Feedback

Miata13BToday at 9:08 AM
@Wolf9466  curious how you would go about making the seed more random to truly make it ASIC resistant? Trouble I see with a truly random seed would be how to deploy it across the network. We already see the effect of Bitstreams not being propagated with the Mini+, imagine the effect of that across all devices.

Wolf9466Today at 9:08 AM
@Miata13B Simple! Glad you asked! :smile:
Block height - NOT timestamp. New seed is the hash of that block's header.
There should be a small grace period where PoW using the old seed (block header hash) will be valid, as well as PoW using the new one - after which, new one only.
It's still POSSIBLE to create an ASIC for, but at least it's not so trivially predictable it helps them out by being such.

Miata13BToday at 9:10 AM
How would an orphaned block effect that scheme?

Wolf9466Today at 9:10 AM
Orphans are orphans - this means, put simply, that block height in the valid chain.
The grace period should allow miners to wait for a couple blocks following it before switching, too!
This way, it'd take a good-sized reorg to make the miners oops.
Even in that case - wouldn't be for very long at all.