Pages:
Author

Topic: Just-Dice.com : Invest in 1% House Edge Dice Game - page 6. (Read 435362 times)

legendary
Activity: 2940
Merit: 1333
If you hold shift and hit TAB twice after editing the 'payout' field, you'll see the payout field gets adjusted to reflect the new rounded chance field.

I just made a couple of changes.

1) as soon as you leave the 'payout' field, the payout field is adjusted to match the displayed 'chance to win' value. So type 222222 into 'payout' then hit TAB or click outside the box and see it instantly change to 247500 (the multiplier for 0.0004%).

2) every time you hit a key, a 5 second timer is started. If you go 5 seconds without hitting a key, the 'payout' field is adjusted to match the displayed 'chance to win'. So type '222222' into the 'payout' field and then do nothing for 5 seconds. The 222222 should change to 247500.

I've put neither feature live, but you can try them out here:

    https://54.155.13.216/

That also has the /escrow, /cancel, /paid, /release commands for your testing pleasure. See Q7 in the FAQ for details, or just type /escrow in the chat for more terse usage information.

All feedback gratefully received, as always.

Edit: I improved the automatic 'fixing' of the payout multiplier. When you type a number that isn't one of the allowed multipliers, the field's background goes orange and a message appears in the top right telling you that it will be automatically fixed after 3 seconds of inactivity. It's up at:

    https://54.155.13.216/
legendary
Activity: 2940
Merit: 1333
That means when you send the coins to such a bitcoin address one could use the private key of that bitcoin address to import the clams into the clam wallet?

Exactly. 'dumpprivkey' in bitcoind, 'importprivkey' in clamd and the CLAMs show up.

I will modify the /dig command in Just-Dice so that it tells you CLAM address for any BTC address you give it:

I found that /help and !help show the commands but the latter seems not to be complete. !w is missing for example.

All Just-Dice chat commands begin with '/'. Typing "!help" simply causes you to say "!help" in the chat. If something is responding to that, it must be some kind of third party bot.

And i think it would be good when you could update the chance to win formfield.

I tried the following some time ago: Entered the value corresponding 1USD into betsize. Was around 0.6 clams. Then adjusted the payout to the value that would match the max profit. Then wanted to bet and... nothing happened. Even though all formfields were white.

I'll see if I can reproduce that. Put 0.6 into 'bet size', then by trial and error adjust the 'payout' box until the profit is the same as the max profit? (Do you realise how mad people are going to be if I bet 0.6 CLAM and win the max profit testing this for you?!) OK, so I see the problem. You're making a relatively small bet to win a huge amount. That's obviously going to need the chance of winning to be tiny. We have only 4 decimal places of granularity in the chance of winning, because we only roll 6 digit dice rolls (00.0000 to 99.9999). At 0.0004% chance of winning (247,500x) a winning 0.6 CLAM bet makes 148,499.4 profit, and at 0.0005% (198,000x) it makes 118,799.4 profit. The max profit (135k) is somewhere between these, and so you want to roll at something like 0.00045%, but we can't do that when we only have a million different roll outcomes. I notice that if I put 0.6 into the 'bet size' field and then edit the 'payout' field, the 'chance to win' field goes to 0.0005 when I type 215000 but to 0.0004 when I type 225000. It looks like I am seeing the chance field to the closest available chance to the one you're requesting:

>>> 99.0 / 215000 = 0.0004604 -> rounded to 0.0005%
>>> 99.0 / 225000 = 0.0004400 -> rounded to 0.0004%

I guess the question is: when the user asks for a 215000x multiplier, and we don't have a bet like that, should we round down to 198000x, up to 247500x, or round to whatever available multiplier is 'closest' in some sense? We round to the closest one. Sometimes that involves rounding the multiplier up, which puts the profit over the max profit, and so the bet is disallowed.

You say "nothing happened". When I tried, I saw this in the text area in the very top right, next to the 'Just-Dice.com' text:
    "That is over the current maximum allowed profit (135886.52)"
but it's not very obvious. The client should really detect that the bet isn't acceptable before sending it to the server, but it's tricky. The 'chance' depends on the multiplier, and the multiplier can only have 4 decimal places, and the multiplier depends on the rounded chance, so we have a circular dependency here. Changing the multiplier changes the chance, which gets rounded to 4 decimal places, which changes the chance, which changes the multiplier. But you're currently typing in the multiplier field, so it's really annoying if we edit it while you're typing it. A few sites do that, and it makes them really annoying to use.

I checked around a bit and found that you calculate bets with the chance to win value. And even though that value is white, there are too few decimals to keep up with the payout number. When i went to chance to win and left the number there how it was then the payout was lowered down to match the percent. Though the profit was only 2 third of max profit then.

If you hold shift and hit TAB twice after editing the 'payout' field, you'll see the payout field gets adjusted to reflect the new rounded chance field.

It seems its not possible to adjust the chance to win to the possible values in payout because the decimals are missing. I think it would be good to adjust it so that they can match each other.

There's a limit on how finely the chance can be adjusted. We only have a million different possible rolls, and so we can't make a bet with a percentage chance of winning which has more than 4 decimal places. We could change the roll algorithm to roll 8 or 10 digit roll numbers, but then I'd have gamblers up in arms. "I was on a 200 roll losing streak and I'm sure I was about to win when you changed everything. You cheater!".

At the moment i have to split my bets into 2 bets and have to reinvest the rest because i cant bet it against max profit.

I think you need to pick either 0.0004% or 0.0005% and adjust your stake accordingly. 0.54903858 CLAM at 0.0004% or 0.68629892 CLAM at 0.0005% will hit the max profit right now. If you put 0.0004 in the 'chance to win' field and then click the 'max' bet button, it will fill in the 0.549... value for you. You may have to hit it again if the max profit drops before you get to submit your bet.
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
And it seems there is no way to identify a winning user under biggest, with his ID, when he didnt post in the chat. If you search such id with !last then nothing happens. Is there another way?

I think it would be good when you would have a "Change" section in your biggest too. Instead losses and wins a change. Quicker to gasp. Maybe even with the amount of stakes? Though sure, that has nothing to do with bets, it would be more like a investors stat. Maybe a new document like that for investors would be nice... Checking with one view what hapenned in last timeframes. Staking and bankroll matters but should be divided.

Only my thought. Tongue
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
Dooglus, you might do something about old accounts from bitcoin just-dice that still have set up the old bitcoin address as emergency withdrawal address. I didnt find info that clams can be sent to a bitcoin address, only can be claimed from one, right?

It's possible to calculate the CLAM address that has the same private key as any given BTC address. So I can 'send CLAMs to a BTC address' in the same way that the initial CLAM distribution was done.

But people should update their emergency address to a CLAM address.

That means when you send the coins to such a bitcoin address one could use the private key of that bitcoin address to import the clams into the clam wallet?

Though that would probably not worth it anymore then. When just-dice dies then clams are dead too most probably. Smiley

I found that /help and !help show the commands but the latter seems not to be complete. !w is missing for example.

And i think it would be good when you could update the chance to win formfield.

I tried the following some time ago: Entered the value corresponding 1USD into betsize. Was around 0.6 clams. Then adjusted the payout to the value that would match the max profit. Then wanted to bet and... nothing happened. Even though all formfields were white.

I checked around a bit and found that you calculate bets with the chance to win value. And even though that value is white, there are too few decimals to keep up with the payout number. When i went to chance to win and left the number there how it was then the payout was lowered down to match the percent. Though the profit was only 2 third of max profit then.

It seems its not possible to adjust the chance to win to the possible values in payout because the decimals are missing. I think it would be good to adjust it so that they can match each other.

