From pool to pool

Sorry for my english. How do you think it is possible after registering a purse on the adapool to register on another adapool. For example, if the original adapool was bad with a large% and a small turnover of funds. How fast can you go from one adapool to another?

Delegating your ADA to a pool is just a transaction that kinda “locks” your ADA, but it never actually leaves your wallet.

Revoking the ADA from a pool is just another transaction, and may be performed at any time. No pool approval is required.

So you just revoke your delegation to pool A and delegate to pool B.

As fast as two transactions (with strict happen-before relation) take to be executed. So somewhere between 40 seconds (for now) and few minutes.


Thank you. Everything is clear. I have several thousand Ada and I’m waiting for the start of the adapools.

1 Like

We expect there will be a lock-in period for staked ADA of at least 1 epoch (~5 days). The reason is that slot leaders are elected for the next epoch in advance. This should ensure that slot leaders will ‘turn-up’, of course assuming the nodes themselves are up and running. We should have more details once Shelley testnet is running.

If I understand it correctly - Ouroboros will take “a snapshot” of all active stake at the beginning of each epoch, so after this point, afaik, there’s actually no difference if pool still have all the same stake - for the whole time of the epoch it is considered to have the same stake as it had at the moment of the snapshot.

So, from the point of a staker user - he will be counted as a part of a pool (and receive reward) only if his stake was present in this pool at the moment of the snapshot. So it would make sense if user could withdraw his stake (certificate) at any point in time, but even if he withdraws his stake from pool A right after the snapshot and moves it into pool B - he is not considered to be a part of the pool B until the next snapshot.

On the other hand, it would open the case when user can lock his stake for the snapshot and then immediately withdraw it and spend it, which is, indeed, and undesirable behaviour (from my understanding), so there might be actually a lock-up time introduced in some form.

Anyways, looking forward to the Prof. Koutsoupias’ paper about all this stuff :slight_smile: