Chris Taylor
Contributor
Best thing I learned in school: Fail Faster. Try it, and iterate on the results.
General Color matters / devotion
- Thread starter CML
- Start date
Best thing I learned in school: Fail Faster. Try it, and iterate on the results.
I've been playing a lot of a variant format called THE DANGER ROOM/BATTLE BOX, inspired by Brian DeMars's article and Ben Stark's Battle Box being mentioned in a CFB video. The critical difference from regular MtG is that you start with one of each basic and allied guildgate/tapland off to the side and can play 1 land a turn from that pile; you can't really ever get mana screwed since you'll hit a land drop for the first 10 turns, you can't get really flooded since every card you draw is a spell, and you generally don't get color screwed unless you sequence your lands poorly.
Also you share a library but not graveyards, which adds a new twist to things like scrying and put-target-whatever-on-top.
DeMars and Stark have lists that have been tuned somewhat. The 2 major differences (land cache, large shared library) restrict the design space quite a bit; LD is awful for the environment, can't really have any super-bomby cards, can't have too many synergy cards, shuffle effects are bad for the experience, scry becoming fateseal, etc. The static land cache also lends itself quite poorly to any kind of format where you actually construct a deck... still prefer it to WoW/Kaijudu(or Z/X...) cards-as-mana or Hearthstone 1-more-rainbow-mana-per-turn.
I just threw a bunch of draft commons and old cube leftovers into a box and called it a day, and it's still pretty fun despite being completely imbalanced. If you have an inconsistent playgroup to begin with, maybe try this format out with whoever is on hand and see how you like it.
Also you share a library but not graveyards, which adds a new twist to things like scrying and put-target-whatever-on-top.
DeMars and Stark have lists that have been tuned somewhat. The 2 major differences (land cache, large shared library) restrict the design space quite a bit; LD is awful for the environment, can't really have any super-bomby cards, can't have too many synergy cards, shuffle effects are bad for the experience, scry becoming fateseal, etc. The static land cache also lends itself quite poorly to any kind of format where you actually construct a deck... still prefer it to WoW/Kaijudu(or Z/X...) cards-as-mana or Hearthstone 1-more-rainbow-mana-per-turn.
I just threw a bunch of draft commons and old cube leftovers into a box and called it a day, and it's still pretty fun despite being completely imbalanced. If you have an inconsistent playgroup to begin with, maybe try this format out with whoever is on hand and see how you like it.
With regards to WoW TCG, I feel like they got a lot of things about their rule system right, like face down cards as resources, and targeted attacks (every combat is a "fight", in Magic parlance). Where things went out of whack was in their set design and individual card development. The game suffered massively - massively - from an on-board complexity problem, with game states that make Lorwyn limited look like brightly coloured Duplo blocks. It got so bad that constructed matches had to be changed from best-of-3 to best-of-1 (!!), because the on-board complexity meant that three games could never be finished within a fifty minute time span. But this is something, again, I blame on the card design, and not the base ruleset. NWO wasn't a thing that existed, and every successive set only upped the bar on swingy activated abilities and triggers with inscrutable timing. If they had the R&D resources of, I don't know, a Magic, I think they really could've had a hit on their hands.
Fair point. I ran a small test and posted some thoughts in the evolution thread.
So I'm going to quickly weigh in on the 4-of rule.
Lets reduce this to a minimum: Imagine magic where there's only creatures, and resources are generated by some mechanism instead of there being lands, and there's no 4-of. One of the better decks will probably be 40 'best one drop', or maybe 'best two drop' depending on how much your extra mana gets you. Clearly there are then some decks in the middle that go 20 ones and 20 twos or something around there, that beats the one drops as its got twos, but loses to the twos because it has ones. Right? Right.
So how does this relate to the 4-of rule? Well clearly a marginally better deck than 20/20 is something approaching 18/12/10 (or whatever), and so on until you put in a single 8 drop and your game vs goes-to-seven.dec is determined by whether you can curve to it. This is, in a nutshell, the variance problem. Assuming you can successfully identify best n drop, and optimal curve, your deck consists of the best cards at each cost in each bracket.
The 4-of rule goes fuck that noise, and throws a great big dollop of variance into the mix, because variance is what makes games fun. This was where the original rules Garfield came up with were aimed: people will spend the price of a game on cards, have like 300 cards, and build what they can with their odd ones and twos of powerful cards, but each person has different power cards so it's interesting. Then people bought ten times as many cards as he expected and built 15 lotus 15 bolt 10 timetwister (or whatever), with zero variance. But really, having to stop at 4 $best_2_drop and then fill out your other 12 2 drop slots (which, incidentally, will change what the best 2 is because the format isn't degenerate) with sub-optimal choices, figuring out what those sub-optimal choices should be to make then optimally sub-optimal, and trying to figure out what choices other people will make so you can adjust your 'best card' function to take advantage is the deckbuilding game.
Now, talk to people who like to think they play competitively, and they'll (probably) tell you that variance is anathema to competitive gaming, but they're wrong because they're playing magic and not competitive RoShamBo, where there's no variance at all and therefore is the perfect competitive game(?). Even within magic, the times that competitive players stopped playing where at the times when deck-building was at a variance-nadir; ravager affinity, caw-blade, people complained about standard delver a bunch (although it was more interesting to play due to the gameplay variance involved), people complain that theros mono-U all day every day was boring. Variance is important for fun, and 4-of saves people from themselves optimising it away.
Conclusion: play netrunner with 45 card decks, 3-of rule, influence costs being the best way to regulate deckbuilding, and a much, much lower risk of not being able to play the game due to card order in your deck.
Lets reduce this to a minimum: Imagine magic where there's only creatures, and resources are generated by some mechanism instead of there being lands, and there's no 4-of. One of the better decks will probably be 40 'best one drop', or maybe 'best two drop' depending on how much your extra mana gets you. Clearly there are then some decks in the middle that go 20 ones and 20 twos or something around there, that beats the one drops as its got twos, but loses to the twos because it has ones. Right? Right.
So how does this relate to the 4-of rule? Well clearly a marginally better deck than 20/20 is something approaching 18/12/10 (or whatever), and so on until you put in a single 8 drop and your game vs goes-to-seven.dec is determined by whether you can curve to it. This is, in a nutshell, the variance problem. Assuming you can successfully identify best n drop, and optimal curve, your deck consists of the best cards at each cost in each bracket.
The 4-of rule goes fuck that noise, and throws a great big dollop of variance into the mix, because variance is what makes games fun. This was where the original rules Garfield came up with were aimed: people will spend the price of a game on cards, have like 300 cards, and build what they can with their odd ones and twos of powerful cards, but each person has different power cards so it's interesting. Then people bought ten times as many cards as he expected and built 15 lotus 15 bolt 10 timetwister (or whatever), with zero variance. But really, having to stop at 4 $best_2_drop and then fill out your other 12 2 drop slots (which, incidentally, will change what the best 2 is because the format isn't degenerate) with sub-optimal choices, figuring out what those sub-optimal choices should be to make then optimally sub-optimal, and trying to figure out what choices other people will make so you can adjust your 'best card' function to take advantage is the deckbuilding game.
Now, talk to people who like to think they play competitively, and they'll (probably) tell you that variance is anathema to competitive gaming, but they're wrong because they're playing magic and not competitive RoShamBo, where there's no variance at all and therefore is the perfect competitive game(?). Even within magic, the times that competitive players stopped playing where at the times when deck-building was at a variance-nadir; ravager affinity, caw-blade, people complained about standard delver a bunch (although it was more interesting to play due to the gameplay variance involved), people complain that theros mono-U all day every day was boring. Variance is important for fun, and 4-of saves people from themselves optimising it away.
Conclusion: play netrunner with 45 card decks, 3-of rule, influence costs being the best way to regulate deckbuilding, and a much, much lower risk of not being able to play the game due to card order in your deck.
good games surprise their players.
variance from pseudorandom sources can make a game more skilltesting, fun, and rewarding. it can also do very much the opposite. it is not required or important for fun games at all.
very successful and fun competitive games like the starcraft series and the entire fighting game genre have little to no variance. they are very difficult to master and also fun.
the only pseudorandom effects i can think of in starcraft: broodwar are:
1) starting positions
2) miss chance on units shooting uphill
on a balanced map, #1 doesn't matter that much. #2 typically happens so much that excluding bizarre outliers, the random aspect tends toward not being that significant. i think sc2 even did away with #2 but im not 100% sure on that, i'm sure Jason or someone else that played that game knows.
most fighting games have little to no pseudorandom elements, typically the ones that do exist are character specific. modern fighters have almost none, and the few instances of it are viewed almost entirely negatively and are considered unfun and remove skill-testing elements from the game. the only one i can think of that i've never heard someone dislike is faust's items in guilty gear.
on the other hand, removing variance from magic would almost certainly decrease both the fun and the skilltestingness of the game. skill in magic comes from having to guess what your opponents have, and also around planning for best case/worst case scenarios of what you may be drawing. my experience playing hearthstone (which has no land cards, you just get "land" every turn guaranteed) really felt less intriguing mostly because of the less flexibility in bluffing or whatever inherent in mtg's system
what matters here is the source of the hidden information in the games.
in Starcraft, this is accomplished mostly by fog-of-war. a big part of the skill of the game is figuring out the hidden information based on shown information. "my opponent should have made roughly 500 gas by now, but i've only seen 100 gas worth of units, where's the rest of that spent" and so forth. the complexity of the starcraft resource system is relatively difficult to model in a player's head and requires a strong sense of timing and observation ("how many workers have I seen? how long has he had them? have those workers been mining gas or minerals?"). a system like this is probably impossible to work in a non-electronic game, most hidden game actions on hidden zones don't work very well without a judging party present (the computer in this case). i suppose you could write them all down and then reveal them but that seems quite tedious.
in fighting games, this is accomplished by many of the options being essentially too fast to react to, so in a sense you get rock-paper-scissors except much more complicated and [imo] much more interesting. in rock-paper-scissors, you make a hidden choice and reveal simultaneously. in fighting games you often make choices to do moves that start their effects within a handful of frames (1/60th of a second in most games), which may as well be hidden information until you've actually started to do it. some actions are fast enough to react to if you are looking for them but not if you aren't specifically looking for that subset of options. again this just can't work for a turn-based game.
so in magic, your hand is hidden, and to some degree the contents of your deck is vague. occasionally you have face down exiled cards or morphs, but mostly the hidden information is your hand. have you ever played magic where everyone plays with their hand revealed? it's a lot less interesting. to some degree cards like thoughtseize negatively affect this aspect of the game, and oversaturating a format with those effects can be negative. but at the same time, you draw again next turn and then there is hidden information back in the mix. and as soon as you have a brainstorm + shuffle or something similar, the peek effect is almost entirely gone. if you watch good players self-commentate, they always talk about playing around a card the opponent could have or not have, or playing with the possibility of drawing or not drawing whatever card. the variance makes the game much much more interesting. the possibility of non-games because of manaflood and manascrew is a bad awful thing, but i honestly believe nobody has come up with a good solution to the problem.
playing around and with hidden information is very important and difficult in all 3 of these games, and it is a major part of the skills involved. it's also fun.
also, i'm not a fan of chess.
variance from pseudorandom sources can make a game more skilltesting, fun, and rewarding. it can also do very much the opposite. it is not required or important for fun games at all.
very successful and fun competitive games like the starcraft series and the entire fighting game genre have little to no variance. they are very difficult to master and also fun.
the only pseudorandom effects i can think of in starcraft: broodwar are:
1) starting positions
2) miss chance on units shooting uphill
on a balanced map, #1 doesn't matter that much. #2 typically happens so much that excluding bizarre outliers, the random aspect tends toward not being that significant. i think sc2 even did away with #2 but im not 100% sure on that, i'm sure Jason or someone else that played that game knows.
most fighting games have little to no pseudorandom elements, typically the ones that do exist are character specific. modern fighters have almost none, and the few instances of it are viewed almost entirely negatively and are considered unfun and remove skill-testing elements from the game. the only one i can think of that i've never heard someone dislike is faust's items in guilty gear.
on the other hand, removing variance from magic would almost certainly decrease both the fun and the skilltestingness of the game. skill in magic comes from having to guess what your opponents have, and also around planning for best case/worst case scenarios of what you may be drawing. my experience playing hearthstone (which has no land cards, you just get "land" every turn guaranteed) really felt less intriguing mostly because of the less flexibility in bluffing or whatever inherent in mtg's system
what matters here is the source of the hidden information in the games.
in Starcraft, this is accomplished mostly by fog-of-war. a big part of the skill of the game is figuring out the hidden information based on shown information. "my opponent should have made roughly 500 gas by now, but i've only seen 100 gas worth of units, where's the rest of that spent" and so forth. the complexity of the starcraft resource system is relatively difficult to model in a player's head and requires a strong sense of timing and observation ("how many workers have I seen? how long has he had them? have those workers been mining gas or minerals?"). a system like this is probably impossible to work in a non-electronic game, most hidden game actions on hidden zones don't work very well without a judging party present (the computer in this case). i suppose you could write them all down and then reveal them but that seems quite tedious.
in fighting games, this is accomplished by many of the options being essentially too fast to react to, so in a sense you get rock-paper-scissors except much more complicated and [imo] much more interesting. in rock-paper-scissors, you make a hidden choice and reveal simultaneously. in fighting games you often make choices to do moves that start their effects within a handful of frames (1/60th of a second in most games), which may as well be hidden information until you've actually started to do it. some actions are fast enough to react to if you are looking for them but not if you aren't specifically looking for that subset of options. again this just can't work for a turn-based game.
so in magic, your hand is hidden, and to some degree the contents of your deck is vague. occasionally you have face down exiled cards or morphs, but mostly the hidden information is your hand. have you ever played magic where everyone plays with their hand revealed? it's a lot less interesting. to some degree cards like thoughtseize negatively affect this aspect of the game, and oversaturating a format with those effects can be negative. but at the same time, you draw again next turn and then there is hidden information back in the mix. and as soon as you have a brainstorm + shuffle or something similar, the peek effect is almost entirely gone. if you watch good players self-commentate, they always talk about playing around a card the opponent could have or not have, or playing with the possibility of drawing or not drawing whatever card. the variance makes the game much much more interesting. the possibility of non-games because of manaflood and manascrew is a bad awful thing, but i honestly believe nobody has come up with a good solution to the problem.
playing around and with hidden information is very important and difficult in all 3 of these games, and it is a major part of the skills involved. it's also fun.
also, i'm not a fan of chess.
Starcraft has loads of variance and you talk about it.
I'm protoss, you're zerg, we're on Xel'Naga Plateau. Am I going four-gate or fast factory? Are you going quick ling speed or going to fast roaches? If I go fast factory and you go speedlings, do you skip roaches to get mutas, and do I try and expand recklessly to go stalker/immortal?
I haven't played SC2 in a long time.
All of that is the variance in the game. There isn't a random element that means that the games vary, but we have to come to the game with a 'deck', in this case a starting build order, and then have to adapt to whatever the other player is doing. In fact, it even has the bad variance that everyone hates when you can get close spawns on maps that favor zerg on close spawns, and then the various tournaments disable close spawns so you actually can only horizontally or diagonally and not vertical spawns. This is the mana screw they're eliminating.
In fighting games, there's plenty of variance. I'm going to use divekick as my example because divekick is esports. If I'm the baz because why wouldn't you be and you're kung pao, I'm going to react differently to if you're dr shoals. I may move forward as my first move instead of backward. I mightjump dive less because dr shoals has a high jump and a shallow kick, whereas kung has a lower jump and a slower kick. That's all gameplay variance. There aren't any 9-1 matchups which would be the bad variance, and if there were, we just wouldn't play the 1. Noone plays pichu competitively in SBMM because its stupid to, and if competitively you were randomly assigned a character to play, everyone would bitch about the variance. Wait, talking smash bros we can observe that no items, fox only, final destination, because that removes as much variance as possible, but no two games like that will really be the same. Even dive vs dive mirror matches in divekick aren't the same.
Hidden information is sort of a modifier on variance. It isn't variance by itself, but you need variance in there to actually make it worthwhile instead of everyone just writing their orders down diplomacy style (and even then there's variance because BACKSTABBING).
BTW, the answer is I'm going dark templar. Because fuck you they're great.
I'm protoss, you're zerg, we're on Xel'Naga Plateau. Am I going four-gate or fast factory? Are you going quick ling speed or going to fast roaches? If I go fast factory and you go speedlings, do you skip roaches to get mutas, and do I try and expand recklessly to go stalker/immortal?
I haven't played SC2 in a long time.
All of that is the variance in the game. There isn't a random element that means that the games vary, but we have to come to the game with a 'deck', in this case a starting build order, and then have to adapt to whatever the other player is doing. In fact, it even has the bad variance that everyone hates when you can get close spawns on maps that favor zerg on close spawns, and then the various tournaments disable close spawns so you actually can only horizontally or diagonally and not vertical spawns. This is the mana screw they're eliminating.
In fighting games, there's plenty of variance. I'm going to use divekick as my example because divekick is esports. If I'm the baz because why wouldn't you be and you're kung pao, I'm going to react differently to if you're dr shoals. I may move forward as my first move instead of backward. I might
Hidden information is sort of a modifier on variance. It isn't variance by itself, but you need variance in there to actually make it worthwhile instead of everyone just writing their orders down diplomacy style (and even then there's variance because BACKSTABBING).
BTW, the answer is I'm going dark templar. Because fuck you they're great.
I'm leaving and never coming back
CML
Contributor
Games with another player will include some amount of variance that is indistinguishable from psychology. When computers get involved psychology is indistinguishable from complexity and quality. There is no way to make a good game without some element of luck and if you disagree with this you are wrong.
Onderzeeboot
Ecstatic Orb
I would argue that chess is a good game without any element of luck. It's completely deterministic, but contains enough moving pieces that the human brain can't think more than a few steps ahead. Any advantage a player gets is based on either a superior capacity to think ahead (calculating the best move) or taking advantage of your opponent's lack thereof (capitalizing on mistakes). The variance in chess is not luck based.
There are good dexterity-based games that aren't really based on luck. Street Fighter games don't really have luck. I feel like you're just spouting CML.
yeah thats actually true but the post-sf2 games only have random chance for very situational character specific things (like vega can throw his mask/claw either forward or backward in sf4). sf2 has other randomness with the length of its input buffers too. most fighting games after sf2 had much less randomness than sf2, but compared to a game of magic even that in sf2 is pretty minimal
I know, but generally the randomness isn't the element that "makes" the game, and I'm sure the ones that have it could easily be functionally implemented sans random elements.
Even in games where there is no luck at all, there's always the other player potentially playing unpredictably, which is close enough to randomness that it counts. If the other player didn't or couldn't play unpredictably, it wouldn't be a game, it'd be an optimisation problem. When chess gets to a high enough level, things are so well discussed and situations prepared for, it is actually an optimisation problem, which is perceived as a problem in some chess circles.
FlowerSunRain
Contributor
"Luck" as can be usefully discussed in games comes from a three major flavors
Mechanically installed randomness: The game rules create events that happen beyond the players' control.
Hidden information: Players make decisions without complete information.
Multiplayer Chaos: The decisions of multiple opponents unevenly effect a player.
So you can have "luck" to some degree in a game of chess. If I don't know my opponent, I don't know what openings he is familiar with and I might pick, for whatever reason, one that he is intimately familiar with and lose a game that I would have easily won if I picked a different opening. The knowledge of my opponent's skill to responding to openings was Hidden Information, so making the correct decision was beyond my control. It was a metaphorical coin flip
On a game design level, chess has no mechanically installed randomness, no hidden information and no multiplayer chaos, so it is not wrong to consider it an abstract strategy game on a theoretical level and discuss it as if it has no luck. The variance common to all interactive games independent of their mechanics is more of a discussion onto itself rather then one useful when discussing a game in particular.
Mechanically installed randomness: The game rules create events that happen beyond the players' control.
Hidden information: Players make decisions without complete information.
Multiplayer Chaos: The decisions of multiple opponents unevenly effect a player.
So you can have "luck" to some degree in a game of chess. If I don't know my opponent, I don't know what openings he is familiar with and I might pick, for whatever reason, one that he is intimately familiar with and lose a game that I would have easily won if I picked a different opening. The knowledge of my opponent's skill to responding to openings was Hidden Information, so making the correct decision was beyond my control. It was a metaphorical coin flip
On a game design level, chess has no mechanically installed randomness, no hidden information and no multiplayer chaos, so it is not wrong to consider it an abstract strategy game on a theoretical level and discuss it as if it has no luck. The variance common to all interactive games independent of their mechanics is more of a discussion onto itself rather then one useful when discussing a game in particular.
That post was crystal clear. 10/10 A+
CML
Contributor
chess isn't great because the psychological element, which does exist, is easily outstripped by the. chess is a game one experiences as being defined by your errors and not your good plays.
playing against another person involves some randomness
i guess what i'm arguing is games that a computer can play well are trivial and dull
series of interesting decisions
Onderzeeboot
Ecstatic Orb
there's whole books on that question
i'm only 7/8ths through with it but i really recommend "The Art of Game Design: A book of lenses" by Jesse Schell, I started reading it a little bit ago because mark rosewater recommended it as his favorite book on game design, which piqued my curiosity. Honestly its become my favorite too at this point. I'm not a the biggest maro fan either, the book isn't really in that direction.
CML
Contributor
series of interesting decisions that matter. who cares what extraordinary subtlety you set up 5 turns ago during a game of EDH if someone just wrathed and someone else is comboing off? who cares about queenside structural weaknesses when you dropped a knight last move?
good games have to stimulate the memory and have to force their players to make and revise their judgments. they cannot, like chess, be games where you are defined not by what you do right but what you do wrong.
Onderzeeboot
Ecstatic Orb
How does chess not stimulate the memory (it's super important at higher levels of play), and how does it not make players revise their judgments (recognizing and acting on suboptimal plays by your opponent)? Every game involving skill is defined by what you do wrong to an extend, and Magic the Gathering is no exception. If I mulligan incorrectly, or make a bad attack or block, that might cost me the game. This does not mean either chess or Magic is a bad game.
Neither does luck make for a good game. I for one think Monopoly is one of the worst game ever because one player is always trailing behind and then has to sit through a miserable two hours of his last money being slowly (or quickly) being drained away once all the streets are sold. Luck does not a good game make, and neither does the absence of luck a bad game make.
I do agree interesting decisions that matter add to a good game. Notably, both Magic and chess are full of interesting decision that matter at every turn of play, whereas Monopoly is almost completely devoid of them.
PS. All of this does not mean you have to like chess of course, but calling it a bad game, that's just silly. I mean, you're contradicting yourself when trying to define what makes a game good and then telling chess hasn't got those characteristics when it clearly does.
Neither does luck make for a good game. I for one think Monopoly is one of the worst game ever because one player is always trailing behind and then has to sit through a miserable two hours of his last money being slowly (or quickly) being drained away once all the streets are sold. Luck does not a good game make, and neither does the absence of luck a bad game make.
I do agree interesting decisions that matter add to a good game. Notably, both Magic and chess are full of interesting decision that matter at every turn of play, whereas Monopoly is almost completely devoid of them.
PS. All of this does not mean you have to like chess of course, but calling it a bad game, that's just silly. I mean, you're contradicting yourself when trying to define what makes a game good and then telling chess hasn't got those characteristics when it clearly does.
A decision that doesn't matter isn't interesting. You could've done anything else.
Chess is a much reduced series of decisions, once you've memorised a bunch of openings, counter openings and counter counter openings. Top level chess is about doing that, although lower level chess is indeed a bunch of decisions. I sit with the ideal that games should be designed/balanced at the top level of play first, and for casual players second.