bustabit v2 – Dilution fee lowered to 1%

[JackpotRacer]

............................................................................... .................as I shopped the offer around to everyone I was aware of that I thought would be honest and competent (although I avoided the whole "open" sale thing, to avoid a repeat of the moneypot disaster).

thx for admitting

good luck

What do you mean? Ryan probably doesn't want bustabit ( a long time project of his ) to be run by scammers, so he just offered around to people he trusted hence him saying ( to avoid a repeat of the moneypot disaster )

my posting was  regarding the moneypot disaster

[1531974983]

1531974983

[DarkStar_]

............................................................................... .................as I shopped the offer around to everyone I was aware of that I thought would be honest and competent (although I avoided the whole "open" sale thing, to avoid a repeat of the moneypot disaster).

thx for admitting

good luck

What do you mean? Ryan probably doesn't want bustabit ( a long time project of his ) to be run by scammers, so he just offered around to people he trusted hence him saying ( to avoid a repeat of the moneypot disaster )

I assume he's thanking for admitting that MoneyPot was a disaster.

[CoinXMan]

I watched a little. very few winners.

[CaliBookie]

I watched a little. very few winners.

Maybe you should ask the owner - Justin Malara why there are so few winners.  Start by looking at this:

The median should be about 1.98 - yet Bustabit is running 1.96 - while it may not sound like much it is actually a HUGE difference.

Now, devans (aka Justin Malara) will say that it is proven and will spit out an equation or 2 that "proves" it is a 1% house edge.  

When asked if he ran the entire 10 million numbers to see what the actual median is he stated no, it was unnecessary work.

If I am advertising a 1% house edge and I am getting a 2% house edge, what would you call that?  I call it cheating.  A stupid ass equation does NOT change the house edge.

Example, if I flip a coin 500,000 times and heads comes up 100,000 and tails 400,000 times - that is NOT variance - that is a fucked up quarter

I have not done the math but i am pretty fucking sure that >500,000 games is a large enough sample that the median should be pretty spot on and not 1.96

In case you want to ask devans aka daniel you can find him here:

Justin Malara
209 S Prospect Ave Apt 217
Redondo Beach CA 90277

And he looks like this:

https://i.imgur.com/bITbxVi.jpg

[RHavar]

The median should be about 1.98 - yet Bustabit is running 1.96 - while it may not sound like much it is actually a HUGE difference.

The expected median should indeed be 1.98x, but the actual median is going to vary significantly over a finite amount of games.  

Now, devans (aka Justin Malara) will say that it is proven and will spit out an equation or 2 that "proves" it is a 1% house edge.

That system is known as "provably fair" and contains absolutely everything you need to prove the game is fair. While i do admit it requires quite a bit of technical knowledge, it has the advantage that anyone can verify the game for everyone. So if you can find someone who understands how the maths primitives work, it should be easy to verify =)

 
When asked if he ran the entire 10 million numbers to see what the actual median is he stated no, it was unnecessary work.

If I was in his position, I would not do it either. If that knowledge affects your actions, I would say it's pretty unethical (as it'll either advantage/disadvantage  investors/gamblers). If that knowledge doesn't affect your actions, it's kind of pointless to know. I would definitely not want to put myself in a position where I could get influenced by it.


BTW, Daniel actually makes more money when the games are good than when the games are bad. His profit is a function of the amount wagered, and players are able to turn over a lot more when they're not getting raped.

If I am advertising a 1% house edge and I am getting a 2% house edge, what would you call that?  I call it cheating.  A stupid ass equation does NOT change the house edge.

The house edge can be 1%, but players lose 2%  .... but that doesn't mean there was cheating, it means there was variance. The provably fair system proves that the game outcomes have been decided from a probability distribution function that results in a 1% house edge. It's totally normal and expected that the games will be "good" and "bad" over a finite period of time

I have not done the math but i am pretty fucking sure that >500,000 games is a large enough sample that the median should be pretty spot on and not 1.96

I'd encourage you to do the maths, or get someone to do it for you. Variance is far more of a bitch than almost everyone realizes. In the end, bustabit has done all that is possible to allow you to verify the games and the maths and check things -- the ball is really in your court on this one.

[KloNEM]

I'm back in game after 14 days, and I see there were some changes made on  bustabit page :

1/ Why is not possible anymore to set up "non integer" bet, let's say 3.5 bits ? It was possible before. Of course real bet was 3 bits, but if one has multiplier (in case of loss), let's say 2.5x, next bet has been calculated as 3.5 * 2.5, not 3 * 2.5 ... which was fine. Actually, it was only one great upgrade (in my eyes), comparing to bustabit v1. Anyway, why's not possible to set up bet in satoshis instead of bits, so, in my example above, 350 sats instead of 3.5 (bits) ?

2/ When I play auto-mode, in case of loss the script has been terminated, even if I have set up multiplier (in case of win and "return to the base bet", everything goes on without interruption).

3/ What is the "Simulation mode" checkbox beside the "Run script" green button ?


Thanks for the answers to my questions.

[CaliBookie]

I have not done the math but i am pretty fucking sure that >500,000 games is a large enough sample that the median should be pretty spot on and not 1.96

I'd encourage you to do the maths, or get someone to do it for you. Variance is far more of a bitch than almost everyone realizes. In the end, bustabit has done all that is possible to allow you to verify the games and the maths and check things -- the ball is really in your court on this one.

Try this on for size, I think both you and Justin Malara both underestimate the probability of such a discrepancy.  Here are some numbers for you to entertain:

Current Median for 650,000 Bustabit games is 1.96 with an expected value of 1.98. (Here is the kicker, that is assuming the game is fair)  
That is a 50.51% vs 50% difference correct?

The likelihood of that being true IF the game is truly legitimate is (GRAB SOME POPCORN) - after 650,000 games is....

 
2.04386 x 10^-18

THAT IS A LOT OF FUCKING ZEROS --  - something like 1/489.3 quadrillion - NOW THAT IS A BIG FUCKING NUMBER - actually a very very small number

So, like I said, either the game is fucked up, the math is wrong or Daniel (aka Justin Malara) is cheating.  You choose

[CaliBookie]

The expected median should indeed be 1.98x, but the actual median is going to vary significantly over a finite amount of games.  

The problem here is the word "finite" - 650,000 games is more than enough games as seen in the post above.

If I was in his position, I would not do it either. If that knowledge affects your actions, I would say it's pretty unethical (as it'll either advantage/disadvantage  investors/gamblers). If that knowledge doesn't affect your actions, it's kind of pointless to know. I would definitely not want to put myself in a position where I could get influenced by it.

If the median for the 10 million games is 1.98 it SHOULD NOT affect either you or Justin Malara's decision.  On the contrary it should reassure you and the people playing that the game is legit.
NOW - if it is 2.0 or 1.96 (which it should not be over 10 million games) then I can see where the problem would be, although it should only occur about as often as guessing Satoshi's private key

The house edge can be 1%, but players lose 2%  .... but that doesn't mean there was cheating, it means there was variance. The provably fair system proves that the game outcomes have been decided from a probability distribution function that results in a 1% house edge. It's totally normal and expected that the games will be "good" and "bad" over a finite period of time

It has nothing to do with players losing, if the median is 1.96 and you advertise 1.98 then you are essentially stealing the extra.  Again "finite" is very arbitrary

[devans]

1/ Why is not possible anymore to set up "non integer" bet, let's say 3.5 bits ? It was possible before. Of course real bet was 3 bits, but if one has multiplier (in case of loss), let's say 2.5x, next bet has been calculated as 3.5 * 2.5, not 3 * 2.5 ... which was fine. Actually, it was only one great upgrade (in my eyes), comparing to bustabit v1. Anyway, why's not possible to set up bet in satoshis instead of bits, so, in my example above, 350 sats instead of 3.5 (bits) ?

bustabit only allows betting of whole bits, so you can only bet multiples of 100 satoshis. Previously a bug in the script editor allowed invalid bets to be accepted and silently rounded. This bug has been fixed and script authors need to ensure that their bet sizes are valid.

Luckily the fix is very simple. You can use Math.round to round your wagers, so for example

Code:

engine.bet(betSize, targetMultiplier)

would become

Code:

engine.bet(Math.round(betSize / 100) * 100, targetMultiplier)

If you always want to round downwards or upwards, use Math.floor or Math.ceil respectively instead.

2/ When I play auto-mode, in case of loss the script has been terminated, even if I have set up multiplier (in case of win and "return to the base bet", everything goes on without interruption).

In case that's what you're using, the Martingale sample script also relied on the bug I mentioned, but has been fixed. To load the fixed Martingale script, delete it from the list of scripts, then click the orange "Restore samples" button.

Otherwise it's hard to say what the issue is without seeing the specific script. Most likely it is trying to place an invalid wager and failing.

3/ What is the "Simulation mode" checkbox beside the "Run script" green button ?

The simulation mode lets you test a script without actually placing bets. If you enable simulation mode you will be able to choose your starting balance for the simulation. Running your script will then cause your simulated balance to increase or decrease, allowing you to see how a script would do on bustabit without risking any bits.

[CaliBookie]

Justin Malara aka devans

I would not respond to those REAL HARD math problems above either.  Leave it up to Ryan to bail you out.


 


===================================================================================

Justin Malara
209 S Prospect Ave Apt 217
Redondo Beach CA 90277

And he looks like this:

https://i.imgur.com/bITbxVi.jpg

[KloNEM]

Luckily the fix is very simple. You can use Math.round to round your wagers, so for example

Code:

engine.bet(betSize, targetMultiplier)

would become

Code:

engine.bet(Math.round(betSize / 100) * 100, targetMultiplier)

Ah, great, it really works. I just see (e.g. in Martingale script) function roundBit(bet), but editing of this function is full enough, it seems. Thanks!

In case that's what you're using, the Martingale sample script also relied on the bug I mentioned, but has been fixed. To load the fixed Martingale script, delete it from the list of scripts, then click the orange "Restore samples" button.

Otherwise it's hard to say what the issue is without seeing the specific script. Most likely it is trying to place an invalid wager and failing.

Yeah, it was that issue. I just deleted the script and used the "Restore sample" button and everything works like a charm now.

The simulation mode lets you test a script without actually placing bets. If you enable simulation mode you will be able to choose your starting balance for the simulation. Running your script will then cause your simulated balance to increase or decrease, allowing you to see how a script would do on bustabit without risking any bits.

Ok, thanks for the explanation, I'll try this, it looks like an useful function.


Thanks a lot for your explanation and also for all your work with bustabit, devans!

[devans]

Over the past few days KLITZ has won 300 BTC, making him the new record holder on bustabit for highest net profit. Congratulations!

[vbond]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

[hell696969]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

the games are predetermined. Nothing has changed besides the part of the chain we're on

[vbond]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

the games are predetermined. Nothing has changed besides the part of the chain we're on

Nothing is predetermined. You are deceived.
https://jsfiddle.net/Dexon95/2fmuxLza/embedded/result/  It shows only the history of the games and there is nothing to prove honesty.
Can you still believe in Santa?

[hell696969]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

the games are predetermined. Nothing has changed besides the part of the chain we're on

Nothing is predetermined. You are deceived.
https://jsfiddle.net/Dexon95/2fmuxLza/embedded/result/  It shows only the history of the games and there is nothing to prove honesty.
Can you still believe in Santa?

Ive played bustabit since 2015... Please learn about how the game works before you try to claim it was tampered. The games are hashed and chained together to prove the games havent been altered. If 1 game was changed or altered the entire chain would be broke and we would be able to tell.

[vbond]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

the games are predetermined. Nothing has changed besides the part of the chain we're on

Nothing is predetermined. You are deceived.
https://jsfiddle.net/Dexon95/2fmuxLza/embedded/result/  It shows only the history of the games and there is nothing to prove honesty.
Can you still believe in Santa?

Ive played bustabit since 2015... Please learn about how the game works before you try to claim it was tampered. The games are hashed and chained together to prove the games havent been altered. If 1 game was changed or altered the entire chain would be broke and we would be able to tell.

How to check the integrity? Tell me please.

[hell696969]

After KLITZ, there's nothing to do here.Bustabit went crazy.
The administrator reconfigured the script.
 Shiba: Median over last 50, 100 games: 1.535x, 1.595x

the games are predetermined. Nothing has changed besides the part of the chain we're on

Nothing is predetermined. You are deceived.
https://jsfiddle.net/Dexon95/2fmuxLza/embedded/result/  It shows only the history of the games and there is nothing to prove honesty.
Can you still believe in Santa?

Ive played bustabit since 2015... Please learn about how the game works before you try to claim it was tampered. The games are hashed and chained together to prove the games havent been altered. If 1 game was changed or altered the entire chain would be broke and we would be able to tell.

How to check the integrity? Tell me please.

You posted the link yourself... if a game was altered the whole chain would be broke and that verifier would show it