- All-OP_RETURN blocks: If miners fill total 4 MB blocks with nothing however giant OP_RETURN outputs, what are the validation and propagation penalties?
They can’t. OP_RETURNs exist in transaction outputs, which aren’t topic to the witness low cost. The most important consensus-valid quantity of OP_RETURN knowledge in a block is just below 1 million bytes, considerably smaller than what is feasible with utterly peculiar segwit transactions. They’re are additionally far less complicated to validate; it is transaction inputs which are costly when it comes to computation and I/O phrases.
Even ignoring that, relay coverage continues to be restricted to 100 kvB transactions. A miner can in fact exceed that, however that was already the case, and their means to take action is unaffected by the (default) coverage adjustments in Bitcoin Core v30.
- Mempool flooding: What occurs if the community is flooded with 1000’s of very giant OP_RETURN transactions that linger unconfirmed? How does eviction carry out towards “fewer however heavier” junk txs?
Fewer bigger transactions is much preferable on this regard than many small transactions.
- Propagation asymmetry: Might sustained megabyte-heavy blocks widen propagation gaps between high-bandwidth and low-bandwidth miners, growing orphan danger and centralization stress?
That’s the reason the block weight restrict exists. Blocks could possibly be stuffed to 1 MvB (and as much as 4 MB serialized dimension) no matter coverage, and incessantly have been. If something, together with giant OP_RETURNs reduces the impression right here, as they don’t seem to be topic to the witness low cost 4x issue, so their inclusion displaces extra witness bytes than their very own serialized dimension.
- Storage development: Sustained 4 MB junk blocks would develop the chain by ~576 MB/day (~210 GB/yr). How effectively does Core’s blockfile and database design deal with such development, particularly for pruned nodes?
Sustained 4 MB junk blocks are already potential. OP_RETURN outputs cut back the disk development house, by not being topic to low cost.
- Charge market impression: If OP_RETURN demand competes straight with financial transactions, has price estimation been stress-tested for non-financial blockspace utilization?
I do not see how price estimation can be affected by, or is in any approach associated to, the consumer’s intent for transactions.
