On April 02 2017 03:49 Charoisaur wrote: Don't really agree. It's not strategic depth but a coin-flip because you have to make potentially game-deciding decisions before having the opportunity of gathering information. PartinG losing a GSL because he guessed wrong in game 7 was bullshit.
It isn't a coin filp. It is a skill.
The Patriots won a Superbowl believing the Seahawks were going to throw a slant based on their formation and the number of timeout Seattle had (Seattle was on the 1 yard line with the best rushing offense in the NFL, everyone thought they would run the ball). The Seahawks did throw a slant, and the Patriots intercepted the ball and won. But it wasn't randomness, it was preparation and calculated risk taking. But if the Seahawks didn't throw a slant and made their formation look like it, they might have been able run the ball in easily, and win the Superbowl.
So it isn't a coin flip at all. That is the kind of decision making that is present in every game, including LOTV (if I build an Oracle and without knowing I have a Stargate you place a Widow Mine in your mineral line, that isn't a coin flip, as Bill Belichick says, something might just not look right). The problem is that LOTV has removed a lot of the decision making from the game, and that is why it is stale.
You have to micro, have to macro, but the behind the scenes is significantly diminished. The preparation and build order planning, the skill I brought to Starcraft, was beating my opponent with preparation before the game began with unique build orders behind the scenes
It's sad that I can't do exactly what Sun Tzu says all warfare is based on: deception, in a strategy game! I used to like to make it look like I'm taking a third and throw an all-in at you. Or make it look like an all-in while I take a hidden base. It forces you to scout, react, and think, not just mindlessly macro and micro. But while you're thinking on your feet, I'm executing a game plan I made long before the game. And that is how I won a lot games in WOL, by out thinking my opponent because I'm not great at micro or macro.
"All warfare is based on deception. Hence, when we are able to attack, we must seem unable; when using our forces, we must appear inactive; when we are near, we must make the enemy believe we are far away; when far away, we must make him believe we are near.” ― Sun Tzu, The Art of War
And that's why the Patriots win, behind the scenes the players are supported by a system that tries to understand what their opponent is going to do, and the counter it, before the game begins.
And thus, the outcome of that play in the Superbowl, just like Parting's GSL 7 game, was decided before the game began. That isn't a coin flip at all. Stating it is disrespectful to the skill and preparation that goes on behind the scene.
There is a reason we had so most repeat GSL Code S champions in the first year of the GSL, the most volatile of the all years in terms of gameplay. Starsense is real and a skill.
what a disgusting way to think about strategy
as far as Im concerned the only justification for Fog of War is removing it would make the game boring
Is this sarcasm? The whole point of strategy games is decision making based on incomplete information. This is a fundamental aspect of many games, ranging from poker to a variety of video games like RTS, MOBA, FPS. You can even find it in games with so-called "perfect information" like chess and fighting games. It is what distinguishes competitive games played by humans against humans from other genres of puzzles, challenges, feats of skill.
So yeah, "removing it would make the game boring". Starcraft is supposed to be a strategy game, not a comparison of who has the best micro execution. If the game and the players can no longer evolve strategically then what's the point?
The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
On April 02 2017 05:26 BronzeKnee wrote: [quote] It isn't a coin filp. It is a skill.
The Patriots won a Superbowl believing the Seahawks were going to throw a slant based on their formation and the number of timeout Seattle had (Seattle was on the 1 yard line with the best rushing offense in the NFL, everyone thought they would run the ball). The Seahawks did throw a slant, and the Patriots intercepted the ball and won. But it wasn't randomness, it was preparation and calculated risk taking. But if the Seahawks didn't throw a slant and made their formation look like it, they might have been able run the ball in easily, and win the Superbowl.
So it isn't a coin flip at all. That is the kind of decision making that is present in every game, including LOTV (if I build an Oracle and without knowing I have a Stargate you place a Widow Mine in your mineral line, that isn't a coin flip, as Bill Belichick says, something might just not look right). The problem is that LOTV has removed a lot of the decision making from the game, and that is why it is stale.
