Luck and Skill in Games
HTML-код
- Опубликовано: 20 июн 2024
- In this 2013 GDC Next session, designer Skaff Elias aims to explain how to balance luck and skill in the art of game design, making them interweave properly to create a proper challenge for players.
GDC talks cover a range of developmental topics including game design, programming, audio, visual arts, business management, production, online games, and much more. We post a fresh GDC video every weekday. Subscribe to the channel to stay on top of regular updates, and check out GDC Vault for thousands of more in-depth talks from our archives.
Follow us on Twitter
/ official_gdc
Check out our Facebook page for GDC exclusives
/ gamedevelopersconference
Visit our site: ubm.io/2ctNvqZ
Whoa. I expected a relatively boring&often debated topic, but after 10 minutes it becomes really insightful.
your comment kept me here and i don't regret watching this. thanks
Two doors. One leads to victory, the other one leads to success. That's a tough decision.
This is why we have day 1 patches.
Metadigital
Actually, that's why we test before release.
That's an optimistic view.
Metadigital
Well, that's the way it should be. The sad thing is that modern technology allows developers to be lazy and fix things post-release. That wasn't possible a few console generations ago. That's one aspect where the industry kinda went backwards.
Games also require more work, more art, and larger teams than back then. So, it's not really laziness. In fact, It's a little offensive to call the people working on games lazy considering the long hours involved.
Finally, It only took them 4 years to upload ... ehehe eh ... eh ... .................................... heh
They have the content behind a paywall if you're in a hurry.
I have a roughly 16% win rate against Kasparov in Rando Chess. That's right, I'm that good at the game.
I really like the part around 20 minutes. You could walk in a store and play Magic with the people you liked, and it was fun for everyone. While when I tried to play Go, there were only a few people in my city's club, it was like asking for private lessons.
I want to her the Q&A section. : (
As a computer scientist, I think it's important to make a clear distinction between between true randomness/non-determinism and player predictability. Randomness, as the term is used here, is actually referring instead to player predictability. If the player isn't able to reasonably predict future events, they call it 'random,' when in reality, with games like Chess and even more complex stuff, given sufficient information, all future events can be algorithmically predicted. It's only random if, given every piece of information about the system, one can still not predict future events. Practically, of course, many games are complex enough that the player isn't able to reasonably predict things, achieving low predictability without actually being random. Just a nit pick that was bothering me.
Whether or not something is perceived as random is solely dependant on whether or not someone can (given his knowledge state of the world) predict what is going to happen next.
If a game does something without giving the player any information that could be used to predict that action, then it is doing something random. Simple as that.
Then nothing literally is random in the universe. Its just that we cant yet simulate a universe to predict state of universe over time at the smallest scale possible.
It's what I'd call extrinsic vs intrinsic luck, or maybe arbitrariness vs randomness
Just because someone has a chance to win with random moves doesn't make it luck base, thats capitalization on human error.
"Because there is an over randomness to it, it gives them something to very clearly fall back on. The problem is over time, the more and more experienced players are, the less this illusion will hold up."
I think this talk lacked some more insight into what makes the luck digestible and maybe enjoyable. IMO luck is great when it determines the environment of the game, not the outcome. In randochess luck was determining the outcome of the game, however in the strategic game, the environment was changing.
Similarly, in standard poker there is a lot more luck involved when you do it the straight way (you draw all hand) as opposed to texas holdem (where you draw 2 cards and cards on the table are your environment upon which you can build your hand). You still know your hand and it is your environment in both types, but with the added guesswork you can do based on the cards on the table, making luck in texas holdem more about environment than the outcome.
For the people reading this like my years later. The presenter mentioned he shortened a 1 hour presentation into like half an hour within 24hours of knowing. Obviously he left out a ton.
Really, really interesting talk. Wish it was longer. Going to check out his book. Anyone have similar videos to recommend?
