I’ve been reading Robin Linus’s "Binohash" paper
(Binohash:transaction introspection without softforks,https://delvingbitcoin.org/t/binohash-transaction-introspection-without-softforks/2288) and
trying to understand the grinding process.
My rough understanding:
- The spender chooses subsets of signatures
from a fixed pool embedded in the locking script - Each subset changes the scriptCode via
FindAndDelete before hashing - The goal is to find a subset where the resulting
sighash satisfies a leading-zero condition
So in practice this looks like a search problem
over combinatorial subsets rather than a simple
nonce space.
Compared to Bitcoin mining:
- Mining hashes a fixed-size header with a
changing nonce - Binohash requires modifying KB-sized script
data (via FindAndDelete) before each hash - Each transaction defines its own independent
search space
My questions:
-
Is it reasonable to think of this as a kind
of PoW, or is that a misleading analogy? -
What actually limits ASIC acceleration here —
the FindAndDelete step, the larger input size,
or something else? -
Would a specialized ASIC still provide
significant speedup, or is this inherently
closer to a GPU-style workload? -
Does the fact that each transaction defines
its own search space make this fundamentally
non-competitive, unlike mining where everyone
races on the same puzzle?