You have to micro, have to macro, but the behind the scenes is significantly diminished. The preparation and build order planning, the skill I brought to Starcraft, was beating my opponent with preparation before the game began with unique build orders behind the scenes
It's sad that I can't do exactly what Sun Tzu says all warfare is based on: deception, in a strategy game! I used to like to make it look like I'm taking a third and throw an all-in at you. Or make it look like an all-in while I take a hidden base. It forces you to scout, react, and think, not just mindlessly macro and micro. But while you're thinking on your feet, I'm executing a game plan I made long before the game. And that is how I won a lot games in WOL, by out thinking my opponent because I'm not great at micro or macro.
[quote]
And that's why the Patriots win, behind the scenes the players are supported by a system that tries to understand what their opponent is going to do, and the counter it, before the game begins.
And thus, the outcome of that play in the Superbowl, just like Parting's GSL 7 game, was decided before the game began. That isn't a coin flip at all. Stating it is disrespectful to the skill and preparation that goes on behind the scene.
There is a reason we had so most repeat GSL Code S champions in the first year of the GSL, the most volatile of the all years in terms of gameplay. Starsense is real and a skill.
what a disgusting way to think about strategy
as far as Im concerned the only justification for Fog of War is removing it would make the game boring
Is this sarcasm? The whole point of strategy games is decision making based on incomplete information. This is a fundamental aspect of many games, ranging from poker to a variety of video games like RTS, MOBA, FPS. You can even find it in games with so-called "perfect information" like chess and fighting games. It is what distinguishes competitive games played by humans against humans from other genres of puzzles, challenges, feats of skill.
So yeah, "removing it would make the game boring". Starcraft is supposed to be a strategy game, not a comparison of who has the best micro execution. If the game and the players can no longer evolve strategically then what's the point?
The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
On April 02 2017 03:49 Charoisaur wrote: Don't really agree. It's not strategic depth but a coin-flip because you have to make potentially game-deciding decisions before having the opportunity of gathering information. PartinG losing a GSL because he guessed wrong in game 7 was bullshit.
It isn't a coin filp. It is a skill.
The Patriots won a Superbowl believing the Seahawks were going to throw a slant based on their formation and the number of timeout Seattle had (Seattle was on the 1 yard line with the best rushing offense in the NFL, everyone thought they would run the ball). The Seahawks did throw a slant, and the Patriots intercepted the ball and won. But it wasn't randomness, it was preparation and calculated risk taking. But if the Seahawks didn't throw a slant and made their formation look like it, they might have been able run the ball in easily, and win the Superbowl.
So it isn't a coin flip at all. That is the kind of decision making that is present in every game, including LOTV (if I build an Oracle and without knowing I have a Stargate you place a Widow Mine in your mineral line, that isn't a coin flip, as Bill Belichick says, something might just not look right). The problem is that LOTV has removed a lot of the decision making from the game, and that is why it is stale.
You have to micro, have to macro, but the behind the scenes is significantly diminished. The preparation and build order planning, the skill I brought to Starcraft, was beating my opponent with preparation before the game began with unique build orders behind the scenes
It's sad that I can't do exactly what Sun Tzu says all warfare is based on: deception, in a strategy game! I used to like to make it look like I'm taking a third and throw an all-in at you. Or make it look like an all-in while I take a hidden base. It forces you to scout, react, and think, not just mindlessly macro and micro. But while you're thinking on your feet, I'm executing a game plan I made long before the game. And that is how I won a lot games in WOL, by out thinking my opponent because I'm not great at micro or macro.
"All warfare is based on deception. Hence, when we are able to attack, we must seem unable; when using our forces, we must appear inactive; when we are near, we must make the enemy believe we are far away; when far away, we must make him believe we are near.” ― Sun Tzu, The Art of War
And that's why the Patriots win, behind the scenes the players are supported by a system that tries to understand what their opponent is going to do, and the counter it, before the game begins.
