In one weekend of playing Street Fighter against my friends, I got 20x better at it. Spending days and weeks in training mode didn't do hardly anything for me. So what? To be the best you need both, plus a generous helping of talent, but to improve from wherever you are, you need to play, and play critically above all.
    KayinN likes this.
  19. ratxt1

    ratxt1 Well-Known Member

    I understand where your coming from. You look at this thread and think what a lazy jerk, he doesn't want to try hard to gt good at something, he doesnt want to work to put time and effort to get competitive. I will say that that is not true. First off i have spent plenty of time getting good at games, and am not against practice as a whole in games. Instead i am looking at this from a game designers perspecitve, and thinking isn't the designers goal to try to make the gaming experience as great as possible for the gamer, and by that logic shouldn't he address anything that puts a gamer off from playing his game. That's all I am doing is expressing something that puts me off from trying to get competitive at most games. The fact is getting good at a game is often quite far removed from the actual game.

    In chess (and most abstracts) you have to memorize openings something that is far removed from the actual elegance of chess, and something that can often be done with out even knowing why that opening came in to being and what it's purpose is. In RTSs, FPSs, fighting games, and tons of games in general huge physical barriers that need to be overcome. In MtG you have to spend something like $300 to get a viable deck. The list of obstructions that could put potential competitive gamers off is huge and very present in almost every game. Which brings us back to my goal.

    Make a game that is as fun as possible to get competitive at. Where getting odd is as connected with the actual game as possible.

    I know I have changed my design goal slightly in this post but it think this is just because I am better ealizing what i want and why I am put off of most games.
  20. Kayin

    Kayin Well-Known Member

    Alright, well lemme try and help you break this down.

    So talking about getting good -- seriously, legitimately good at any game worth getting good at is going to take study, and work, and "boring shit"(that some people find interesting, but thats another story). This requirement isn't inherently set by the game, but by the community. Playing a long dead RTS at a capable level is much easier than playing Starcraft at a capable level because people have pushed the game so far. The most popular games are often the hardest to get into, largely because they are hugely competitive.

    For the most part, this is a good thing. Having people love your game so much that they study it and practice shit and dedicate a lot of time to it is awesome! Clearly though, the game designer has not only some, but a LOT of influence over this. He has to provide the depth necessary for the game to even develop to such a level, but he also controls the 'resistance'. Smash Bros isn't a very easy game to get into seriously, but it is an easy game to 'get into'. It has very little early resistance. Fighting games have a very strong early resistance that, depending on the game, ebbs and flows but is always fairly hard. By making things in the game easier to do or learn, you allow people to advance faster and hit less walls. You can't make the game easy or free from study at high levels, but you can make things nicer.

    Now the tricky part is that often making things nicer impacts the depth. Most competitive games involve doing difficult things, and it's hard to make things that scale nicely and it is often the combination of a million little details, rather than one mechanic that leads to the 'feel' of the game. Each aspect on a whole might make you go "This could be more friendly without really hurting anyone!" but if you do that to every aspect of the game, now it's a simplier and possibly boring game. This is, obviously, problematic! It takes a lot of intuition to figure out what stuff you can remove and what stuff you can remove if you replace it elsewhere and what to never touch (unless you have some stroke of genius).

    Starcraft 2 has some interesting examples and is a very very popular game. So in Starcraft 2 they decided "Hey, we can remove a lot of the macro stuff and fighting the AI and some home keeping stuff. Some of of this has to be replaced -- we'll add macro abilities and new micro abilities and stuff. Hopefully then the game will be just as, or nearly as hard as SC1 only the difficulty is offset to 'funner' stuff. A lot of this was hit or miss (Larva injects are like SC1 style tedium :D ), but generally it made a game with a lot less resistance early on that maintained it's integrity. Also Starcraft 2 brings up an interesting point...