At the moment i have to split my bets into 2 bets and have to reinvest the rest because i cant bet it against max profit.
legendary
Activity: 2940
Merit: 1333
Dooglus, you might do something about old accounts from bitcoin just-dice that still have set up the old bitcoin address as emergency withdrawal address. I didnt find info that clams can be sent to a bitcoin address, only can be claimed from one, right?

It's possible to calculate the CLAM address that has the same private key as any given BTC address. So I can 'send CLAMs to a BTC address' in the same way that the initial CLAM distribution was done.

But people should update their emergency address to a CLAM address.
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
Dooglus, you might do something about old accounts from bitcoin just-dice that still have set up the old bitcoin address as emergency withdrawal address. I didnt find info that clams can be sent to a bitcoin address, only can be claimed from one, right?
member
Activity: 62
Merit: 10
Here is something I posted regarding a scammer going under the name Bittrex-Richie on Just-Dice & ON IRC. Please take time to read it and let us know if you've been affected.

https://bitcointalksearch.org/topic/scammer-alert-fake-bittrex-richie-360-clams-stolen-just-dicecom-1105662
legendary
Activity: 1199
Merit: 1047
If we have 100 bets per second and 1000 investors, currently the site has to do ~100 calculations per second. Each investor's share of the bankroll is fixed, and so I only need to keep track of the total bankroll.

If each investor has their own different Kelly factor, then their share of the bankroll changes after every roll, by a factor which depends on their Kelly factor and their share of the bankroll. That means doing 100k calculations per second. It's a huge difference. I wonder how other sites do it.

Well, even if adjusting the offsite investment after each roll would be perfect, I'd already be happy if it was adjusted automatically less often, like once per day.
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
There arent not even enough clams to make sense of this max profit nowadays. Nobody would gamble such a high amount of a currency, clams being in that small state have many new factors to think of.

Don't forget that you can play with a 990,000x payout multiplier. You can bet just 0.14 CLAM at the lowest chance of winning and win the max profit. So it's not about gambling a high amount, it's about a player getting lucky. If they won the max profit they would have a hard time cashing out into a different currency without crashing the price of CLAM, but they could invest their winnings, or just stake it, and have a very nice regular income. You can think of it as if they are playing to win a large share of JD's future earnings.

Using just dice as a kind of lotto. Interesting idea. I mean every invested Dollar in lotto is only a half dollar invested because the state takes the other half. At jd you can invest your full value and you would only lose 1% on average.

I think ill give it a shot some times. Tongue

* There's no way of setting your "Kelly factor". You can only approximate it using /offsite, and have to regularly adjust it to keep near the Kelly factor you want. Dropping the maximum profit to 0.1% from 0.5% only makes that situation worse.

At first glance my idea seemed like a good one, but now I'm much less happy with it. I think I would rather just reduce the maximum offsite multiplier to 20x than mess with the 0.5% figure the way I'm currently thinking.

No, why you think so? I first feared you would do that move, that would in fact hinder investors, that had 100x, of getting their coins back. I was happy when i heard your idea with lowering kelly because its ideal. Its lowering the risk very much and doesnt change anything to the way jd worked before in fact.

The optimal strategy for investors is to risk 1% of their investment on every bet, and so that's what some of them want to do. If I only let them risk 0.1%, they need to set their offsite amount to 9 times their onsite amount to compensate. That causes their onsite amount to chance 10 times more quickly than it would without the offsite setting, and means they need to regularly adjust their offsite amount to be 9 times their new onsite amount. I've not modelled this, but I'm sure if I did we would find that a "10x offsite, 0.1% risk" investment would diverge from the 'full Kelly' exponentially more quickly than a "2x offsite, 0.5% risk" investment does.

The reason i used 100x is because kelly only applies when the gamblers really would play to win max profit all the time. In fact thats not the case practically everytime. So all the normal bets have a risk where you could raise your kelly factor to thousands time offsite investment. Even the 2700? clam profit that king took in one bet yesterday are only 2% of max profit, which means everyone who had 100x enabled at that time would only have a kelly factor of 2 only. And 1 is the optimum. Thats why i asked for the feature on old jd, because i realized that i would gain an advantage nearly all the time. You only run into problems when bets are done that go near max profit.

