Duelingbook Forum

greg503 wrote:But how many seed values does it have?

Seed values? What do you mean?

Christen57 wrote:

greg503 wrote:But how many seed values does it have?

Seed values? What do you mean?

Machines are everything but random, I think he means, how many different values are generated to decide the outcome or which are the criteria to decide which outcome.

Lil Oldman wrote:

Christen57 wrote:

greg503 wrote:But how many seed values does it have?

Seed values? What do you mean?

Machines are everything but random, I think he means, how many different values are generated to decide the outcome or which are the criteria to decide which outcome.

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random

It says here the Math.random command generates a random number between 0 and 1, including 0, but not including 1. This means the possible numbers it could generate would be something like 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9.

The code I provided in my other comment is one that uses the Math.random command to generate one of these 10 random numbers, then uses the Math.round command to round it to the nearest integer. Simple as that. This means that if it generates 0, 0.1, 0.2, 0.3, or 0.4, it outputs 0, since any of those numbers rounded to the nearest integer equals 0, but if it generates 0.5, 0.6, 0.7, 0.8, or 0.9, it outputs 1, since any of those numbers rounded to the nearest integer equals 1.

So since there are 5 numbers it could generate that lead to an outcome of 0, and 5 numbers it could generate that lead to an outcome of 1, that makes it 50/50. It's equivalent to a coin flip.

Christen57 wrote:

Lil Oldman wrote:

Christen57 wrote:
Seed values? What do you mean?

Machines are everything but random, I think he means, how many different values are generated to decide the outcome or which are the criteria to decide which outcome.

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random

It says here the Math.random command generates a random number between 0 and 1, including 0, but not including 1. This means the possible numbers it could generate would be something like 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9.

The code I provided in my other comment is one that uses the Math.random command to generate one of these random numbers, then uses the Math.round command to round it to the nearest integer. Simple as that. This means that if it generates 0, 0.1, 0.2, 0.3, or 0.4, it outputs 0, since any of those numbers rounded to the nearest integer equals 0, but if it generates 0.5, 0.6, 0.7, 0.8, or 0.9, it outputs 1, since any of those numbers rounded to the nearest integer equals 1.

So since there are 5 numbers it could generate that lead to an outcome of 0, and 5 numbers it could generate that lead to an outcome of 1, that makes it 50/50. It's equivalent to a coin flip.

Saw the page and it says that

Is Pseudo random, thats what I (and probably greg) where reffering to. You Cannot just tell a Machine to do x thing randomly, it follows a criteria or a seed based on different factors (like the date and time of the day for example). On paper it may seem like it is 50/50, but in reality it may not be as fair.

Lil Oldman wrote:

Christen57 wrote:

Lil Oldman wrote:Machines are everything but random, I think he means, how many different values are generated to decide the outcome or which are the criteria to decide which outcome.

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random

It says here the Math.random command generates a random number between 0 and 1, including 0, but not including 1. This means the possible numbers it could generate would be something like 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9.

The code I provided in my other comment is one that uses the Math.random command to generate one of these random numbers, then uses the Math.round command to round it to the nearest integer. Simple as that. This means that if it generates 0, 0.1, 0.2, 0.3, or 0.4, it outputs 0, since any of those numbers rounded to the nearest integer equals 0, but if it generates 0.5, 0.6, 0.7, 0.8, or 0.9, it outputs 1, since any of those numbers rounded to the nearest integer equals 1.

So since there are 5 numbers it could generate that lead to an outcome of 0, and 5 numbers it could generate that lead to an outcome of 1, that makes it 50/50. It's equivalent to a coin flip.

Saw the page and it says that

Code: Select all

math.random

Is Pseudo random, thats what I (and probably greg) where reffering to. You Cannot just tell a Machine to do x thing randomly, it follows a criteria or a seed based on different factors (like the date and time of the day for example). On paper it may seem like it is 50/50, but in reality it may not be as fair.

It's still more fair than rock paper scissors. Besides, if it does in fact use the time of day, that too would be equivalent to a coin flip. In other words, if the current time of day is for example 6 o'clock, you could use the last digit of milliseconds in unix time or the last digit of milliseconds or something before it becomes 6:01, like this: https://currentmillis.com

So if it's 6:01:30:0987, where 6 is the hour, 01 is the amount of minutes after 6, 30 is the amount of seconds after 6 hours and 1 minute, while 0987 is the amount of milliseconds that passed since 6 hours, 1 minute, and 30 seconds, you could use the last digit, which in this case is 7, to determine whether to output 0 or 1. If that last digit is even, you treat that as 0, and if it's odd, you treat it as 1. Since the last digit constantly changes literally every millisecond, a millisecond is equal to 0.001 seconds, and the average human reaction time is only 0.2 seconds, that makes it impossible to accurately and consistently predict what the 50/50 outcome would be.

I assume duelingbook's coins and dice features are already coded this way, where they use the last digit of the amount of the milliseconds of the current time of day, and base heads/tails on the last digit of unix time in milliseconds.

Christen57 wrote:

Lil Oldman wrote:

Christen57 wrote:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random

It says here the Math.random command generates a random number between 0 and 1, including 0, but not including 1. This means the possible numbers it could generate would be something like 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9.

The code I provided in my other comment is one that uses the Math.random command to generate one of these random numbers, then uses the Math.round command to round it to the nearest integer. Simple as that. This means that if it generates 0, 0.1, 0.2, 0.3, or 0.4, it outputs 0, since any of those numbers rounded to the nearest integer equals 0, but if it generates 0.5, 0.6, 0.7, 0.8, or 0.9, it outputs 1, since any of those numbers rounded to the nearest integer equals 1.

So since there are 5 numbers it could generate that lead to an outcome of 0, and 5 numbers it could generate that lead to an outcome of 1, that makes it 50/50. It's equivalent to a coin flip.

Saw the page and it says that

Code: Select all

math.random

Is Pseudo random, thats what I (and probably greg) where reffering to. You Cannot just tell a Machine to do x thing randomly, it follows a criteria or a seed based on different factors (like the date and time of the day for example). On paper it may seem like it is 50/50, but in reality it may not be as fair.

It's still more fair than rock paper scissors. Besides, if it does in fact use the time of day, that too would be equivalent to a coin flip. In other words, if the current time of day is for example 6 o'clock, you could use the last digit of milliseconds in unix time or the last digit of milliseconds or something before it becomes 6:01, like this: https://currentmillis.com

So if it's 6:01:30:0987, where 6 is the hour, 01 is the amount of minutes after 6, 30 is the amount of seconds after 6 hours and 1 minute, while 0987 is the amount of milliseconds that passed since 6 hours, 1 minute, and 30 seconds, you could use the last digit, which in this case is 7, to determine whether to output 0 or 1. If that last digit is even, you treat that as 0, and if it's odd, you treat it as 1. Since the last digit constantly changes literally every millisecond, a millisecond is equal to 0.001 seconds, and the average human reaction time is only 0.2 seconds, that makes it impossible to accurately and consistently predict what the 50/50 outcome would be.

I assume duelingbook's coins and dice features are already coded this way, where they use the last digit of the amount of the milliseconds of the current time of day, and base heads/tails on the last digit of unix time in milliseconds.

true, a roll/coin is more fair than RPS. I was just answering the question of "Seed values? What do you mean?".

If you acknowledge who goes first or second should be random, and then look at two methods, one of which is random and the other one is "random" but has room open for manipulation (knowing a persons patterns because you follow them, ygoscope, etc...), there is no reason to not go with the more random method, unless you take advantage of the old system.

I think RPS has a little bit of skill in it, even without the whole YGOscope aspect. Throw YGOScope out of the picture, and you have no issues, and I'd even argue that having RPS rewards people who are good at prediction. It throws another aspect of strategy in the game, but not in a way that seems unfair. IRL, RPS would NEVER work, because of how easy you can cheat at it, but considering how none of that is possible on DB, I think it's pretty clear what the issue is. Allowing people to remove their names from YGOScope is fine I guess, but it then largely defeats the purpose of that part of YGOScope in the first place, since pretty much everyone would instantly remove their name from it. Instead, I think a timer for picking is fine, however, instead of merely making it so that the person is forced to lose the RPS if they do not chose in time, make DB automatically pick a random option when that time is up. This way, even if lag occurs, a viable option will be chosen, and the person wouldn't suffer an unfortunate consequence for something out of their control, but also wouldn't have enough time to check the other person's YGOScope and see their replays, unless by some chance they manage to tie RPS (which depending on how quick they can check the replays, they may have to tie RPS twice).

itsmetristan wrote:I think RPS has a little bit of skill in it, even without the whole YGOscope aspect. Throw YGOScope out of the picture, and you have no issues, and I'd even argue that having RPS rewards people who are good at prediction. It throws another aspect of strategy in the game, but not in a way that seems unfair. IRL, RPS would NEVER work, because of how easy you can cheat at it, but considering how none of that is possible on DB, I think it's pretty clear what the issue is. Allowing people to remove their names from YGOScope is fine I guess, but it then largely defeats the purpose of that part of YGOScope in the first place, since pretty much everyone would instantly remove their name from it. Instead, I think a timer for picking is fine, however, instead of merely making it so that the person is forced to lose the RPS if they do not chose in time, make DB automatically pick a random option when that time is up. This way, even if lag occurs, a viable option will be chosen, and the person wouldn't suffer an unfortunate consequence for something out of their control, but also wouldn't have enough time to check the other person's YGOScope and see their replays, unless by some chance they manage to tie RPS (which depending on how quick they can check the replays, they may have to tie RPS twice).

Skill isn't supposed to be a factor though. The method is supposed to be entirely random.

Monarch Snow wrote:

itsmetristan wrote:I think RPS has a little bit of skill in it, even without the whole YGOscope aspect. Throw YGOScope out of the picture, and you have no issues, and I'd even argue that having RPS rewards people who are good at prediction. It throws another aspect of strategy in the game, but not in a way that seems unfair. IRL, RPS would NEVER work, because of how easy you can cheat at it, but considering how none of that is possible on DB, I think it's pretty clear what the issue is. Allowing people to remove their names from YGOScope is fine I guess, but it then largely defeats the purpose of that part of YGOScope in the first place, since pretty much everyone would instantly remove their name from it. Instead, I think a timer for picking is fine, however, instead of merely making it so that the person is forced to lose the RPS if they do not chose in time, make DB automatically pick a random option when that time is up. This way, even if lag occurs, a viable option will be chosen, and the person wouldn't suffer an unfortunate consequence for something out of their control, but also wouldn't have enough time to check the other person's YGOScope and see their replays, unless by some chance they manage to tie RPS (which depending on how quick they can check the replays, they may have to tie RPS twice).

Skill isn't supposed to be a factor though. The method is supposed to be entirely random.

So can xteven or any of the admins implement this now or do we have to have a poll/vote on it first?

Christen57 wrote:

Monarch Snow wrote:

itsmetristan wrote:I think RPS has a little bit of skill in it, even without the whole YGOscope aspect. Throw YGOScope out of the picture, and you have no issues, and I'd even argue that having RPS rewards people who are good at prediction. It throws another aspect of strategy in the game, but not in a way that seems unfair. IRL, RPS would NEVER work, because of how easy you can cheat at it, but considering how none of that is possible on DB, I think it's pretty clear what the issue is. Allowing people to remove their names from YGOScope is fine I guess, but it then largely defeats the purpose of that part of YGOScope in the first place, since pretty much everyone would instantly remove their name from it. Instead, I think a timer for picking is fine, however, instead of merely making it so that the person is forced to lose the RPS if they do not chose in time, make DB automatically pick a random option when that time is up. This way, even if lag occurs, a viable option will be chosen, and the person wouldn't suffer an unfortunate consequence for something out of their control, but also wouldn't have enough time to check the other person's YGOScope and see their replays, unless by some chance they manage to tie RPS (which depending on how quick they can check the replays, they may have to tie RPS twice).

Skill isn't supposed to be a factor though. The method is supposed to be entirely random.

So can xteven or any of the admins implement this now or do we have to have a poll/vote on it first?

Xteven is open to the idea of implementation and with a survey.
I've conducted 3 surveys and have also used someone else's poll.
I'm waiting to conclude my final survey. I must acquire 58 more responses before I can break down the data and report my findings.

Three individual surveys and a poll created to address this issue found that participants in them all favored Dice (plural 2) for the method of being selected for RPS.

Monarch Snow wrote:Three individual surveys and a poll created to address this issue found that participants in them all favored Dice (plural 2) for the method of being selected for RPS.

Can you link me to this?

Christen57 wrote:

Monarch Snow wrote:Three individual surveys and a poll created to address this issue found that participants in them all favored Dice (plural 2) for the method of being selected for RPS.

Can you link me to this?

I can link you to the public opinion poll and one of the surveys.

https://strawpoll.com/sbszhf8xc (Poll - Created by Genexwrecker.)
https://www.survey-maker.com/Q4L2XUAOC (Survey #3)

For survey data, I'd have to export it on Discord.

Monarch Snow wrote:

Christen57 wrote:

Monarch Snow wrote:Three individual surveys and a poll created to address this issue found that participants in them all favored Dice (plural 2) for the method of being selected for RPS.

Can you link me to this?

I can link you to the public opinion poll and one of the surveys.

https://strawpoll.com/sbszhf8xc (Poll - Created by Genexwrecker.)
https://www.survey-maker.com/Q4L2XUAOC (Survey #3)

For survey data, I'd have to export it on Discord.

You should have made it be on this forum instead of an external site like strawpoll. Also aren't the "Coin toss" "Die roll" and "Let duelingbook automatically decide" options the same? If duelingbook tosses a coin or rolls a dice, it's the same as duelingbook automatically deciding.

Christen57 wrote:

Monarch Snow wrote:

Christen57 wrote:
Can you link me to this?

I can link you to the public opinion poll and one of the surveys.

https://strawpoll.com/sbszhf8xc (Poll - Created by Genexwrecker.)
https://www.survey-maker.com/Q4L2XUAOC (Survey #3)

For survey data, I'd have to export it on Discord.

You should have made it be on this forum instead of an external site like strawpoll. Also aren't the "Coin toss" "Die roll" and "Let duelingbook automatically decide" options the same? If duelingbook tosses a coin or rolls a dice, it's the same as duelingbook automatically deciding.

It's about the feel