Pool Reward Functions

[Marianna Blt]

Hello to everybody
I am a master student from Politecninco di Milano.

I am working on pool reward functions (how to redistribute the reward among miners).

What I am focusing on to is the incentive compatibility property proposed by Okke Schrijvers, Joseph Bonneau, Dan Boneh, and Tim Roughgarden.

In their work they assume that in every moment of the round (time interval between a full solution and another one) every miner knows exactly the total number of submitted shares.

This seems to me really far from reality.

By the way how can miners understand when a round is long or short if they do not have this information?

Thank You all in advance

[jonnybravo0311]

Your assumption of that information being far from reality is incorrect.  Virtually every pool provides this.  You, as a miner, know exactly how many shares you've submitted, and how many total shares the pool has hashed towards this round.

[Marianna Blt]

Why this information is still available if it is well known that it encourage pool hopping and other attacks?

Is there any way that prevent miner to know this exact number of total already submitted shares?

[VRobb]

The search button is your friend.  Learn it, use it, love it.  Then come back and ask some questions that actually mean something. 

[kano]

Why this information is still available if it is well known that it encourage pool hopping and other attacks?

Is there any way that prevent miner to know this exact number of total already submitted shares?

e.g. PPLNS - knowing the numbers provides no information about getting a better or worse payout in the future.

You are probably reading about Prop or some other recent bazaar payouts being used where knowing the information does allow you to game the pool.

Most pools are not stupid enough to use such payout mechanisms.