Of course its still risky with higher bets, when they would happen. Id like a feature to automatically lower offsite investment when it looks like the bets get too big. Though i dont see that being implemented. Smiley

I see where the investors come from that want offsite max to be disabled but i surely would not be happy since i invested 100x exactly at the start of the "crash".  Roll Eyes Guess i have too much look in love. Cheesy
hero member
Activity: 568
Merit: 500
https://bit-exo.com/?ref=gamblingbad
Gamblers cant make the same arguement, when no one ever have been close to max bet and never lost on a max bet. And new max bet would be higher then anyone has bet before.

We don't have the concept of "max bet". We have "max profit", and gamblers are definitely making bets which would win the maximum profit if they won. So I think their (hypothetical) argument is valid: "I've been betting 0.14 CLAM at 0.0001% over and over, trying to hit the max profit. I failed so far, and now you want to cut the max profit by 80%? What's wrong, am I just about to win it? Unfair!"

Yes, but that arguement can be used whatever jd profit is
legendary
Activity: 2940
Merit: 1333
Gamblers cant make the same arguement, when no one ever have been close to max bet and never lost on a max bet. And new max bet would be higher then anyone has bet before.

We don't have the concept of "max bet". We have "max profit", and gamblers are definitely making bets which would win the maximum profit if they won. So I think their (hypothetical) argument is valid: "I've been betting 0.14 CLAM at 0.0001% over and over, trying to hit the max profit. I failed so far, and now you want to cut the max profit by 80%? What's wrong, am I just about to win it? Unfair!"
legendary
Activity: 2940
Merit: 1333
There arent not even enough clams to make sense of this max profit nowadays. Nobody would gamble such a high amount of a currency, clams being in that small state have many new factors to think of.

Don't forget that you can play with a 990,000x payout multiplier. You can bet just 0.14 CLAM at the lowest chance of winning and win the max profit. So it's not about gambling a high amount, it's about a player getting lucky. If they won the max profit they would have a hard time cashing out into a different currency without crashing the price of CLAM, but they could invest their winnings, or just stake it, and have a very nice regular income. You can think of it as if they are playing to win a large share of JD's future earnings.

* There's no way of setting your "Kelly factor". You can only approximate it using /offsite, and have to regularly adjust it to keep near the Kelly factor you want. Dropping the maximum profit to 0.1% from 0.5% only makes that situation worse.

At first glance my idea seemed like a good one, but now I'm much less happy with it. I think I would rather just reduce the maximum offsite multiplier to 20x than mess with the 0.5% figure the way I'm currently thinking.

No, why you think so? I first feared you would do that move, that would in fact hinder investors, that had 100x, of getting their coins back. I was happy when i heard your idea with lowering kelly because its ideal. Its lowering the risk very much and doesnt change anything to the way jd worked before in fact.

The optimal strategy for investors is to risk 1% of their investment on every bet, and so that's what some of them want to do. If I only let them risk 0.1%, they need to set their offsite amount to 9 times their onsite amount to compensate. That causes their onsite amount to chance 10 times more quickly than it would without the offsite setting, and means they need to regularly adjust their offsite amount to be 9 times their new onsite amount. I've not modelled this, but I'm sure if I did we would find that a "10x offsite, 0.1% risk" investment would diverge from the 'full Kelly' exponentially more quickly than a "2x offsite, 0.5% risk" investment does.
hero member
Activity: 568
Merit: 500
https://bit-exo.com/?ref=gamblingbad

[/quote]

Nothing is decided at all. I wanted to get feedback from people.

Two thoughts I hadn't had:

* Investors just lost a bunch; it would be unfair to drop the maximum profit now - but if we wait for profits to go back up, couldn't gamblers make the same argument? The fact is that almost nobody is playing for the maximum profit anyway, so the point is mostly moot.