And thus, the outcome of that play in the Superbowl, just like Parting's GSL 7 game, was decided before the game began. That isn't a coin flip at all. Stating it is disrespectful to the skill and preparation that goes on behind the scene.
There is a reason we had so most repeat GSL Code S champions in the first year of the GSL, the most volatile of the all years in terms of gameplay. Starsense is real and a skill.
what a disgusting way to think about strategy
as far as Im concerned the only justification for Fog of War is removing it would make the game boring
Is this sarcasm? The whole point of strategy games is decision making based on incomplete information. This is a fundamental aspect of many games, ranging from poker to a variety of video games like RTS, MOBA, FPS. You can even find it in games with so-called "perfect information" like chess and fighting games. It is what distinguishes competitive games played by humans against humans from other genres of puzzles, challenges, feats of skill.
So yeah, "removing it would make the game boring". Starcraft is supposed to be a strategy game, not a comparison of who has the best micro execution. If the game and the players can no longer evolve strategically then what's the point?
The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
That aside, removing Fog of War in StarCraft wouldnt make it a comparison of who has the best micro execution. It would still be about understanding the game strategically and tactically, and mechanics. What would make it boring, though, is how silly the games would play out.
The point of poker is that you can read your opponents and therefore determine the cards they have, which affects your calculations. In this sense it is not mathematically trivial, because your data is based on psychological interpretation using factors as your knowledge of his playing strength, 'tells', patterns, capacity for deception, preparation.
This is also why I mentioned fighting games, as you clearly see the same structure there: the necessity to predict your opponent's actions based on 'tells', or patterns in his movement, because you can't block an attack if you play in a purely reactive manner due to limits to human cognition such as reflexes.
Applying this to Starcraft 2 we discover the following: you are not blind, you have the option to scout and investigate your opponent's behaviors and react appropriately. If you fail to get sufficient information you can invest resources into acquiring more of it (scans, sacrificial scouts etc.). Based on the context of the game (your opponent's strength, history etc.) you can elect to play more safely, or to take more risks. You are not playing a computer who blindly gambles every game with perfect execution.
Maybe you can't win every game, but it's idiotic to pretend like the better player is not statistically favorable in a match with a system like this. Structurally it's sound, but Blizzard needs to secure that all the parts are in working order.
How one can label that which is out of one's control not as coin-flip is beyond me. If you are playing against someone who only ever has 6-pooled, there is no way to know they are going to 6-pool in their next game. You may expect it, you may assume it, you may account for that eventuality when deciding upon your opening of choice, etc however there is no certainty.
The reason why Fog of War can be tolerated is that a game such as StarCraft allows a strategically and mechanichally superior player to overcome bad luck reasonably consistently.
You're using coin-flip as a derogatory word, but I think it would be very foolish to dismiss all random behavior in games like this. Given sufficient random events, parity is the overwhelmingly likely possibility. Every successful competitive game is rife with randomness (or at least actions you could not predict), its advantages are manifold including creating more excitement for viewers, more options for players, deeper strategies. It is part of the wider idea that you want players to be able to react to unpredictable events, because that is what shows true skill. Anyone can memorize an opening, but not everyone can respond well to an unexpected move.
The pitfalls of "coin flips" in SC2 are well known, of course, but it's an implementation problem. Honestly, the fact that you would so casually dismiss a core aspect of RTS gameplay just tells me you don't know what you're talking about and are distracting from more useful discussion.
Reading is hard? If you read what I actually wrote you would have realized that I dont disagree with you. I literally wrote removing Fog of War would be a worse evil than keeping it. Somehow that statement makes you think I dismiss Fog of War???
The point is that there is sound theoretical justification for having hidden information in the game, and you haven't demonstrated an appreciation of this. If you remove fog of war as a mechanic, and the concept of hidden information in general, you are changing the game in a very fundamental way that will completely alter gameplay and in my opinion destroy the game. It's not just that removing it would break the balance and lead to "silly games", the point is that it is categorically a bad idea because you've just annihilated the border between e.g. chess and poker.
Poker where you can see all the cards is not fun, it's not just different, it's broken beyond repair and it can't be saved.
I do appreciate Fog of War despite of the inherent coinflipping it forces upon the game. I do appreciate all the clever ways one can go about dealing with incomplete information: how you scout, how you adjust, how you do small things to account for possibilities, oh trust me I appreciate these things a lot more than just about anyone.
On April 02 2017 20:27 BlackPinkBoombayah wrote: [quote] what a disgusting way to think about strategy
as far as Im concerned the only justification for Fog of War is removing it would make the game boring
Is this sarcasm? The whole point of strategy games is decision making based on incomplete information. This is a fundamental aspect of many games, ranging from poker to a variety of video games like RTS, MOBA, FPS. You can even find it in games with so-called "perfect information" like chess and fighting games. It is what distinguishes competitive games played by humans against humans from other genres of puzzles, challenges, feats of skill.
So yeah, "removing it would make the game boring". Starcraft is supposed to be a strategy game, not a comparison of who has the best micro execution. If the game and the players can no longer evolve strategically then what's the point?
The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
On April 02 2017 23:53 Grumbels wrote: [quote] Is this sarcasm? The whole point of strategy games is decision making based on incomplete information. This is a fundamental aspect of many games, ranging from poker to a variety of video games like RTS, MOBA, FPS. You can even find it in games with so-called "perfect information" like chess and fighting games. It is what distinguishes competitive games played by humans against humans from other genres of puzzles, challenges, feats of skill.
So yeah, "removing it would make the game boring". Starcraft is supposed to be a strategy game, not a comparison of who has the best micro execution. If the game and the players can no longer evolve strategically then what's the point?
The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
On April 03 2017 02:14 SKNielsen1989 wrote: [quote] The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
It simply means that how to calculate it is easy, straight-forward, trivial, etc
How fast computers can calculate in its entirety is something else which I did not comment on.
On April 03 2017 02:59 Puosu wrote: [quote] This is false. The least complex commonly played form of poker, Limit Texas Hold'em was only "kind of" (it's ~close~) solved last year, after years of work in academia. More complex (and more popular) forms of poker such as No Limit Texas Hold'em and Pot Limit Omaha, on the other hand, are, depending on how you calculate it, more complex, and possibly harder to solve than Chess.
edit: Try to think of a game that becomes more complex when each player is given perfect information. Now think of a game that becomes less trivial (more complex) due to withholding information. Now it should be plain to see that the amount of information given to each player at time of play is essential to complexity. (and ~strategy~) If anything, StarCraft would become more trivial with full maphacks.
?
At any given time, it's trivial to calculate the chances of all eventual scenarios.
Because which scenario occurs is out of your control it is by definition gambling.
Not so trivial. This is an example game from a paper released on the subject.
Solving this game using a standard CFR implementation (2 double-precision floats per canonical infoset-action) would require 1 093 904 897 704 962 796 073 602 182 381 684 993 342 477 620 192 821 835 370 553 460 959 511 144 423 474 321 165 844 409 860 820 294 170 754 032 777 335 927 196 407 795 204 128 259 033 (1.094 × 10138) yottabytes of RAM
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
It simply means that how to calculate it is easy, straight-forward, trivial, etc How fast computers can calculate in its entirety is something else which I did not comment on.
Sure, but that's irrelevant. By that definition, with only a few tweaks, Starcraft 2 is a game which is mathematically solvable given infinite computing power. But because we live in the real world this information is only interesting for mathematicians and not for people that play games. Not unless you can apply this knowledge to a real world situation.
Because that's what you (infamously) said: "The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game. "
You are implying that the fact you could potentially write out an equation to solve poker (if you had near infinite computing power and if you ignored human psychology) has some bearing on its status as a strategy game, but this connection does not exist. Poker mapping to some mathematical construct which fits a certain category is not necessarily related to poker as a strategy game played by humans. You can't jump from one to another, because when humans play poker they don't have a magic 8-ball with the required 1000000000000000000000000000000 bytes of memory to store the solution with them. Even if it existed (e.g. chess computers) they could just be made illegal and it still wouldn't matter.
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
It simply means that how to calculate it is easy, straight-forward, trivial, etc How fast computers can calculate in its entirety is something else which I did not comment on.
Sure, but that's irrelevant. By that definition, with only a few tweaks, Starcraft 2 is a game which is mathematically solvable given infinite computing power. But because we live in the real world this information is only interesting for mathematicians and not for people that play games. Not unless you can apply this knowledge to a real world situation.
Because that's what you (infamously) said: "The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game. "
You are implying that the fact you could potentially write out an equation to solve poker (if you had near infinite computing power and if you ignored human psychology) has some bearing on its status as a strategy game, but this connection does not exist. Poker mapping to some mathematical construct which fits a certain category is not necessarily related to poker as a strategy game played by humans. You can't jump from one to another, because when humans play poker they don't have a magic 8-ball with the required 1000000000000000000000000000000 bytes of memory to store the solution with them. Even if it existed (e.g. chess computers) they could just be made illegal and it still wouldn't matter.
You really dont know what you are talking about, huh. StarCraft has infinitely many scenarios and starting to apply algorithms to guide a computer to find desirable scenarios is anything but trivial - in a math sense and otherwise. Hence why people with an interest in developing machine learning have been looking in StarCraft's direction (DeepMind).
On April 03 2017 04:37 Puosu wrote: [quote] Not so trivial. This is an example game from a paper released on the subject. [quote] Michael Johanson 2013, Measuring the Size of Large No-Limit Games https://arxiv.org/abs/1302.7008
As per your comment on anything that is out of one's control being gambling... I don't care how you define gambling, but even if something happening is out of your control, you may just as well adjust to the chances of specific things happening and therefore guarantee yourself better chances of winning a game. This is very much so strategy, and games of imperfect information indeed become super complex because there are so many things you have to prepare for.
You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
It simply means that how to calculate it is easy, straight-forward, trivial, etc How fast computers can calculate in its entirety is something else which I did not comment on.
Sure, but that's irrelevant. By that definition, with only a few tweaks, Starcraft 2 is a game which is mathematically solvable given infinite computing power. But because we live in the real world this information is only interesting for mathematicians and not for people that play games. Not unless you can apply this knowledge to a real world situation.
Because that's what you (infamously) said: "The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game. "
You are implying that the fact you could potentially write out an equation to solve poker (if you had near infinite computing power and if you ignored human psychology) has some bearing on its status as a strategy game, but this connection does not exist. Poker mapping to some mathematical construct which fits a certain category is not necessarily related to poker as a strategy game played by humans. You can't jump from one to another, because when humans play poker they don't have a magic 8-ball with the required 1000000000000000000000000000000 bytes of memory to store the solution with them. Even if it existed (e.g. chess computers) they could just be made illegal and it still wouldn't matter.
You really dont know what you are talking about, huh. StarCraft has infinitely many scenarios and starting to apply algorithms to guide a computer to find desirable scenarios is anything but trivial - in a math sense and otherwise. Hence why people with an interest in developing machine learning have been looking in StarCraft's direction (DeepMind).
If you remove hidden information, then mathematically speaking Starcraft is not different from chess and can be solved with the same method. Let's say that you have 30 frames/cycles per second, that the average game length is 30 minutes and that every frame you have about 1000 possible actions. By my count that leaves a mere 10^150000 game states to calculate, which is only a trifle if you have infinite computing power.
If DeepMind is really serious about creating an SC2 AI they will cease their pitiful efforts using machine learning and switch to this obviously superior system of building a database of every theoretically possible game ever, which requires repurposing only a few trillion universes for data storage.
Face-up poker is mathematically trivial. Limit variants like Razz would still require proper play with pot odds evaluations. No-Limit would just be shoving blinds to whoever got the best starting hand.
Hidden information poker – the way we know and love it – is not trivial, in part because one player doesn't know what strategy another player is using. The margins for play decisions in certain situations (e.g. preflop raise range) are based on all the hands played before the current one and thus fluctuate constantly.
On April 03 2017 02:14 SKNielsen1989 wrote: The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game.
Now I wonder what your definition of a strategy game is... Unless you disqualify poker based on there not being a movement of army pieces (classic idea of strategy) I don't see why chess wouldn't be trivial too. The result of the game is fixed: it's either white always wins or (more likely) it's always a draw. Basically overcomplicated tic-tac-toe.
Bonus note: Triviality is one of the few things in mathematics that kind of has a vague definition in an otherwise very strictly defined world; what is considered 'easy' is to some degree a subjective verdict.
On April 02 2017 03:49 Charoisaur wrote: Don't really agree. It's not strategic depth but a coin-flip because you have to make potentially game-deciding decisions before having the opportunity of gathering information. PartinG losing a GSL because he guessed wrong in game 7 was bullshit.
You forgot Squirtle losing to Mvp (one of the greatest finals ever on sc2, btw).
On April 02 2017 03:49 Charoisaur wrote: Don't really agree. It's not strategic depth but a coin-flip because you have to make potentially game-deciding decisions before having the opportunity of gathering information. PartinG losing a GSL because he guessed wrong in game 7 was bullshit.
You forgot Squirtle losing to Mvp (one of the greatest finals ever on sc2, btw).
I think that a lot of this argument in this thread arises from the difference between tournament and ladder play. In tournament play a pro's opponent is often some one they have played/seen play before. this allows them to make educated guesses about there opponents strategy The ladder on the other hand which most of us play has a meta that we gamble on but we can't make the same kind of bets as we would if we knew our opponent well, rarely do you see the same player enough to form an accurate picture of how they play. This changes the amount of mind games that go on significantly and the amount of strategy vs coin flipping involved in decision making. On the ladder you often will lose because you made incorrect predictions about your opponents strategy without getting enough information from scouting to make informed decisions. In a tournament you can take more calculated risks because you can make better predictions about the range of strategies your opponent will feel comfortable using.
Knowing about your opponent's play style lets you make all sorst of educated guesses about how they will play. For instance my friend tends to focus a lot on offbeat zergling timings and runby harassment but rarely does he allin. Knowing this I can prepare myself against his preferred style, I can build extra bunkers and make sure I'm walled up while expanding safety but not being overly defensive. Tournament play Is a lot like this when innovation plays SoS he knows to expect a lot of crazy off beat builds and that he should be defensive. If he is playing zest he can have a higher expectation of a macro game with some kind of harassment or pressure opening. If he is playing stats he can expect the phoenix adept build. Based on his knowledge of his opponents Innovation can select strategies that accommodate there preferred style of play.
Most of us mortals who play sc2 mostly experience the game through playing the ladder. The ladder definitely has a predictable meta game that you can use to make educated guesses about what your opponents doing but outside of the meta all you have is the scouting information you can grasp. In some matchups at some times this information has been substantial. Sometimes you have been able to see gas timings, expo timings, first tech choice, ect. When matchups are like this you have enough information to make good educated guesses about what your opponent is doing, just like tells in poker can let you make good educated guesses. At other times though match ups have allowed players to lock out scouting and hide there build enough that you prity much have to flip a coin and hope you pick the right answer to there strat. For instance in hots Protoss had several Allin builds that could all. Branch from the same opening, same stalker+msc timing, same gas timing, same pylon placement and they could lock Terran out of scouting once they had units. In this meta the best you could usually do was make a blind guess of what build they were going. Sc2 has had a fair number of ladder metas that felt arbitrary and coin flippy . That's why some players view strategic choices as coinflippy. and others as poker like. It's match up and meta dependent. Since you have limited knowledge of your opponent you often can't make calculated plays and tend to have to rely more on luck.
^ In my opinion this is why this sort of completely anonymous ladder with a billion opponents to cycle through is not the objective, ideal implementation. If you look at virtually any game, like chess or poker, or sports, people prefer to play it against real opponents, whom you get to know and study. I think this culture exists to a degree in video games as well, if you look at pc-bangs, practice partners, old methods of finding a game where you first had to meet someone in chat and then play one or more games, I think all of these are superior to an anonymous ladder from a certain perspective? Of course all of these are inconvenient in some way.
At some point they introduced the dungeon finder in WoW, which let you almost immediately find anonymous people to group with, instead of the slog of having to advertise in public chats or bully guild mates. This had many advantages(time), but in my opinion reduced the quality of groups, because you no longer had the chance to screen players and you were no longer limited to players from your realm, who are known and reliable; and whose guild affiliation makes for interesting gossip. I guess this sort of development was inescapable, and you get a feel of progress and convenience marching on, at the expense of turning people into anti-social sheep. There were still solutions available, like being in the right guild, but you almost deliberately have to step outside the mainstream, and you are always tempted to just give in.
Other games like CS and LoL, being team games, are more naturally about human interaction, which I suspect leads to different cultures. I suppose a lot of people like the individualism and anonymity of Starcraft, it is just very nice to have the instant gratification of an immediate opponent in a 1v1 game, whom you don't have to interact with. In that sense, even if you were to promote or enable all these social features, like clans, local tournaments, pc-bangs, rematch features, team games, friend lists and so on, it still wouldn't make a significant dent unless you either sabotaged the ladder or radically changed the culture because the primary hurdle is the passivity of the player weened on the anonymous ladder.
An additional point about build orders is that you often have no information about your opponent during early phases of the game. And I don't mean scouting, what worries me is that the early clash of build orders can be pivotal even though you haven't had any real chance to interact with your opponent yet and therefore it is a blind guess. It is not like you got tricked by a friend who defied expectations, it is just a stranger whose play style is still a mystery, so even this element of human psychology is missing. But this is not the complete story, because obviously there is the metagame to provide information. Players can reason about each other, but it is mediated via this nebulous concept of the meta game. But players don't have an obligation to copy common play styles. It reminds me of some IdrA stream where he would whine that his opponent was obviously an allin player because of his stereotype about protoss players, and when they met the expectation IdrA was angry, and when they defied the expectation IdrA was angry also because his reaction was wrong.
On April 03 2017 04:52 SKNielsen1989 wrote: [quote] You dont seem to understand what trivial means in a mathematical sense.
All in all I dont really think we disagree much at all - our choice of words and definitions simply appear to be different.
So what exactly do you mean? You said something vague and then every time something takes a reasonable interpretation of your comment you respond by saying that we are all speaking a different language.
Also, something being "trivially mathematically solvable" is an absolutely irrelevant statement given realistic constraints. If poker is "trivially solvable", but it will just take 1000000000000000000000000 years to calculate all to the end, then it has no bearing on the world we live in.
Mathemathically trivial is not at all vague. It's not my fault you dont know what it means. Feel free to go educate yourself about its well-established and well-defined meaning. It's funny how you so readily claim a statement you dont even understand to be false.
Can you please explain what you mean by mathematically trivial, so that we can all marvel in your wisdom.
Why would I throw pearls before a swine such as yourself?
I would gladly explain it to you if you were honestly interested.
