Field note
What a Crypto-to-Fiat Conversion Really Costs
“Cheapest” is not a trading-fee percentage. The amount of fiat that becomes usable depends on the offered pair, explicit fees, custody exposure, settlement delay, trust assumptions, failure risk, and records needed for tax and accounting.
That framework matters especially when comparing a live exchange with a protocol that is still a published architecture. PayMoney has no production price, fee, latency, or failure distribution to put in a comparison table yet.
Price and fee are different fields
In the PayMoney target, the maker commits one FungibleV1 asset and quantity plus one ISO fiat currency and expected integer-minor-unit amount. The protocol does not consult a rate feed or reprice the pair. A taker can compare those fixed terms with other routes before claiming it.
Two fees have different jobs. The inherited FeeV1 output pays for aggregator liveness on every Action. On Solve, FiatOfferV1 also splits the selected escrow asset between the taker and PayMoney's collector according to immutable offer terms, rounding toward the taker. Neither policy has a defensible public figure before deployment measurement.
Custody and trust are separate costs
A centralized route temporarily turns the user's asset into a claim on an exchange and may add withdrawal review. PayMoney leaves the selected asset in resource-owned escrow instead of a platform account, but its launch fiat-evidence leg trusts one self-hosted notary. A colluding notary and taker can forge a claimed payment fact.
Those exposures are different, not absent. A useful comparison asks who controls value at every stage, who can forge the evidence, and what remedy exists if the expected next event never arrives.
Time is a distribution, not a promise
PayMoney's target needs direct MPC-TLS, local proof generation, epoch inclusion, and Ethereum confirmation. Network bandwidth, bank behavior, proving capacity, and L1 congestion can dominate. Any single “settles in N minutes” claim before real rail-template measurements would be a guess.
Compare percentile completion times and failure rates from production evidence, not happy-path demos. Expiry is measured in absolute settled epochs rather than wall-clock minutes.
The amount bound removes one ambiguity and leaves a hard edge
The observed fiat amount must satisfy expected ≤ observed ≤ expected + Δ, where Δ is one global deployment policy. Currency, recipient, memo, claimant, rail template, and completion status must also match.
The upper bound limits overpayment loss, but an irreversible mistaken transfer cannot be repaired by an operator. Preflight reduces the chance of paying when the only evidence path is already unavailable; it cannot eliminate a failure that begins later.
The honest comparison today
For a live route, collect the actual pair, fees, completion distribution, custody terms, evidence trust, and failure remedy. For PayMoney, evaluate the architecture and launch gates, not a fictional price. A fair cost comparison becomes possible only after a deployment publishes offers, fee policy, rail templates, notary terms, and measured completion data.