@asaefstroem congratulations brother, and great to have you back on core zk development. kip 16 was a master piece, cant wait to see what you're going to do next.
@KaspaHub@maxibitcat the only exception would be based zk covs, which are effectively rollup style and introduce separate state. that framework is for heavy computation, not asset accounting. balances should remain native via inline covenants where enforcement happens in the utxo model.
@KaspaHub@maxibitcat no, that’s not true. if you’re using native kas, there’s nothing to bridge. asset state lives in the utxo set, and state transitions are enforced via covenants. nothing leaves l1.
zk proving is hardware intensive on any chain, so the per prover cost is similar everywhere. if eth zk proving is around 100k in hardware for competitive latency, vprogs provers will be in roughly the same order of magnitude. the difference is not the price of a prover, it is how many provers the system needs. eth uses zk for scaling and execution, so aggregate compute demand is massive. kaspa only uses zk for programmability, so total proving demand is far lower even if each prover is similarly priced. proving is also getting cheaper over time as hardware improves and proving systems get more efficient. vprogs are zk agnostic, so they are not locked into one proving system and can adopt cheaper and faster proving options as they become available.
@Hashdogsonkaspa@aKaspamaxi@michaelsuttonil powerful applications drive demand. kaspa is designed to host the most powerful applications securely at scale. everything else is noise.
eth chose this path largely due to constraints in its original execution model. the assumption has been that general purpose execution at scale must be pushed offchain, with l1 reduced to verification and settlement. vprogs challenge that assumption by allowing computation offchain while l1 still performs the native state transition. that model is generally viewed as infeasible today, which is why it has not shaped eth’s design. when eth claims to have solved the trilemma it is mostly a reframing where tradeoffs are redistributed rather than eliminated, with execution authority no longer native to the base layer.
no they’re different eth pushes ledger execution offchain and then asks l1 to verify a foreign state transition the base layer is no longer the system advancing state it just checks another ledger so the tradeoff is moved not solved vprogs push computation offchain but l1 itself still performs the native state transition by updating utxos there is no external ledger and no foreign state being verified execution authority stays on the base layer