I am honestly interested, please tell me. And it's not just me, think of what other people might learn.
It simply means that how to calculate it is easy, straight-forward, trivial, etc How fast computers can calculate in its entirety is something else which I did not comment on.
Sure, but that's irrelevant. By that definition, with only a few tweaks, Starcraft 2 is a game which is mathematically solvable given infinite computing power. But because we live in the real world this information is only interesting for mathematicians and not for people that play games. Not unless you can apply this knowledge to a real world situation.
Because that's what you (infamously) said: "The fact that you refer to Poker as a strategy game says everything about how our definitions of strategy clearly differ from one another. Poker is a mathematically trivial gambling game. "
You are implying that the fact you could potentially write out an equation to solve poker (if you had near infinite computing power and if you ignored human psychology) has some bearing on its status as a strategy game, but this connection does not exist. Poker mapping to some mathematical construct which fits a certain category is not necessarily related to poker as a strategy game played by humans. You can't jump from one to another, because when humans play poker they don't have a magic 8-ball with the required 1000000000000000000000000000000 bytes of memory to store the solution with them. Even if it existed (e.g. chess computers) they could just be made illegal and it still wouldn't matter.
You really dont know what you are talking about, huh. StarCraft has infinitely many scenarios and starting to apply algorithms to guide a computer to find desirable scenarios is anything but trivial - in a math sense and otherwise. Hence why people with an interest in developing machine learning have been looking in StarCraft's direction (DeepMind).
If you remove hidden information, then mathematically speaking Starcraft is not different from chess and can be solved with the same method. Let's say that you have 30 frames/cycles per second, that the average game length is 30 minutes and that every frame you have about 1000 possible actions. By my count that leaves a mere 10^150000 game states to calculate, which is only a trifle if you have infinite computing power.
If DeepMind is really serious about creating an SC2 AI they will cease their pitiful efforts using machine learning and switch to this obviously superior system of building a database of every theoretically possible game ever, which requires repurposing only a few trillion universes for data storage.
On April 02 2017 03:49 Charoisaur wrote: Don't really agree. It's not strategic depth but a coin-flip because you have to make potentially game-deciding decisions before having the opportunity of gathering information. PartinG losing a GSL because he guessed wrong in game 7 was bullshit.
You forgot Squirtle losing to Mvp (one of the greatest finals ever on sc2, btw).
Except Squirtle got a build order win and choked.
It still hurts reading these comments after all these years ;__;
I still feel like I fell prey to bait and switch tactics in the LotV beta. I was drawn into the beta because of the removal of macro boosters. While Terran had a hard time during the short time that macro boosters were gone, I played some of the most enjoyable games ever in Starcraft 2. There was a clearly defined early game, mid game, and late game, and it felt like a real RTS game to me.
And then the macro boosters were added back, and the SCV count bumped to 12.
I'll say it again... People watch the games they enjoy playing. This is why Minecraft consistently has more viewers than SC2 despite being not as flashy as the latter. The game should be patched to make it fun to play and not just fun to watch. When few people have fun playing the game, few people will bother to watch.
On April 04 2017 23:28 Eternal Dalek wrote: I still feel like I fell prey to bait and switch tactics in the LotV beta. I was drawn into the beta because of the removal of macro boosters. While Terran had a hard time during the short time that macro boosters were gone, I played some of the most enjoyable games ever in Starcraft 2. There was a clearly defined early game, mid game, and late game, and it felt like a real RTS game to me.
And then the macro boosters were added back, and the SCV count bumped to 12.
I'll say it again... People watch the games they enjoy playing. This is why Minecraft consistently has more viewers than SC2 despite being not as flashy as the latter. The game should be patched to make it fun to play and not just fun to watch. When few people have fun playing the game, few people will bother to watch.
The macro mechanic removal was fun, but the queen's autoinject was terrible. They should have removed inject entirely and mucked with larva spawn rate/hatchery cost.