An idea for all game programmers/developers

[Arkhan]

Randomness isn't just luck. Skill is still involved. An example would be sports. Every time an athlete plays a sport it is a different game every time. It isn't just luck and since the randomness is controlled by rules it is an even playing field. The real world is governed by rules and randomness.

 

Randomnes still involves luck.

 

 

Look at a game like Gradius.

 

Predictable. You can memorize the enemy patterns and beat the game via practice and skill..

 

Imagine playing it where the starting waves that offer you crucial powerups just fly out at random.

 

Sometimes it would be physically impossible to shoot them all.

 

 

So, you'd have Hank who plays and gets an option and a lazer already...

 

and Steve who's like DAMNIT ALL I GOT WAS A SPEED UP AND A MISSILE.

 

 

Hank can just laugh his ass off and hold the button down and pummel crap.

 

Steve will be crying instead of laughing.

 

 

 

 

 

I got away with my randomness in Insanity (Berzerk clone) because half of the skill of the game comes from knowing how far you can go before the robot is alert enough to even think about shooting. The original Berzerk game was relentless anyways. If you don't know how to pick the bots off before you're in their firing radar, you're often hosed. The randomness I introduced though, was in shot probability, and it levels out to where the slight differences go pretty un-noticed. Once you practice at the game, you can consistently get high scores and try to out-do each other. I know this is the case, because there was a high score contest for the game where a few people really busted their asses to get better and farther.

 

I made some errors in the game at its later stages that make it sort of daunting. I went back and fixed them for the Xbox 360 release. The Otto head was way too OP and awful . Rookie mistake.

 

if the randomness I introduced had been something besides the shot chance... say, random shot speeds or random walk speeds, the game would be broken.

 

 

One person could get rooms where the bots move slower than dick and their shots might as well go backwards...

 

and then the next person could end up with a room where the robots run like they are in the Olympics, and shoot at you at warp speeds. It would be unfair.

 

If you want to have a truly fair scoring comparison, you need to do so with games that run the same every time, or the randomization that occurs is something that player skill/reflexes more than compensate for, such as the Berzerk example.

 

The shot probabilties boil down to "dumb, almost alert, kind of alert, alert, extra alert, and LOL don't even try to walk by".

[Schizophretard]

Yes, in a sense randomness involves luck. In my sports analogy, part of the randomness is the other player's freewill. To use your Gradius example, imagine if you bought a football game that worked the same as Gradius. All the AI players would have the exact same behavior. Only your behavior would change the game. If you did the exact same thing then the game would be identical every time. You would be able to "memorize the enemy patterns and beat the game via practice and skill." That wouldn't be football because not knowing what the other players will do is part of the game.

 

I see what you are saying about Gradius but the first time you play it it feels random because you don't know what the enemy patterns will be the first time you see them. You have to learn of the future before you can memorize it. You can only do that once because after you beat the game you can't restart the game and have that learning and memorizing the future experience. Now imagine that you had the option to have the learning and memorizing the future experience every time you played it and still had the option to play "time lines" that you already beat. It would be Gradius as you know it now plus Gradius as you knew it the first time you played it.

[Arkhan]

The thing is, Gradius still has SOME randomness to it. The enemies don't shoot the same every time. The patterns are the same, but there are slight differences in the timing of shots that makes it still require alertness.

 

 

I can beat a few old sports games by running a Blitz every turn on defense and a hail mary on offense every turn.

 

Sports games on old systems all have very predictable behavior. Sports Talk Baseball for Genesis, I can run a home run every time as long as I get to first base. You just screw with the basemen til they chase you and then throw the ball and then you run back, etc. until you get home.

 

I cant think how many times I played that game and beat it in the first inning because I got 11 points.

 

 

now, what if that goofy rule about 10 runs changed every game.

 

 

Or what if sometimes the bases went the other way, or the bases got closer?

 

What if sometimes you couldn't steal and sometimes you could.

 

What if sometimes, homeruns count as foul balls?

 

 

Those are random things that if you introduced them, would ruin the game for everyone playing. You'd never know how it was going to work.

[TMR]

If I read that right, it sounds like you're saying the Interstate was badly designed because you don't know where the cars will be.

 

i'm not sure how you get there from what i've said, but if you're planning on using the Interstate for a game then the answer is yes, it's badly "designed" because it was never meant to be used that way.

 

In contrary to your statement, Tetris is a prime example of great randomization. The opponents in Tetris are time and randomization.

 

But i didn't actually say anything about Tetris or other randomly generated games in that context; my point has always been the one you directly replied to, that random games have "no even playing field to compete against anyone else for score" and, as you've just said, Tetris isn't designed to have human components competing for score in that way.

 

[SlowCoder]

i'm not sure how you get there from what i've said, but if you're planning on using the Interstate for a game then the answer is yes, it's badly "designed" because it was never meant to be used that way.

 

 

 

But i didn't actually say anything about Tetris or other randomly generated games in that context; my point has always been the one you directly replied to, that randommotion * games have "no even playing field to compete against anyone else for score" and, as you've just said, Tetris isn't designed to have human components competing for score in that way.

"Calm ...calm ..." *waves hands in a relaxing motion and smiles kindly*

 

I was responding to KidQuaalude's post, not yours.

 

But since you mention it, your argument to my argument about the Interstate is a nonpoint. What does whether it was designed for it have to do with the fact that traffic is random? The same argument could be made for city traffic.

 

Here's another thing to think about. Auto racing. 2 guys pull up to a start line. One of them has a 200HP Mustang, the other a 120HP Civic. Somehow the Civic smashes the Mustang. Why? Because maybe the Civic driver, though he had a lower powered car, had the skill to drive, whereas the Mustang driver didn't so much. I think the differences in the stats of the cars could relate to "randomisation", and the driver's skill is the ability to overcome the randomisation.

 

And are you sure Tetris was never designed for multiplayer? I'm sure I saw 2 player arcade cabinets at the arcade.

Edited September 20, 2012 by SlowCoder

[TMR]

I was responding to KidQuaalude's post, not yours.

 

You didn't say and it could equally have applied to either, so i randomly answered it. That's the problem with randoms, they cause things you weren't expecting to happen. =-)

 

But since you mention it, your argument to my argument about the Interstate is a nonpoint.

 

Because it's answered as a response to my post. Now i have context, i'm not exactly sure why you introduced it as a response to KidQuaalude.

 

I think the differences in the stats of the cars could relate to "randomisation", and the driver's skill is the ability to overcome the randomisation.

 

At the start of the race the playfield is uneven but remains that way throughout whilst in Tetris the playfield starts off even (an empty well) and is constantly changing throughout.

 

And are you sure Tetris was never designed for multiplayer? I'm sure I saw 2 player arcade cabinets at the arcade.

 

Apparently it initially was meant to be two player, but all the prototypes and early iterations are for one person. But since i've just read that many iterations of Tetris aren't actually random, that sort of knocks the discussion on the head until we find out which ones are random?

[Arkhan]

There has to be some sort of pattern to tetris:

 

http://www.youtube.com/watch?v=jwC544Z37qo

 

 

It might just be in knowing every possible playable combination of current block + upcoming block, and stuff.

 

 

I'm not sure. I only like Tetris on NES and Gameboy, and it's mostly because of the music and the rocketship.

[TMR]

There has to be some sort of pattern to tetris:

 

Many iterations are pseudo-random rather than truly random.

[Schizophretard]

The thing is, Gradius still has SOME randomness to it. The enemies don't shoot the same every time. The patterns are the same, but there are slight differences in the timing of shots that makes it still require alertness.

 

 

I can beat a few old sports games by running a Blitz every turn on defense and a hail mary on offense every turn.

 

Sports games on old systems all have very predictable behavior. Sports Talk Baseball for Genesis, I can run a home run every time as long as I get to first base. You just screw with the basemen til they chase you and then throw the ball and then you run back, etc. until you get home.

 

I cant think how many times I played that game and beat it in the first inning because I got 11 points.

 

 

now, what if that goofy rule about 10 runs changed every game.

 

 

Or what if sometimes the bases went the other way, or the bases got closer?

 

What if sometimes you couldn't steal and sometimes you could.

 

What if sometimes, homeruns count as foul balls?

 

 

Those are random things that if you introduced them, would ruin the game for everyone playing. You'd never know how it was going to work.

 

Then it wouldn't be baseball. You can't just put randomness anywhere. Some places it makes sense and some it doesn't. It doesn't make sense to randomize the rules of baseball but it does make sense to randomize the behavior of the AI in a way that you aren't sure exactly what they will do.

[Schizophretard]

At the start of the race the playfield is uneven but remains that way throughout whilst in Tetris the playfield starts off even (an empty well) and is constantly changing throughout.

 

What if two players were taking turns playing a racing game that is based on score and the AI makes different decisions for all the other cars every time you play it? In other words, there is no pattern to the cars to memorize and they behave like it's a real race. Would the play field be uneven?

[Arkhan]

What if two players were taking turns playing a racing game that is based on score and the AI makes different decisions for all the other cars every time you play it? In other words, there is no pattern to the cars to memorize and they behave like it's a real race. Would the play field be uneven?

 

 

Yes.

 

In a real race, you don't race for a highscore. You all go at the same time and see who wins.

 

duh.

 

 

 

 

and with respect to the baseball analogy: The point is, some things are not made to be random. "random" enemy behavior AI in baseball is sort of mental anyways. What would you do, have the shortstop randomly decide to run the wrong way, or have an outfielder trip and fall sometimes?

 

Or have the first basemen decide to catch the ball and run away from the base?

 

 

To compete in baseball games, you do a player vs. player match. There is no AI. You control the people.

Edited September 21, 2012 by Arkhan

[Schizophretard]

 

 

 

Yes.

 

In a real race, you don't race for a highscore. You all go at the same time and see who wins.

 

duh.

 

 

 

 

and with respect to the baseball analogy: The point is, some things are not made to be random. "random" enemy behavior AI in baseball is sort of mental anyways. What would you do, have the shortstop randomly decide to run the wrong way, or have an outfielder trip and fall sometimes?

 

Or have the first basemen decide to catch the ball and run away from the base?

 

 

To compete in baseball games, you do a player vs. player match. There is no AI. You control the people.

 

Then Enduro has an uneven play field?

 

For baseball and sports games in general, I would make the players act as close to real players as I could with the limitations of the hardware.

[+Random Terrain]

Just about any game that can't be saved would probably be horrible. Imagine trying to beat Mario on NES if it was random every time you turned on your console. At least for games where there's a definite end to reach. Not so bad for high score games.

 

If those old games didn't toss you back to the beginning of a level or all the way back to the beginning of the game, controlled randomness could at least be used to place enemies and bonus items. The kind of platform game I'd like to play would use controlled randomness to place platforms, enemies, bonus items, and so on so I'm actually playing the game instead of simply dying and remembering. I love this quote about text adventure games and think it should apply to most other games too:

 

 

Consider, for example, the case of a room full of poison gas. The way to get through the room is to give the command HOLD BREATH before entering. If the character has no reason for holding his breath except that he choked to death in that room the last time he played, his actions become illogical.

 

However, things can be kept reasonable if the description of the previous room states that wisps of green mist are coming from under the door. Giving the command SMELL MIST might elicit a stronger warning, and then it would make sense that the character should take precautions. The point isn't that a really good player should be able to get through the adventure on the first try, but that the character should stay within the bounds of the game's reality.

 

~Gary McGath (From COMPUTE!'s Guide to Adventure Games)

 

 

Related links:

 

Randomness and Replayability

 

Alternatives to Constant Restarting

[Arkhan]

Then Enduro has an uneven play field?

IIRC, Enduro is the same every time you play it.

 

For baseball and sports games in general, I would make the players act as close to real players as I could with the limitations of the hardware.

 

Good luck.

 

 

EDIT: Are you done grasping at straws yet?

Edited September 21, 2012 by Arkhan

[TMR]

What if two players were taking turns playing a racing game that is based on score and the AI makes different decisions for all the other cars every time you play it? In other words, there is no pattern to the cars to memorize and they behave like it's a real race. Would the play field be uneven?

 

Yes it would because, although the patterns can't be memorised, there's still the problem of truly random car placement and movement not generating a consistent difficulty level throughout each race. Sometimes the field will be spaced out, others it'll be clumped together and, depending on a couple of other factors, one of those will be easier to deal with than the other, skewing the score of whoever lucks out.

 

[+RevEng]

There has to be some sort of pattern to tetris

 

Licensed Tetris games are supposed to adhere to the Tetris Guideline, which specifies how the random generator is supposed to work.

 

Random Generator generates a sequence of all seven one-sided tetrominoes (I, J, L, O, S, T, Z) permuted randomly, as if they were drawn from a bag. Then it deals all seven tetrominoes to the piece sequence before generating another bag.

[Arkhan]

Ah, so its not entirely random. That means you can develop strategies and know what sort of things could be coming up.

 

 

I bet the speed players keep a mental note of the 7 tetrosaurus rexes in their head and plan accordingly.

 

 

That's pretty cool.

[+Random Terrain]

Ah, so its not entirely random.

 

You wouldn't want the same piece coming up 20 times in a row, so of course you'd have to use some form of controlled randomness.

[Schizophretard]

 

IIRC, Enduro is the same every time you play it.

 

 

 

Good luck.

 

 

EDIT: Are you done grasping at straws yet?

 

I recommend you look again. Just tell me the pattern, order, color, locations,... of the first ten cars.

 

I would need a lot of luck because I don't know how to program. If I could, for an example, I would make sure that the player wouldn't know if the AI is going to hit a home run. There would be no patterns to memorize.

 

I don't feel like I am grasping at straws. This isn't something new. There are many games that aren't as identical as watching a movie every time you play them. I don't believe a game has to be 100% predictable to be an even playing field. In the game Yahtzee the die land randomly. It would be ridiculous to think that the game is unfair because the players don't get the same numbers as other players and the same order every time they play. The game is already fair, it has an even playing field, and the scoring opportunities are the same because all players are dealing with the same odds. It wouldn't be Yahtzee and would be boring if there wasn't some randomness to it.

[Schizophretard]

 

 

Yes it would because, although the patterns can't be memorised, there's still the problem of truly random car placement and movement not generating a consistent difficulty level throughout each race. Sometimes the field will be spaced out, others it'll be clumped together and, depending on a couple of other factors, one of those will be easier to deal with than the other, skewing the score of whoever lucks out.

 

But then you wouldn't know who is the better race car driver. You would just know who is better at memorizing a pattern.

[+Random Terrain]

I don't feel like I am grasping at straws. This isn't something new. There are many games that aren't as identical as watching a movie every time you play them. I don't believe a game has to be 100% predictable to be an even playing field.

 

I haven't read all of this thread, but is he trying to say that video games must be devoid of randomness? We've had dice games, card games, and board games that use dice or cards or both before video games came along. Those of us who were born before video games grew up on randomness. We expected a different experience every time we played.

 

Instead of reinventing the wheel, I'll just post what I've said before:

 

www.randomterrain.com/game-design-randomness-and-replayability.html

 

I was disappointed when I found out that Pac-Man had patterns you could learn. I wanted the ghosts to have a bit of randomness thrown in. They would still have their personalities, but you wouldn't be able to count on some stupid pattern to win. You'd actually have to play the game. Sure, the patterns can be ignored, but millions of people were trained that gaming was more about learning 'dance steps' and less about on-the-spot decision making and pure fun.

 

The public became used to games that were devoid of randomness where everything from bonus items to enemies were always in the same place. Level bosses with predetermined patterns to learn became the norm. Most people were brainwashed into believing that all games were supposed to be one-time static action puzzles to solve or a string of 'dance steps' to learn. It's as if they forgot all of those great board games they used to play that were full of randomness and replayability.

 

The reason I became interested in video games was because of all of the amazing possibilities. A computer is ideal for creating replayable games. If you want randomness, what could be better than a computer? You can go beyond anything that is possible with a board game, but that awesome potential is usually ignored or scoffingly dismissed.

[TMR]

But then you wouldn't know who is the better race car driver. You would just know who is better at memorizing a pattern.

 

That's probably why we don't use the same model for races in the real world, having a series of lone racers competing one at a time against a field of drivers who don't matter in the context of the competition isn't something you'll see on the telly.

 

This model can be made to work for games though; pseudo-random generators can produce what appear to be randomly generated races, but the same race can be repeated if the seed values plugged in at the start are used again. That means any number of players can all compete one at a time but, whilst they can't memorise the race in advance, it'll remain constant between games.

[Schizophretard]

 

 

I haven't read all of this thread, but is he trying to say that video games must be devoid of randomness? We've had dice games, card games, and board games that use dice or cards or both before video games came along. Those of us who were born before video games grew up on randomness. We expected a different experience every time we played.

 

Instead of reinventing the wheel, I'll just post what I've said before:

 

www.randomterrain.com/game-design-randomness-and-replayability.html

 

I'm not sure if Arkhan shares the exact same opinion as TMR's. If I'm understanding TMR correctly then he believes that if a game is based on score and two people are competing for a high score then the game has to be exactly the same with no element of randomness for it to be an even playing field. He views the element of randomness to be an element of good or bad luck and therefore each player has different scoring opportunities. Like if I play a game that has an element of randomness against you and I win. I may have beat you because the dice rolled in my favor. For an example, your game Seaweed Assault. It is full of controlled randomness. TMR would consider it an uneven playing field. For it to be a fair game all of the controlled randomness would have to be subtracted out. Every piece of seaweed would have to plop at the same times and at the same places. Every Wrothopod and canister would have to fall from the same spots. The only thing that would keep both of our turns from being like watching a rerun would be the behaviors of the players. If you were playing against me and on your turn you did the exact same thing as me then your turn would look like a recording of mine. In other words, even playing field equals no "dice".

 

My opinion is the opposite. I believe that it is an even playing field and the scoring opportunities are the same as long as the "dice" has the same number of sides and the players are using their skills against the same odds. An example would be Mexican Train Dominoes. Each player draws dominoes randomly during the game. I consider it an even playing field because all players are playing against the same odds with the same rules. The dominoes I drew ,any of the other players could have drawn. If I won all 13 games in a session I would believe I won a far game and it was my skill dealing with what dominoes I drew that made me the winner. If Arkhan and TMR were two of the other players they would probably say that I was just lucky because what I drew gave me better scoring opportunities and it wasn't fair that we all didn't have the same dominoes.

[Schizophretard]

 

 

That's probably why we don't use the same model for races in the real world, having a series of lone racers competing one at a time against a field of drivers who don't matter in the context of the competition isn't something you'll see on the telly.

 

This model can be made to work for games though; pseudo-random generators can produce what appear to be randomly generated races, but the same race can be repeated if the seed values plugged in at the start are used again. That means any number of players can all compete one at a time but, whilst they can't memorise the race in advance, it'll remain constant between games.

 

I understand that people taking turns playing Jeff Gorden doesn't happen on the telly but controlled randomness does. No race is identical.

[+Random Terrain]

I'm not sure if Arkhan shares the exact same opinion as TMR's. If I'm understanding TMR correctly then he believes that if a game is based on score and two people are competing for a high score then the game has to be exactly the same with no element of randomness for it to be an even playing field.

 

I already think that people should mostly play for fun and not for scores anyway:

 

http://www.randomterrain.com/atari-2600-memories-game-design-guidelines.html#resist_frustration

 

http://www.randomterrain.com/game-design-play-vs-competition.html

 

 

But a high score thread can get people interested in playing a game that they might ordinarily overlook, so it's not always a bad thing.

 

Real life isn't fair. There is no level playing field. In sports, conditions change throughout the game. Sometimes you get a better side of the court. Sometimes the sun is in your eyes. Sometimes you slip in a puddle of puke while the other guy is fresh as a daisy. The score is a fickle bitch that holds skill in one hand and luck in the other.

 

 

Sometimes you eat the bear and sometimes the bear eats you:

 

http://www.youtube.com/watch?v=B48D60wH8gA