Question: Ev output about mithril

[pyrros]

Seems ok to me. Not sure if we need more precision, but might be good to align with what the rest of Cardano is doing if they are in fact using higher precision.

Adding after discussion w/ @markulf : Stake representation should also align with Cardano.

[abakst]

Adding after discussion w/ @markulf : Stake representation should also align with Cardano

Do you know what Cardano is using?

[abailly-iohk]

AFAICT, stakes are represented as Integer inside Cardano codebase

[iquerejeta]

Let's try to close this issue. @pyrros , @markulf , when you mentioned that stake representation should also align with Cardano, what were you expecting?

[pyrros]

The stake representation point was in addition to the initial one. For the stake representation as long as we are not casting to something smaller or different (integer to floating point or vice-versa) we should be ok.

The initial comment was about the bit precision on the vrf comparison.

[iquerejeta]

Oh, I see. And how would you go for the conversion? True that in the paper the mapping is defined as {0,1}*-> Z_p, and the image of phi is [0,1], and cannot see anywhere how to do the conversion. Actually, in our implementation we define the mapping as {0,1}* -> Z_{2^64}. Do we also want to change that?