tag:blogger.com,1999:blog-11719805.post3022309783184394421..comments2022-12-02T19:19:58.206-08:00Comments on Lost Garden: Understanding randomness in terms of masteryDaniel Cookhttp://www.blogger.com/profile/10437870541630835660noreply@blogger.comBlogger11125tag:blogger.com,1999:blog-11719805.post-77782838609696202972013-02-15T06:50:10.606-08:002013-02-15T06:50:10.606-08:00Having just found this web space I must say I'...Having just found this web space I must say I'm pleasantly surprised not only by the quality of the article but by that of the comments.<br /><br />While I enjoyed the article I didn't share of your analysis on the application and desirability of randomness as a whole.<br /><br />In particular, the topics of outliers and your conclusion ("A well rounded designer does not remove randomness from their games. The world is a random place...") were the ones that I found particularly bothersome. Jotaf, Keith Burgun and Mistify have been more than eloquent already but I'll add a thing or two.<br /><br />On the topic of outliers, I believe the emphasis should be placed on what these do to the spectrum of effective player actions. <br /><br />One of the things that makes random events interesting is that they create a state in which optimum strategies must be selected and implemented dynamically; in other words, they help us move away from a complete separation of planning and implementation towards a state in which one informs the other in a loop during play.<br /><br />If an outlier destroys the action space in such a way that a big enough disadvantage/advantage is created, player action might be completely (or overly) devalued. Even in cases in which these can be overcome by experienced players it creates a separation of the difficulty level of play when the outliers are absent and when they appear that might impair a player's ability to produce an accurate model of the situation.<br /><br />Yes, drama shall abound, but not all drama is desirable emotional involvement.<br /><br />I applaude you taking a stand in your conclusion, but I couldn't disagree more. Randomness can be used to make good games, but arguing that there's causality between the quality of these games and the inclusion of randomness is simply bad logic. As Keith Burgun mentioned, there can definitely be great games in which judging odds is the core skill... in retrospect, however, randomness has been often used to obfuscate the underlying model in order to inflate gameplay or present something as more complex than actually is; in the worst cases it has been a tool to hide the inadequacies of a design (this is, of course, a personal observation.)the_blunderbusshttps://www.blogger.com/profile/18375917747859710391noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-21086543871694455562013-01-16T20:36:14.480-08:002013-01-16T20:36:14.480-08:00I really enjoyed this post. I posted some reflecti...I really enjoyed this post. I posted some reflections on it exploring the question of what exactly is a 'skill' and, in agreement with your ideas above, suggesting that noise is actually an integral part of skill rather than something that opposes or degrades skill in a game. thanks!<br />http://synaptogaming.com/2013/01/16/is-randomness-antithetical-to-skill-based-games-or-a-crucial-element-of-mastery/Unknownhttps://www.blogger.com/profile/03833846934540200942noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-18843766381954269032013-01-12T11:58:02.740-08:002013-01-12T11:58:02.740-08:00Really useful article Dan. It's been pretty he...Really useful article Dan. It's been pretty helpful for me as i've been thinking about eSports and black box game design.<br /><br />Also, I found some typos as well.<br /><br />A good hidden object game players is measurably better than a new player. -> A good hidden object game player is measurably better than a new player.<br /><br />Probability and statistics provides use with a set of mathematical skills for dealing with randomness. -> Probability and statistics provides us with a set of mathematical skills for dealing with randomness.<br /><br />Instead, use system at are reasonably easy to figure out. -> Instead, use systems that are reasonably easy to figure out.<br /><br />Thanks for writing :)<br />Shervin Ghazazanihttps://www.blogger.com/profile/11350258070713471695noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-66700846817280344932013-01-12T11:56:12.457-08:002013-01-12T11:56:12.457-08:00This comment has been removed by the author.Shervin Ghazazanihttps://www.blogger.com/profile/11350258070713471695noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-42882320737406945602013-01-09T03:07:29.604-08:002013-01-09T03:07:29.604-08:00Well informed, well written, useful. Thanks!Well informed, well written, useful. Thanks!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-11719805.post-78087298605316389112013-01-04T19:30:00.369-08:002013-01-04T19:30:00.369-08:00Thanks Elyandarin. Fixed the typos. :-) Thanks Elyandarin. Fixed the typos. :-) Daniel Cookhttps://www.blogger.com/profile/10437870541630835660noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-74893903142787732912013-01-04T18:06:53.655-08:002013-01-04T18:06:53.655-08:00I will hold that randomness should not determine t...I will hold that randomness should not determine the outcome of a game. It can add interest and variety to the middle of the play, and when you are progressing close to the expected outcome, it is fine. But even then, you hit an outlier, and despite any skill you have, or how well you have mastered the randomness, through absolutely no fault of your own or skill of your opponent, you fail. In a competitive space, randomly doing well is also problematic, as that means the opponent is losing, again through no fault of their own. This invalidates your gameplay. Every good decision the player has made, every bit of skill they have applied, is rendered pointless. <br /><br />Mystifyhttps://www.blogger.com/profile/00150647144490780496noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-83805813325322861792013-01-04T16:23:45.014-08:002013-01-04T16:23:45.014-08:00I enjoy randomness in games, to a point. As long a...I enjoy randomness in games, to a point. As long as I can have consistently rotten luck and still win by using good strategy, then it adds flavor and depth. Otherwise, I feel it takes something away both from my wins and losses. <br /><br />...Speaking of injecting unrelated noise into an otherwise coherent system, I noticed a number of typos in the article - dropped, added or substituted words. <br /><br />what that result might. -> what that result might be.<br />the help teach -> to help teach<br />There is These is uncertainty, -> These are uncertainty, (or possibly simply starting with "Uncertainty, ") <br /><br /><br />Anyway, nice article :)Elyandarinhttps://www.blogger.com/profile/08289813912842329057noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-3991830795328812462013-01-02T11:15:06.844-08:002013-01-02T11:15:06.844-08:00Actually, even a bit of noise fundamentally change...Actually, even a bit of noise fundamentally changes the way you play a game. In a fully predictable game, you *know* there is an optimal decision; every other choice is suboptimal. You only consider your target in game-space and the route that gets you there, and every other solution is irrelevant.<br /><br />With noise, you have to regularize your solution so you don't overfit -- you must aim for a general area in game-space where you probably win, considering that noise will cause slight deviations. The neighboring solutions of your target solution will matter a lot. This is much harder to consider.<br /><br />On another note, the comparison to poker is very relevant and makes it clear why some games are easier to master than others. To make the rules easier to learn, you need to reset most of the game's state very often, like the multiple rounds in poker. A game that keeps a large bag of state information for a long time will be harder to learn. The large number of small rounds makes such a random game much more tractable for a new player.<br /><br />In probability theory terms, the results of poker rounds are fairly IID (independent and identically distributed), which makes bayesian inference easier. In conclusion, to make a game easier you can segment it into smaller events, and carry over as little information between them as possible.Jotafhttps://www.blogger.com/profile/11804130701206652599noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-15420019933449667852013-01-01T06:45:31.748-08:002013-01-01T06:45:31.748-08:00>The world is a random place and learning to de...>The world is a random place and learning to deal rationally with randomness is a critical life skill.<br /><br />Since when do games have to teach us critical life skills to be good, and who said they have to reflect "the world"? Was Reiner Knizia being a "not well-rounded designer" when he designed the luck-less Through the Desert?<br /><br />Randomness is noise. It doesn't matter if you have tools with which to understand odds; that doesn't change the fact that when you roll two dice and it comes up snake-eyes, that was meaningless noise from outside the system. You can understand that you have a 99.9% chance of getting a desired outcome, and still not get that desired outcome, because RANDOM NOISE - that's the only reason.<br /><br />The only situation in which it's justified to use randomness (output randomness) as a tool in game design is if you're building a game whose entire core mechanism is judging odds. So, in Blackjack or Poker or whatever, I understand it.<br /><br />But if you have a situation where it's a tactics game like X-Com or Summoner Wars but then you throw in random dice rolls, that's just injecting unrelated noise into an otherwise coherent system.Keith Burgunhttps://www.blogger.com/profile/09146704232808889882noreply@blogger.comtag:blogger.com,1999:blog-11719805.post-72738831878992562142013-01-01T05:58:06.729-08:002013-01-01T05:58:06.729-08:00Yes, you can certainly become skilled at luck! Alt...Yes, you can certainly become skilled at luck! Although I am surprised you did not mention anything about some kind of negative feedback loop to counteract outliers. Or do you think the nature of the random distribution is more important to tweak? (Such as using "a deck of cards" rather than "dice")Anonymousnoreply@blogger.com