* There's no way of setting your "Kelly factor". You can only approximate it using /offsite, and have to regularly adjust it to keep near the Kelly factor you want. Dropping the maximum profit to 0.1% from 0.5% only makes that situation worse.

At first glance my idea seemed like a good one, but now I'm much less happy with it. I think I would rather just reduce the maximum offsite multiplier to 20x than mess with the 0.5% figure the way I'm currently thinking.
[/quote]


Gamblers cant make the same arguement, when no one ever have been close to max bet and never lost on a max bet. And new max bet would be higher then anyone has bet before.
But who increased their investment or just started investing in JD have lost, and will use much longer time to get it back. 20x is still high if u want have a safe bankroll.
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
Well this is slightly unfortunate because now I think a lot more people will 20x and it'll be skewed a lot more towards 20x.

You can't do anything about setting your own Kelly. This is the system and basically you have to live with it.

You could possibly have trigger orders when there's a big move. Say you're 2x, if it goes to 3x you set 2x again. There'll be limit to how many trigger orders you can set. It might be a compromise.

While as someone who uses high leverage, 0.1% sounds great. It'd give me as an investor a greater share of the site profit versus a 0.5% floor. But, 0.1% is barely risking any capital.

Huh? You realize that the kelly criterion dooglus means is used for the max profit? Thats the 137k number. We never saw bets in that area, as far as i know, so you cant say that changing the kelly criterion will mean to barely risk any capital. We never risked capital worth 0.1% kelly. So it would not be a difference.

And more people will 20x? You know that the average already is above that? Its 47x.

I would not like seeing the offsite max lowered. It would screw investors that used 100x and lost big now.
legendary
Activity: 2674
Merit: 1083
Legendary Escrow Service - Tip Jar in Profile
* Investors just lost a bunch; it would be unfair to drop the maximum profit now - but if we wait for profits to go back up, couldn't gamblers make the same argument? The fact is that almost nobody is playing for the maximum profit anyway, so the point is mostly moot.

Yes, its exactly like you say in your last sentence. Nobody plays in the league of max profit. The highest wins were only fraction of max profit, maximum 2%. So its no argument that investors cant get their investment back then anymore. You could set down max profit to 2% and still nothing would have changed because simply no one played amounts like that yet. In fact i think it wouldnt be bad to lower the max profit even more. There arent not even enough clams to make sense of this max profit nowadays. Nobody would gamble such a high amount of a currency, clams being in that small state have many new factors to think of. For you, for people trading, investors, allowing liquidity, not crashing the market, making jd stablie and so on.

So i would say you could safely lower max profit to 0.01% and nothing would have changed as to before. Investors could not complain. In fact when an investor complains, the risk of losing even more than that is there, like tiger showed today with his single 2700? clams win. That was far away from max profit. And it still hurt very much.

* There's no way of setting your "Kelly factor". You can only approximate it using /offsite, and have to regularly adjust it to keep near the Kelly factor you want. Dropping the maximum profit to 0.1% from 0.5% only makes that situation worse.

At first glance my idea seemed like a good one, but now I'm much less happy with it. I think I would rather just reduce the maximum offsite multiplier to 20x than mess with the 0.5% figure the way I'm currently thinking.

No, why you think so? I first feared you would do that move, that would in fact hinder investors, that had 100x, of getting their coins back. I was happy when i heard your idea with lowering kelly because its ideal. Its lowering the risk very much and doesnt change anything to the way jd worked before in fact.
member
Activity: 65
Merit: 10
Hi, WTB 125 BTC worth of CLAMs. More details here: https://bitcointalk.org/index.php?topic=1103935.new#new

hero member
Activity: 756
Merit: 500
Well this is slightly unfortunate because now I think a lot more people will 20x and it'll be skewed a lot more towards 20x.

You can't do anything about setting your own Kelly. This is the system and basically you have to live with it.