    So you say the purpose of a game designer is to give the player the greatest experience? Well, how do you define that? Starcraft 2 is very popular and clearly people find it a great experience, else they wouldn't spend so much time practicing builds and mechanics. Is it proper to look at that as bad thing to be designed away? It's a manifestation of passion -- a sign of success. Also what people find great and enjoyable is often very.. subtle. I made, I Wanna Be the Guy, which is a horrible torture game that some people reaaaaaally love. Why would they love something so horribly caustic to their existence? Well, a lot of reasons, from the feeling of success to masochism... is... masochism part of a 'great experience'? I guess sometimes, yes!

    Tuning the resistance a game gives it's player and when that resistance happens can be really important, and if your game is competitive, moving it farther along the path is, generally speaking, the optimal way to do things. You want people to be able to get in on the base level without hurting themselves if you can.... But that resistance isn't actually all that important. The appeal a game has and the fun players have while playing it is important. If the game captures their imagination, it doesn't matter if improving is a lot of work -- they'll do it. That's why most successful games competitive don't put accessibility first. Creating an exciting, interesting game is the important part. Usability is only important if people actually want to play a game!

    So hopefully that gives you a clearer picture about what's 'going on' with these games and help you articulate what you're trying to get at. I sorta feel you -- a lot of games seem to do things that are blatantly disrespectful to the player and their time and some players love the games for it. It's a conundrum! Just remember, some people juggle. Juggling is not intuitive. It is not a barrier you must pass through to reach some 'real game'. It's just juggling. People just juggle and move the shit out of it and practice over and over against just because they enjoy the experience of juggling. What people find to be a 'great experience' is highly subjective, but, often, is far weirder on average than you might think!
    Delha, deluks917, rozencrantz and 2 others like this.
  21. CrystalChaos

    CrystalChaos Moderator Staff Member

    I think that many FPS games are examples of games that you pretty much only improve by playing.
  22. SW

    SW Active Member

    Sometimes the game itself provides such a poor job at feedback on its potential that the best way at gaining strength (out side of copying better players outright) is spreadsheets and math at many points of the game. This is quite common in games with a lot of "pregame prep" like RPG character build plans, table top armies and likes, where a player can move to upper middle in strength without having ever played the game, especially in the days where information on games is hard to find.

    There are cases where you actually exploit this property. Gambling games are something that actively misleads players into thinking that they can gain strength by playing, when it is simply impossible (too much data to process even with focus, and no useful schema for thinking about the problem without knowing math) very, very quickly.
    ----
    I have thought about games where getting better is about playing more. I think anything that uses pre-conscious skills and data processing that do not exceed working memory should do well. (processing a large, interconnected game tree is not part of that) Something like sports, for example. While there are training tricks that makes training faster and more effective, doing it a lot can get you pretty good reliably.
  23. ratxt1

    ratxt1 Well-Known Member

    I never meant that SC2 or games that require a lot of practice were bad games. I am just trying to think aobut what would it be like if a game was like this.

    For example almost all of sirlin's games are addressing problems he sees in what most people would consider already good games. (chess, dominion, fighting video games, MtG)

    I understand that even if this game meets the goal it will still require a lot of time and effort to get good because that's just how games are. As you said all I am thinking about is making that experience "nicer" and I think they're are defintly ways that don't impact depth.
  24. Kayin

    Kayin Well-Known Member

    I'm not putting words in your or anything, I'm just using SC2 because it's a good example, don't worry. I'll have more to say later!
  25. infernovia

    infernovia Well-Known Member

    I think about as far as I can agree with the topic is removing hidden stuff like z-cancelling or non-intuitive glitches and making all the mechanics pretty simple to understand like in Soul Calibur. This lowers the necessity of players to rely on outside sources, especially if there is not enough competition wherever you are.

    But outside of that, I don't really get where you are coming from at all. I think you are complaining about going into "training mode" but for the vast majority of people, playing RTS and fighting games with some study is the way to get better (play a game, why did I lose? -> find answer, play another game). Anyway, having a deep game usually means that unless you are one of the first people, there is usually a lot of the game already picked apart by the trailblazers. This usually means that during the post-newbie phase, it is helpful to pick up tricks that others have found if you want to get to top level competition quickly. I don't really see a way around that.
  26. Logo

    Logo Well-Known Member

    To be fair fighting games, MOBAs and RTS games do a TERRIBLE job of explaining how to play the actual game from within the game.

    Imagine how far a 'training mode' for SC2 would go. Something that would help you learn to manage constant worker production or some mode that helps you learn the hotkeys and disables mouse input on anything that should be done by keyboard. That already would vastly raise the skill level of the bronze/silver leagues.

    If you can get players in the habit of constant production (or at least a reasonably efficient production rate) then it'd go a long way to getting someone sucked into the game. As it is right now RTS games just kind of spit you out into skirmish/competitive without any direction at all. Even the idea of constant worker production is foreign concept to someone just starting out.

    Or take DotA-likes. LoL doesn't have the problem as much, but in HoN/Dota people are TERRIBLE at picking teams. Something as simple as a rating system for picking teams* could go a LONG way in helping people learn how to pick teams better. So at least people who suck at picking teams will have some feedback about what they're doing wrong. As it is now even trying to explain that stuns = good is beyond many players.

    *Perhaps done like LoL's justice thing where an independent 3rd party gets a list of a teams picks (with no stats/names) in the order they were made and has to rate #2-5's picks (First guy to pick doesn't get rated since there's nothing to compare to). Maybe some extra supporting stuff like when people soft-pick or what time they make their actual pick so you get an accurate picture (sometimes people pick at the last second invalidating your soft-pick when there's only 1-5s left on the counter).

    ------
    Overall I think it's pointless to try to design gameplay that also teaches people to play when most competitive games don't even feature a rudimentary way of learning the game to begin with.
    Leartes likes this.
  27. infernovia

    infernovia Well-Known Member

    Having replays usually gets around that imo, but that's learning the game outside of the game lol (I like to think of it as being able to learn from another, but that's just me).

    Anyway, my point there wasn't that RTS teach people to play their game well (they don't), but to point out that playing the game with some study is the best way to progress after the post newbie phase. The other point was to present a deep game with a simple easily understood concept (Soul Calibur) with the caveat that you need constant competition. But like I mentioned above, it's always going to be fundamentally faster to transmit massive amount of information verbally or with a replay than try to figure out the game for the vast majority of newbies.
  28. Fenrir

    Fenrir Well-Known Member

    Not always. I mean, I'd argue that the best way to get good at fighting games is not to play them but mostly to spend your time in training mode. Training mode is not the game, but when I'm trying to improve at Street Fighter that's where I spend most of my time.

    Eventually you'll hit the point where that stops having value and you improve more by playing against good opponents than you will by training but that takes at least a hundred hours in training mode for most fighting games. IMO.

    One of my friends bought Super Street Fighter IV back when it came out. I would knock him down once and then throw him 'til he died. When this got dull I explained how to tech throws and explained about reversal moves. But I could still mostly just repeat this tactic because he couldn't counter it consistently. He wasn't going to get good until he spent some time learning to execute the moves in training mode.
  29. BeastofBurden

    BeastofBurden Well-Known Member

    Pro tip: play training mode with a friend or a small group once in awhile. It's closer to "playing the game" and it's way more fun/educational than grinding stuff out in training mode by yourself. Then go to your local tournament and split the prize pool!
  30. Fenrir

    Fenrir Well-Known Member

    That's more of a Pro Tip for players than it is a Pro Tip for designers (or just people thinking about design) who're trying figure out ways to have games teach you while you're playing them.

    I think the chess example is interesting too you can go a long way in chess by reading books and memorising opening theory. I think probably Chess 960 is an improvement on standard chess for this reason.

    I have to say that I agree with those who've said that FPS games a quite close to being games that you learn by playing. You're not likely to gain as much from running around an empty map as you will from playing a game against skilled players.

    I think the key difference is between those games that test in game skills, creativity, quick thinking, appraisal, yomi. And thoee games that test skills learned out side the game, like execution or running fast.
  31. FortMag

    FortMag New Member

    I'm confused by the OP's question. The current state of (almost) all game knowledge is or can be acquired in-game, by playing. You can figure out an optimal build order for Protoss vs Terran given the time to play enough matches. You can figure out the most bleeding fast way to Rocketjump to the center of Badlands in TF2 given enough attempts during actual play. You can figure out not to feed the enemy in LoL after getting yelled at, etc.

    The info-gathering, watching people play, optimal strategy discussion metagame? That's entirely a player invention to quicken the process. YES, you can figure out the fastest rocketjumps on any given map given enough play-- but it's much, much faster to simply check online, where an ongoing iteration of people improving on other people's strategies is taking place. You can most certainly operate in a vacuum. But you're going to fall behind people who are collaborating, metagaming, iterating on the latest 'optimal' strategy.

    If the OP's question is asking if it's possible for a game to be tuned towards "play only", I really have to say it is not. Tic-Tac-Toe can be solved with little effort, yet even LESS effort needs to be expended by googling "tic tac toe solution". But can you figure all this strategy out on your own grit? Most definitly. You're just going to be miles behind everyone else -- you're one man vs the collective.

    That said, if we're talking entry barriers to competitive play here, sure, a game can better explain its base concepts through play than another game. You can figure out how to get good at competitive SmashBros:whatever or CS without google's aid long before you'd break into SC2's scene. (I'm not saying it'd be an easy task for the former, either) Wether this is due to overall game complexity or a more intuitive nature of some games/some games being better teachers, I'm not sure. Likely both.
  32. Leartes

    Leartes Well-Known Member

    Why do people always pick these extreme stances?

    I'm 100% sure it is possible. Possible means does a game exist among all games invented by mankind and all games mankind can potentially at any point in the future develop.

    Take a purely luck based game like roulett. Everyone is equally good or bad at it as it is as everyone is losing the same amount in average. You can read as long as you want about it and won't improve. Unfortunately you can't improve by playing either but at least we found a game that does not inherit learning by reading. Depending on the strictness definition on improvement (< or <=) this might be an answer to the OP (but an unsatisfying one). Again claiming impossibility of existance is a overly strong stance in my opinion.

    The more interesting questions are:
    What properties lead towards such a game? How hard is it to make this game? Is such a game fun, or does it inherit properties that aren't enjoyable?



    A game such like the games Keith tries to make with tons of randomization and only valuation of game states you never saw before and will never see again is promising for this. If every situation you run in has never occured to anyone else ever playing the game and all other gamestates are significiantly different, than you can't prepare for it by reading about it beforehand.
    On the other hand the question comes up if you can learn in this game at all as learning in some way consists of patterns (which are learnable outside the game as well).

    EDIT:
    A little extension to the roulett example: We know mathematically you can't learn anything about the game inside or outside. On the other hand as roulette tables are rarely perfect you might hypothetically learn something about the table you are playing at while playing. Therefore if you are a regular at a specific table for a long (most likely very very long) time you can theoretically learn something about this table and play above average. As long as this information is not public - and it isn't if it is hard enough to come by - only playing is a way to improve.
    Using this the game does not rely on some strictness definition and fulfills the requirements.
    Obviously the game is not satisfying to play regardless. It is an example of a game that has some properties but is not fun regardless. Some might even claim it is no game, but if you define it as a game about exploiting physical imperfection in the table it might be a game even in the other strange definition.
  33. infernovia

    infernovia Well-Known Member

    The question, again, is that if no one else has run into every situation while playing the game ever, is how the heck do you actually improve? Such a game would be impossible to create, imo, and even if it was possible, wouldn't be very good. Also, Keith's dream game involve stuff like Civillization, which you can rapidly enhance by learning stuff from guides and forum posts. And that's kinda inherent in all deep games.

    And like I mentioned above, as long as you have a replay system and deep enough mechanics, it's always fundamentally going to be faster if someone tells you what to do/shows you what to do instead of figuring it out on your own. That's kinda the reason we invented communication.
    zem likes this.
  34. Leartes

    Leartes Well-Known Member

    Agree, I was not saying such a game would be good as in fun to play. It was merely a suggestion how a game without external learning might look like.

    EDIT:
    Also it remains an open problem if such a game can be constructed.
  35. SW

    SW Active Member

    While it is extremely unlikely for games to be optimal in training skills, there are still some where this works.

    The biggest one would be things that can NOT be encoded in communication and is learned on a pre-conscious scale. People can tell me how to swim fast, to read a curve ball or to make headshots, but they are simply far too subtle to be learned that way. One can probably just put in fairly basic instructions and leave the learning almost entirely in playing without large loss of efficiency.

    "Strategy games", which is almost entirely optimized on a conscious level, easily and strongly encoded in a conceptual level, does not work. While one could recruit pattern recognition to help, the game simply provide unclear feedback that had to be preprocessed by the consciousness before the brain can recognize useful patterns.

    edit: The existence of alternative, effective means of solving games is what makes "playing" less effective. At some point there might be a powerful algorithm general purpose for playing a large subset of games, and the best way to get started is to memorize the output before even getting started on it....

    edit: 700th discussion and 20,000 post on this subforum-lol
    Waterd103 likes this.
  36. Logo

    Logo Well-Known Member

    Well one thing RTS games teach that you can't learn via being told is a sense of timing. I can tell you every 30s to larva inject, but that's almost completely meaningless beyond the strategic level. How long is 30s? You're a lot worse judge of that than you think*. That's something you only really pick up on by playing a lot. Early on you may check every 10s for larva inject or every 50s and miss a bunch of injects, but as you play more you gain a rhythm and you end up being able to check less and check more accurately (as in you'll check in +/-5s of inject finishing instead of having to check every 15s to be even remotely close). It's also the sort of thing that's hard to learn outside of the game. 30s in Starcraft feels way different than 30s of sitting there doing nothing.

    I'm pretty sure a lot of frame and pixel spacing stuff in fighting games is the same way.

    *Fun exercise, start a stopwatch, look away, then try to turn around when you think the stopwatch is at 1 minute WITHOUT counting in your head. You'll probably be way off.
  37. rozencrantz

    rozencrantz Active Member

    Any game that involves non-conscious learning could have that property (that you can't communicate how to play) though it isn't necessarily the case. If your game tests a bunch of skills, training them individually might be too much of a benefit (practicing combos for example.) A game that tests one particular skill would be most likely to have this property in its purest form.

    You might look at Dual-N-Back (though it's debatable whether or not that is a game) as well as Multitask and Multitask 2. Having one core skill means that the best way to practice that skill is probably by playing the game itself.

    I do not know if it's possible for a game involving contested skills to have that property.
  38. infernovia

    infernovia Well-Known Member

    Note that things that are not highly regulated by consciousness, like physical activity (Tennis, Sprinting, Swimming etc.) require focused training to make significant process.
  39. Kayin

    Kayin Well-Known Member

    Even more so, most activities like that have an even higher barrier to entry. Like, even playing Golf, developing a basic, reliable stroke can be a huge pain in the ass.

    Though that said, Golf might vaguely fit, as would say, Bowling. You might go to a driving range or practice putting or practice hitting splits or something, but the similarity between 'play' and 'practice' is pretty small and if someone enjoys golf, they'll probably enjoy knocking off balls at a driving range. Both games are also not directly competitive, so if you do refuse to practice, well, no one will care and you won't suffer from playing with better players.

    Really, the best way to get better at a game by just playing it is just find people who play with who also don't want to practice and let skill development go naturally. You'll probably never be truly good, but it's all relative anyways, so who cares?
  40. infernovia

    infernovia Well-Known Member

    Yeah, I think the stance is definitely workable, and it's even a great way to play and enjoy games. You just have to stop worrying about being "the best of the best," it's really going to be anathema to the whole idea.
  41. sloppyJoe

    sloppyJoe Member

    Roulette isn't purely luck based, if it was it would be completely random, which it isn't. Each table and roulette has different physical qualities, and the ball is also influenced by several factors, one of the main ones being the relative strength and habits of the casino guy/girl starting the ball. There will then be an average speed of the ball, and with knowing the speed of the ball you'll be able to calculate how far it will travel and as such be able to somewhat determine where it would end up. However, since there will be multiple people throughout any given night/week/month/year and ball being swapped about all the time you'll have a more than hard time to be able to game the system. Then again, this being so complex and having too many uncertainties the game becomes mostly random and a game of mostly luck. However the knowledge that your highest chance of winning is to go for red or black comes from mathematics, not from playing the game, hence you get to get better at the game by studying it from outside, not that getting good at roulette is a skill worth pursuing.

    Anyhow, that was just me being difficult.
  42. sloppyJoe

    sloppyJoe Member

    In order to determine what a good game for getting good by just playing the game means you'll first have to define what kind of game it should be. Since this is being discussed on Sirlin's forums I don't think it be too wrong to assume we're talking about a competitive game of some kind.

    If we assume a game to be a competitive game we've already found a problem with games such as SC(2). The campaign and multiplayer are essentially different games, though both you'll get better by outside means. You'll improve your campaign skills by playing multiplayer as you'll (hopefully) get into habits of producing more of things, while playing campaign won't improve your multiplayer skills much beyond a basic understanding of each race and unit. Same in basically all other games which offers two distinct gaming experiences.

    Looking outside gaming we could compare to sports. Yes, the best way of getting better is to play games, there's nothing better overall than playing an actual game, however you'll only get so far by playing games. In say basketball, you'll have to work on your stamina, otherwise you'll never be able to play a full game. Work on your shots, else you'll loose. If you can't dunk you'll either have to rely on lay-ups (and adjust your play accordingly) or work on your jumping skills. And so on. Same with all sports, and consequently I disagree with the statement that external practice won't do much in golf and bowling, and no they are both not non-competitive games (golf for instance is advised to be played as a bet every time, where the looser of the flight buys the rest coffee/beer/food/whatever after the game).

    If we look back to games again we'll see that all successful competitive games follow a similar pattern where you as a player get to choose what to improve upon, thus making your actual play better as a whole. However, even if SC2 lets you externally work on your APM, or your constant production, or your mental stamina, or your tactical insight, or your micro abilities, everything comes together in your game and that's where you'll improve the most as a whole.

    Then again, would a game where the absolute best way of getting better be just to play the game be game you'd want to keep playing? Would it be able to sustain enough motivation long enough (disregarding the motivation that is money)? I would argue that any such game would plateau (relatively) quickly and thus become boring.
  43. rozencrantz

    rozencrantz Active Member

    What about Table Tennis? When I played that semi-seriously, I didn't find anything particularly useful out-of-game that I couldn't get more from by practicing in-game. I guess if I went really serious I'd probably want to spend time practicing techs, but it's easier to do that if you have someone on the other side of the table returning the ball, so you don't have to go fetch it every time... at which point you might as well be playing.

    How do we count learning-games? As in: playing matches where players don't try to win, but focus more on pulling off techs or setting them up for the opponent. You're still playing, but you are also practicing. Actually I think that might be a fundamental flaw of the question, that the line between playing and practicing is not sharp.
  44. Leartes

    Leartes Well-Known Member

    I doubt that. I mean sure you can calculate some estimates which is good starters. But I think you need a ton of pratice ingame to see the fine nuance if the ball is faster or slower than average and which result is favored in the current setting.
    Humans are terrible at this kind of stuff but I don't think math helps much here as you cannot do calculations during a round.

    100% agree :D

    I guess we should stop talking about roulett as there is noone good enough to provide first hand information on howto get strong at it.
  45. JackShandy

    JackShandy New Member

    Really, the only type of game you can succeed at without using any outside knowledge is a game that relies on totally different skills to everything else in existence. If it involves skills that are useful in some other field - mathematics, valuation, reflexes - then it's always going to be easier for people who are good at that field.
  46. sloppyJoe

    sloppyJoe Member

    Would such a game a game exist at all? Since all games are a result of ideas, they are based on something, thus a skill from a certain field.

    And honestly, there isn't a single game where you don't benefit from improving your physical stamina.

    @Leartes: Oh I would argue it would be practically impossible to actually calculate where the ball will end up in real time from the limited info you get at the roulette table (being semi-drunk/drunk obviously doesn't help). Perhaps if you have perfect memory (of which there are several individual humans in the world who does) you could learn to recognize patterns and as such predict where the ball will end up, or perhaps you could intuit it, but then again. Getting good at roulette is a rather large waste of time and resources (not to mention a waste of humanity's resources in total, just imagine what else a smart enough person who could pull off such a feat could contribute to humanity through interesting research).

    @rozencrantz: Table tennis also have outside factors. Practicing serves, where you do nothing but serve for a set amount of time, or practicing hitting a fast screwball right in the corner over and over, or playing against a controlled partner who will set up your ideal practice situation all the time. When doing these activities you aren't playing the game, you are practicing specific skills which will boost your game as a whole, but not directly.

    I'd like to bring up a different example, the little cute 360 game Ilomilo. For those of you who don't know it is a game about Ilo and Milo who want to meet in the park and have tea, only problem, the park radically changes between each meeting. In other words it is a three-dimensional puzzler where you gather stuff and find your way through the maze. Anyhow, when my girlfriend plays it she's quite careful and spends relatively much time to move around and solve her problems, when I play I run around essentially brute forcing my way through the game. The main difference between the two of us is that she isn't a gamer, while I am. I play games like I've learnt through playing for almost 20 years (problems are to be solved), she plays games more wondering than me.

    My point? Even such a simple concept as Ilomilo you'll improve greatly from outside experience. So much so I would think no gamers will ever truly find themselves in a position where they have no prior knowledge which will help them directly or indirectly in the game. Games are systems after all, and if there is one thing you learn from playing games it is learning systems easily.
  47. Warskull

    Warskull Active Member

    FPS games, the primary skill test of an FPS is aiming. You get better by practicing. Outside theory can help to a degree with maps knowledge, but you still need to play most FPS games to improve at them. Playing other FPS games will help you improve at any given FPS game, but it is mostly due to the shared mechanics.
  48. Leartes

    Leartes Well-Known Member

    Agree on the line about stamina.

    Mostly agree on the rest of the quote. I'm not saying learning roulette is useful or even possible for every human. I'm just saying I can imagine people doing it which is enough to counter the non-existance claim.

    It is a good example as training other games is not as useful as training the real game.
    In CS there are training maps to improve aiming, they are not really the same as regular maps. One can argue that they are as close to ingame training as possible as you usually try to win and the win condition is there in _cs or _de maps as well.
    It is really close on the fine line of practice vs practise match. A practice match with nothing to lose except honor is still the real game.
  49. Delha

    Delha Active Member

    Even in a game very heavily focused on the testing of a specific skill (eg Golf, FPS), it's indisputable that external input will greatly accelerate the development for most individuals. If it didn't, people wouldn't take golf lessons at the country club.

    The vast majority of people left to their own devices (ie. just playing) will not chance upon the optimal path to progression. Any time that's true, it's pretty much given that advice provided on forums or by a coach could improve their learning process. Any self taught skill tends to be about trial and error, and pooling knowlege gives you answers without needing to test everything yourself.
    zem likes this.
  50. CrystalChaos

    CrystalChaos Moderator Staff Member

    What external input will accelerate development for FPS games?

Share This Page