This seems quite insightful. I'm not sure how to apply it, but it seems like it has unintuitive truths. I think it probably can be applied well, but not without more game making skill than I currently have. Really cool talk.
Really enjoyed this video. I'm currently designing my own game with multiple randomizers... It's a lot more complicated than I thought... :D
Really interesting! Thanks!
If game is solvable, then there is a "best move" in every state. Therefor, every move is either the best move or some weaker move. The magnitude of error (distance from the best move) for each move is determined by two factors, player skill and random error. That is, a chess GM sometimes makes an "error" that improves his position better than his average move. Because error can be beneficial, even solvable games with no hidden information must include variance to be fully explained.
Digital Extremes should watch this.
Awesome :D
"You probably heard of him and - erm - not me.." aaaaaaw
Some segments in the video are stamped not adjacent to each other
hashtag gdc_slowpoke, hashtag better_late_than_never
Now if only that Richard Garfield talk had the same level of audio quality...
Hmm, that tank-dragon example didnt help me much understanding what he means. I mean i understand that he wants a system that brings in randomness to experienced players, but doesnt really change randomness for lower level players. But the example seemed to be as vague as the description of this idea :)
The point was that by adding randomness, he gave room for skill in a different dimension from what was previously used. If everyone sticks to the same strategy they always do, then it's still mechanically based skill game where the ratings will be the same (over time), but more variance caused by the additional randomness. However, if you consider that players may now consider doing different build orders and army compositions due to the cost of units, there is now an additional layer of strategy to the game. Some players that were not very mechanically skilled, but understand the cost efficiency of units well (player C) can use this to their advantage. While players who don't, will suffer (player B). At a glance it might seem that the new ratings are changed because of variance (extra randomness), even though it was actually the strategic decisions that made the difference. Of course this RTS seems to be a poorly balanced and designed game either way, but the example makes sense.
I wonder if anyone ever said "There's too much skill in the game".
A game doesn't need to be complex to be hard (Dead Cells, Dark Souls)
Perhaps people who don't enjoy being beaten by much more experienced players. Or said experienced players who don't enjoy fighting such opponents, whether out of sympathy or lack of challenge?
Thank you RUclips algorithm for this video
Now go ahead and screw over some more creators
Acctually, the 50'347'200 decimal of Pi is 9 as in "...10618584960860592185130723697130919..."
Throwing darts at a darts board is not random is it? I have seen pro darts players who prove this.
Great talk, but he doesn't know enough about darts.
Unfortunately, the speaker seems to not understand the meaning of randomness. It does not mean uncertain or unpredictable or inscrutable.
What do you mean? If a programmer generates a random number it is for creating something unpredictable. Usually you know that there are for example two possible states this randomness can lead, and you can calculate with that as a player, but you don't know what it will be.
The assertion that there is randomness in chess borders on the absurd. The action of the other player may be unpredictable, but chess is a total information game. The other players actions are intelligently (with more or less skill) directed, not random. The speaker is continually using the word randomness when he really means unpredictability or uncertainty from the point of view of a single player.
I think he meant that randomness appears because you as a player cant possibly fathom all the possible game states that can develop from a situation. In fact, IF the universe is deterministic, then the concept of random doesnt exist. And everything just appears to be random. From our perspective, when we start a game of chess, we go to a journey of randomness. If we would be infinitely intelligent, chess would be a booring game to us, as booring as tic tac toe.
So randomness depends on your point of view. From a humans point of view, random generators are random. From a computer's view, they are deterministic calculations.
You have repeated the flawed argument very well. But lack of knowledge is not randomness. There is nothing random about tic tac toe, nor about chess. Both are total information games. The real point that is being made, which is a good one that should not be obscured by misusing the the word random, is that games are more fun when the player cannot predict the outcome--but there are a lot of reasons why an outcome might be prohibitively difficult to predict, which do not rely on randomness. In the case of chess, the complexity is too great.
No one likes R.N.G. in Mario...