You could possibly have trigger orders when there's a big move. Say you're 2x, if it goes to 3x you set 2x again. There'll be limit to how many trigger orders you can set. It might be a compromise.

While as someone who uses high leverage, 0.1% sounds great. It'd give me as an investor a greater share of the site profit versus a 0.5% floor. But, 0.1% is barely risking any capital. Those investors are basically just utilizing JD as a staking wallet. And as you know you sort of want to discourage that. Investing in JD is for backing the bankroll, you don't want people using it just for staking.
legendary
Activity: 2940
Merit: 1333
If I write "/offsite 2x" on chat, will the invested offsite amount automatically be adjusted so it's always the same as invested onsite? I have just tried it, but the invested offsite doesn't seem to change. I'd like to just bet 1% kelly, but having to manually adjust the invested offsite amount is really annoying. I'd have to do that even more often if you change to a default 0.1% of the bankroll.

No, /offsite 2x sets your offsite amount to be equal to your onsite amount at that time. It will immediately begin to drift from there.

I never found an efficient way of having your investment stay at 2x as it grows and shrinks, although I can see that that would be useful.

It seems there are two independent but related factors:

1) different people want to use different "Kelly" factors
2) different people want to keep different amounts of their wealth "offsite".

There's currently no way of expressing (1) at JD. You can only approximate it by setting (2) regularly, which I can see would be annoying.

It's more of Doog's way of staying "accountable," so to speak. While he is trusted not to do anything, there's nothing *stopping* him from doing it (or someone else, if they somehow got ahold of the site and its wallets). He's trying to keep it from being a centralized system (51% would effectively be centralization, in a currency that's meant NOT to be centralized).

I'm obviously conflicted. The more CLAMs staking at JD, the bigger my 10% cut of that staking is, and so ideally we'd have all the CLAMs.

But on the flip side, it simply isn't healthy for a coin to be so concentrated in a single location. Even if the site has perfect security and I am the perfect custodian for CLAMs, it still looks bad for a single wallet to contain more than 50% of the active CLAMs.

I guess if you see CLAM as "JD's casino chip" then it doesn't matter, but that's kind of rude. CLAM existed before JD, and is still cared about by its creators. I expect they have the same kind of conflict: they like seeing the increased demand for their coin that JD brings, but at the same time don't like seeing it monopolised by JD.

Im not sure why it shouldnt be viable. I mean the just-dice server has constantly to calculate things for the investors. Another var in the calculation doesnt matter really.

If we have 100 bets per second and 1000 investors, currently the site has to do ~100 calculations per second. Each investor's share of the bankroll is fixed, and so I only need to keep track of the total bankroll.

If each investor has their own different Kelly factor, then their share of the bankroll changes after every roll, by a factor which depends on their Kelly factor and their share of the bankroll. That means doing 100k calculations per second. It's a huge difference. I wonder how other sites do it.

What time will this happen? Will u set a date or when JD reach a new all time high?

Nothing is decided at all. I wanted to get feedback from people.

Two thoughts I hadn't had:

* Investors just lost a bunch; it would be unfair to drop the maximum profit now - but if we wait for profits to go back up, couldn't gamblers make the same argument? The fact is that almost nobody is playing for the maximum profit anyway, so the point is mostly moot.

* There's no way of setting your "Kelly factor". You can only approximate it using /offsite, and have to regularly adjust it to keep near the Kelly factor you want. Dropping the maximum profit to 0.1% from 0.5% only makes that situation worse.

At first glance my idea seemed like a good one, but now I'm much less happy with it. I think I would rather just reduce the maximum offsite multiplier to 20x than mess with the 0.5% figure the way I'm currently thinking.
hero member
Activity: 756
Merit: 500
dooglus, 10% might be so safe that everyone does it. You should adjust it to 0.1% to 20% if you're really going to do that.

Also I think if you only made the change if the bankroll was over 100k.
hero member
Activity: 568
Merit: 500
https://bit-exo.com/?ref=gamblingbad
Pages:
Jump to: