After Midnight’s mainnet launched in March, I spent a week putting its fee model next to TRON’s Energy model. The clearest difference is the cost structure: sending one USDT transfer on TRON means either locking about 7,000 TRX for resources, renting Energy per transaction for a few cents to a few dollars, or burning roughly 6.4 TRX with no resources banked. Sending a transaction on Midnight only requires holding NIGHT — 1 NIGHT generates up to 5 DUST over a week, the transaction burns a little DUST, and the burned part is collected by no one.

One model charges per use; the other charges by holding. This post answers three questions: what DUST actually is, where it matches TRON’s Energy and where it doesn’t, and whether the rental market Midnight deliberately deleted will reappear in another form.

What DUST actually is

Midnight splits “capital asset” and “network fuel” into two components. NIGHT is a transferable token handling governance, block rewards, and DUST generation; DUST is a shielded resource with exactly one job — paying transaction fees. The whitepaper’s phrasing is that DUST “has value but cannot retain value”: it cannot be traded, cannot be gifted, and burns on use.

I checked the generation rules more than once. Each NIGHT maps to a cap of 5 DUST, filling linearly over one week — both rate and cap scale with the holding, so any amount takes exactly one week to fill. Move the NIGHT and the DUST decays at the same rate, reaching zero in a week; it stays spendable while decaying, and once at zero it stays zero. Fees follow a fixed structure: a fixed floor price, plus a weighted cost for the operations, multiplied by a congestion factor. Above 50% block utilization the factor rises, pricing demand out; below it the factor falls, encouraging use; the target is to hold utilization near 50%.

flowchart TB
    A["Generating<br/>NIGHT untouched<br/>fills to 5 DUST in a week"] --> B["Constant<br/>cap reached<br/>stays steady"]
    B --> C["Decaying<br/>NIGHT spent or moved<br/>decays at the same rate, still spendable"]
    C --> D["Zero<br/>permanently exhausted"]

Two details in this mechanism are the ones I’d write down first.

The first is the registration deadlock. Spending DUST on Midnight requires registering first — binding a NIGHT key to a DUST key — and registration itself costs DUST, which an unregistered wallet doesn’t have. The protocol’s answer is retroactive generation: unregistered NIGHT counts as having “virtually accumulated” DUST, and the registration transaction can pay its own fee out of that virtual balance. The catch is equally concrete: freshly bought or freshly moved NIGHT has almost no accumulation, so registration can fail outright — wait a week, or let an already-registered wallet sponsor the fee. The side effect: newly moved NIGHT always waits a week before it works, which objectively blocks a batch of mass-registered spam accounts.

The second is the 3-hour grace window. DUST balances are shielded, so the network cannot see what a wallet holds; value is therefore computed from transaction creation time rather than inclusion time, with 3 hours of slack to tolerate network latency. I sat with that rule for a while: the cost of privacy reaches even fee accounting — the ledger cannot see its own balances, so the protocol uses a time window instead.

One counterintuitive fact deserves its own paragraph: burned DUST is collected by no one. Block producers take no fees and there is no tip mechanism — the resource is non-transferable, so there is nothing to collect. TRON earns protocol revenue directly by burning TRX; Midnight removed that revenue source entirely and left all the value on the generation rights of the NIGHT asset.

TRON’s Energy: the same mechanism, running for seven years

Now TRON. Stake TRX and receive Energy and Bandwidth proportional to the share of network stake, rolling back over 24 hours; with no resources, TRX is burned automatically to cover the cost. The mechanism is isomorphic to DUST’s: asset leads to time leads to a renewable resource, both replacing the older “burn coins per transaction” model, both adding sponsorship so users never pay gas directly — TRON calls it deployer energy sharing, Midnight calls it DUST sponsorship. Midnight’s own figure: DApps lose on average 80% of users at registration because of the gas barrier.

But what stands out about TRON is not the mechanism — it’s the market. In the first half of 2026, network resource revenue reached 1.31 billion US dollars, of which about 1.13 billion was Energy-related. None of that came from users burning directly; it was earned by a rental market of more than twenty platforms, where renting 65,000 Energy costs 0.8 to 5 TRX, saving 70 to 90 percent versus burning. What underpins that market is the single most consequential design in TRON’s model — Energy is delegable to any account.

That delegability is what created an industry Midnight doesn’t have. Sending one USDT transfer per day on self-staked TRX would lock up about 7,000 TRX, which small holders simply can’t justify; so wallet vendors and rental platforms absorbed all the complexity, and users pay a few cents per transfer. Even the fee itself is negotiated by that market: in August 2025 a single governance proposal cut the Energy price from 210 sun to 100, nearly halving costs — enterprises budgeting on TRON depend on the direction of the next proposal.

Midnight made a radical decision at exactly this point: DUST is non-transferable and cannot be purchased. No rental market, no gas speculation, and no paying to jump the queue under congestion. Cardanians put it precisely: every chain allocates block space by price; Midnight allocates it by “capital held × time.” The cost is that the rental business TRON ran for seven years — and that earned 1.13 billion dollars — was deliberately removed by Midnight.

Three key differences

Placing the two mechanisms side by side, I count three key differences.

The first is transferability — and here I agree with Midnight. Non-transferable DUST means the resource is not a financial asset, which eliminates speculation and regulatory risk in one move; it is the precondition for Midnight’s compliance argument to institutions. Open the transfers and the precondition fails, so I don’t think opening is the answer.

The second is who gets paid. TRON charges for usage — burned TRX flows into the protocol ledger, a genuine revenue source. Midnight charges indirectly through holding: zero fee revenue, network security funded by reserve-issued NIGHT rewards with an initial subsidy rate of 95%, and all value resting on NIGHT. At the start of September NIGHT’s market cap sat around 300 million dollars, still under the selling pressure of a 450-day quarterly linear unlock — with an asset that small, whether it can support settlement at TRON’s scale is my biggest open question.

The third is who goes first under congestion. TRON’s rule is highest bidder wins; Midnight’s rule is “whoever holds more, whoever can wait goes first” — when demand spikes there is no buying a way forward, only queueing for generation. And congestion rate is the one core parameter never validated under real load: five months of mainnet has produced only about 138,000 transactions, so the first real congestion will likely arrive on the very day demand spikes.

Will it become the next TRON

So, the unavoidable question: will Midnight become the next TRON? My answer is no — a gap of several orders of magnitude in volume, no stablecoin-driven settlement volume, capacity constrained by NIGHT holdings. But its benchmark shouldn’t be TRON anyway: TRON built the lowest-cost dollar settlement rail; Midnight is betting that “institutions arrive when privacy, compliance, and predictable costs arrive.” MoneyGram is both a long-standing TRON USDT channel and a founding Midnight node operator; Monument Bank plans to tokenize up to 250 million pounds of retail deposits on-chain — the kind of public institutional partnership TRON never landed.

What I find genuinely worth observing is whether the deleted rental market reappears in another form. The ledger spec states explicitly that the NIGHT-to-DUST address mapping supports redesignation at any time, precisely to reserve the future use-case of “leasing out Dust generation.” In April, a full DUST Capacity Lease design already appeared on the community forum — lock NIGHT, designate the generation to a lessee, irrevocable until expiry. The key of that design is the collateral: touch the NIGHT and the DUST decays weekly; default needs no arbitration, the protocol enforces it automatically. That property is unique to the DUST model — TRON’s delegation can’t do it, since delegated Energy can be withdrawn at any time and rental platforms depend on off-chain trust and prepayment.

So my read: what Midnight needs is not a copy of TRON’s rental market, but its compliant version — rent out the generation flow, never transfer the stock. The same market structure, with TRON operating outside regulatory oversight, while Midnight could run the same structure inside a compliance framework, because what gets rented is “the right to use a service,” auditable by design.

Four things I’ll watch over the next six months

• Whether Mōhalu, the incentivized phase, gets a verifiable hard date — mainnet launched at the end of March, the originally Q2-targeted switch still hadn’t happened by early September, staking rewards unlaunched, zero SPOs connected;
• Whether congestion rate gets a public stress test under real load;
• Whether Capacity Lease graduates from a forum post to an audited contract;
• The first verifiable institutional flow — Monument Bank’s deposit tokenization or a MoneyGram remittance leg showing up on-chain.

Seven years of TRON data point to one conclusion: a resource rental market drives usage growth the fastest. Midnight removed that market, and with it the revenue and the platforms. Once the incentivized phase switches, I’ll be watching whether the market reappears in compliant form — that will be the key evidence in my judgment of whether Midnight makes it.

Article Link: