Topic: [Pokereum] - An Ethereum-Telehash based provably fair decentralized Poker - page 3. (Read 9774 times)

interesting project, wish you all luck with it

I wonder how many other people in that list were included without their notification

So just to clarify, I looked at Pokereum's whitepaper a bit and provided a few comments; I'm not any kind of official advisor. But will be interesting to see this progress.

remember guys we are seeking collaborations,  board members, community manangers, more devs etc...there are even marketing bounties for menial work see here : https://nxtforum.org/stark-industries-(nxtdice)/marketing-5288/

Looks crazy the project! Lets see!

Following this thread for sure, electrum has a lot to offer im glad this team is taking the advantage of it.

Best of luck, sign up on the website!

Thanks for sharing, I'll have to check it out.

Ok, these are pretty important distinctions.  Perhaps add that to the paper, saying it would be initially Tournament or Sit and Go style games.  Then yes, you argument about player randomization is valid.

Happy to help proof and contribute to the idea.

That's a very important distinction. When I said "based on tables" I meant to imply seats at those tables (obviously, full tables are immaterial to these stats). Sorry for being imprecise. Still, the rest of the point still stands: the more games are being played, the more table seats open up, ergo the more unlikely that two colluders will land at the same table.


Cash tables are on the roadmap, but because of the collusion aspect, we're going to be putting some very cool mechanics in place to rate-limit joins/quits by activity (i.e. the more you stick around in games, the more easier it will be to leave). So cheaters that keep table-hopping will quickly find themselves throttled in this activity.

corrected  this is why we need community participation. You all own this project!

Has anyone seen the bounties yet? There are free tokens available here: https://nxtforum.org/stark-industries-(nxtdice)/marketing-5288/

Ill sign up to the nxt forum tonight, and comment over there.

How can I add comments to the document?  You have a spelling mistake on page 12.

How will the system handle slow to act players?

"The punishment may rage from losing a round to getting kicked out from the table and losing the table stake depending on the frequency of offences."

Not entirely true.  The number of games in play does not matter; the number of open seats does.  Sites will table balance to ensure full tables are running.  By randomly allocating, as opposed to table balancing, it may degrade the player experience.

Cheats could wait for opportune times, such as when all tables are full, or a table has < empty seats.

So its tournament only, not cash games?

Just read the white paper.  I saw that solution.  You might have issues with table balancing.


No offence taken.  I've played over 100,000 hands online (hit a RFlush online on Friday too!)  Been playing since 2002 and cashed in tournaments in the UK, Poland, Canada and Australia.  Just trying to make sure it cant be gamed.

One of my old work colleges used to work for Stars, doing heuristics on cheats.  So we often have discussions on this topic.

A few points of clarification:

• You don't have to be part of the DOA to enjoy the game. It will be made available to everyone
• Collusion requires players being able to play at the same table. Pokereum removes the ability for people to pick and choose their tables (randomized assignment as per the white paper description), thereby making it a statistical improbability that two or more colluders will land at the same tables.

I certainly don't want to appear hostile or defensive, but I get the sense that the criticism about collusion are born out of an incomplete understanding of what we are talking about (for this aspect of security). I'll try to sum it up colloquially here:
To collude in traditional online poker systems, two or more people are typically in contact with each other outside the game system and share knowledge of their hands without others at the table knowing. This happens today when online poker games let people pick their tables, and they can direct their colluders to the same table. Pokereum does not let you pick a table - you are assigned (according to our matchmaking algorithms). So two people are randomly assigned to tables. There are stats and maths to discusss how likely/unlikely these two people land in the same game, but intuitively, we know that as the number of games increase, the chances of them getting in the same game drop (rapidly).

So, the cheaters face two options (once they've chosen to take their chances at collusion):

• play the table assigned, thereby be unable to cheat
• quit the table (and lose the buy-in) and try again

Any team looking to try option number two will quickly find that the costs of quitting tables will always be higher than whatever they can get IF they ever land at the same table (esp. since BOTH cheaters have to walk away from their buy-ins).

I thought this concept was pretty clearly explained in the paper - but maybe the confusion around the topic is a good indicator that we should take another stab at that section.

Hope this clarifies things for you!

EDIT
PS: There will be a "play with friends" feature, but it's not fully spec'd out yet, so I don't have a lot to share with you on that part just yet.

I see "Ron Gross" discusses this point in the white paper.

Isn't this an oxymoron?  By implementing it as a DOA and anonymous players, collusion becomes very easy.

Reading the white paper now.  I'd been thinking about a similar project, but wouldn't have the resources to do it.  Looks good!

Thanks

will follow this thread. good luck

Hey guys, I'm part of the core dev team, and I'm lurking here on this thread. So if you have any questions about the project, I'd be happy to answer as best I can.