In 2015, LOUIS XIII sealed a John Malkovich film until 2115. Bitcoin has its own long clock, written in block heights rather than years. Its last new satoshi is expected around 2140. That is not the end of mining. It is the end of the subsidy.
On November 18, 2015, a film had its grand non-premiere.
It had a famous actor, a famous director, a real crew and a finished story. John Malkovich wrote it. Robert Rodriguez directed it. Then LOUIS XIII placed the only film reel inside a safe designed to open automatically one hundred years later.
The premiere is scheduled for November 18, 2115, in Cognac.
The people who made the film were never its intended first-night audience. The invitations were made for descendants.
The project is called 100 Years: The Movie You Will Never See. Its title is less a description than a dare.
LOUIS XIII could have bought another page of advertising and written that the house thinks in generations. Instead, it commissioned an expensive piece of culture, completed it, and removed the payoff from the lifetime of almost everyone involved. The company's own annual report described the century as the brand's natural unit of time.
It does not explain patience. It performs patience.
Bitcoin made a different kind of century-scale bet.
There is no film reel. No cellar master. No single safe. Yet the network carries a promise that also reaches beyond the lives of its creators and its current audience: new bitcoin is issued on a declining schedule, block after block, until the subsidy reaches zero at a height expected around the year 2140.
People often compress this into one sentence: "The last bitcoin will be mined in 2140."
Memorable, yes. Technically complete, no.
There is no final whole bitcoin waiting behind a vault door. There is a last possible unit of new issuance, one satoshi, released through a block subsidy. And when that subsidy reaches zero, mining does not automatically stop.
The real story is better than the slogan.
The 140-year correction
First, the number.
Bitcoin does not have 140 years left from today. The network began in 2009. Under the nominal ten-minute block target, the zero-subsidy height lands roughly 131.8 years after the genesis block, around 2140. From 2026, the remaining distance is closer to 114 years.
So where does "140 years" belong? At best, it is loose conversational rounding for a century-plus issuance story. It is not a field in the protocol, and it should not be presented as the remaining countdown.
This correction makes the analogy sharper.
The safe is scheduled to open on November 18, 2115.
The subsidy becomes zero at height 6,930,000. The year is an estimate.
On Bitcoin mainnet, the subsidy interval is 210,000 blocks. The target spacing is ten minutes. Multiply the two and one subsidy era works out to just under four nominal years. But blocks arrive probabilistically, and the network periodically adjusts mining difficulty to steer the average back toward the target. That is why 2140 is an estimate, not an appointment engraved on a physical lock.
The safe watches a clock.
Bitcoin watches height.
New bitcoin does not wait in a vault
The phrase "mining the last bitcoin" creates the wrong mental picture. It sounds as if 21 million finished coins were placed underground in 2009 and miners have been digging them out ever since.
That is not how issuance works.
The first transaction in a valid block is a special coinbase transaction. It can pay the miner a block reward made of two parts:
The subsidy is newly issued value permitted by the consensus rules. The fees come from transactions included in the block. A miner proposes the coinbase transaction, but the miner does not get to invent the amount. Bitcoin Core checks the allowed subsidy for that block height, adds the available fees, and rejects a block whose coinbase pays too much.
This is the important cultural difference between a promise and a protocol.
A company can say it will keep a safe closed. Bitcoin participants do not have to trust a future company, foundation or mining executive to remember the supply schedule. Independent validating nodes can check the same rule for every block.
The schedule survives because the limit is verified in the present tense.
The staircase to one satoshi
Bitcoin's initial subsidy was 50 BTC per block. Every 210,000 blocks, that number is divided by two.
The sequence is familiar:
| Era begins at height | Subsidy per block | Calendar context |
|---|---|---|
| 0 | 50 BTC | Genesis era, 2009 |
| 210,000 | 25 BTC | First halving era |
| 420,000 | 12.5 BTC | Second halving era |
| 630,000 | 6.25 BTC | Third halving era |
| 840,000 | 3.125 BTC | Current era in 2026 |
| 6,720,000 | 1 satoshi | Final non-zero subsidy era |
| 6,930,000 | 0 | Zero-subsidy era, expected around 2140 |
In the code, the operation is brutally compact. Start with 50 BTC in satoshis. Divide it by two once for each completed interval. Because Bitcoin's smallest on-chain unit is one satoshi, fractions smaller than that are discarded. After the one-satoshi era, another division produces zero.
This also explains why "21 million" is the elegant headline, not a pile counted down to the last decimal. Applying the current Bitcoin Core schedule across all subsidy eras produces a theoretical maximum of 20,999,999.9769 BTC from block subsidies. Actual issued supply can be lower if a miner claims less than the permitted reward. Lost coins affect spendable supply, but they do not change the issuance rule.
The number is famous because it is simple.
The mechanism is durable because it is exact.
What actually happens in 2140?
One part of the reward falls to zero. The other part does not have to.
Transaction fees already exist today. They are paid by users who want transactions included in blocks. The Bitcoin white paper explicitly described a future in which the incentive could transition entirely to transaction fees after the predetermined issuance had entered circulation.
So the last new satoshi would mark the end of the block subsidy, not the automatic end of:
- proof of work
- block production
- transaction fees
- mining equipment
- mining competition
- network validation
Whether fee revenue will be sufficient to support the level and distribution of mining security people want in 2140 is a serious economic question. Nobody can prove the answer from 2026. Transaction demand, fee markets, energy systems, hardware, regulation and Bitcoin itself can change over more than a century.
The honest statement is narrow: the current protocol permits fees to remain after the subsidy reaches zero. It does not guarantee the future price of those fees, the hashrate they will support, or the economics of any future miner.
Anyone selling certainty about 2140 is trying to turn a consensus rule into a forecast.
Why this matters to a miner who will not be there
No ASIC running today will see 2140. Most will not see many subsidy eras at all.
That does not make the long horizon irrelevant.
Every mining model already lives inside the same transition. The subsidy changes in discrete steps. Fees vary block by block. Difficulty changes. Bitcoin price changes. Energy price, uptime, fleet efficiency and accepted work determine whether a specific operation can live through those changes.
The useful lesson is not to build a 114-year spreadsheet. It is to keep four variables separate now:
- Subsidy: Newly issued bitcoin allowed at the current block height.
- Fees: Transaction value offered for block inclusion and collected by the successful miner.
- Hashrate: The work competing for that reward.
- Site economics: The actual cost and reliability of turning electricity into accepted work.
Those variables interact, but they are not interchangeable. A halving changes the subsidy. It does not automatically halve electricity prices, difficulty or every operator's revenue. A fee spike can make one block unusually valuable. It does not prove a permanent fee market. A more efficient machine changes one site's cost curve. It does not rewrite the issuance schedule.
ROI ASIC's guide to flexible load and curtailment makes the same discipline practical: mining economics belong to a measured site, a real operating window and explicit assumptions.
Two machines for carrying time
The LOUIS XIII film and Bitcoin are not the same kind of object, and pretending they are would weaken both stories.
The film is a work of branded culture. Its scarcity comes from controlled access. One artifact is kept away from the audience until one date.
Bitcoin is an open protocol. Its monetary scarcity comes from verification. The rules are visible, the software is inspectable, and each accepted block has to remain inside the limit recognized by the network.
One keeps the object secret.
The other keeps the rule public.
One asks a safe to stay closed.
The other asks independent nodes to keep rejecting the wrong amount.
Yet both projects understand something most marketing forgets: a claim becomes powerful when the action itself carries the meaning.
LOUIS XIII did not merely say, "We think a century ahead." It created an audience in the future.
Bitcoin does not merely say, "Supply is limited." Its participants repeatedly verify the issuance limit in software.
The strongest promise is the behavior that remains when the people who wrote the sentence are gone.
The 2140 Card
Use these four lines the next time someone says Bitcoin mining ends in 2140.
| Claim | More accurate version |
|---|---|
| "Bitcoin has 140 years left" | The nominal schedule spans about 131.8 years from 2009. Roughly 114 years remain from 2026. |
| "The last bitcoin is waiting" | New value is issued through each valid block's coinbase transaction. The final non-zero subsidy is one satoshi per block. |
| "Mining ends in 2140" | The subsidy reaches zero. Fees can remain part of the block reward. Future mining economics are not guaranteed. |
| "A miner could issue more" | A block paying more than the allowed subsidy plus fees is invalid under current consensus validation. |
The date is approximate. The height is the rule. The subsidy ends. The network is not programmed to stop.
That is the version worth passing on.
What will outlive the pitch
In 2115, if the mechanism works and the chain of custody survives, a safe in Cognac will open. An audience that did not exist when the film was made will finally see what John Malkovich and Robert Rodriguez put inside.
Around 2140, if Bitcoin continues under its current issuance rules, miners will reach a block height at which the code permits no new subsidy. The audience will have changed. The hardware will be unrecognizable. The economics may look nothing like today's.
But the underlying gesture is already visible.
LOUIS XIII trusted a safe to carry a story through time.
Bitcoin trusted a rule that strangers could keep checking.
Great marketing proves what a brand believes.
Great protocols make belief unnecessary.
Sources and data freshness
- LOUIS XIII, Cognac and Time, checked September 8, 2026. Official confirmation of the film title, 2115 release and century theme.
- Rémy Cointreau, 2016 shareholder letter and 2015/2016 annual report. Official historical records for John Malkovich as writer, Robert Rodriguez as director, the single reel, automatic safe and November 18, 2115 premiere.
- Bitcoin Core, mainnet consensus parameters, checked September 8, 2026. Used for the 210,000-block interval and 600-second target spacing.
- Bitcoin Core, block subsidy and coinbase validation, checked September 8, 2026. Used for the halving formula and rejection of excessive coinbase value.
- Bitcoin Developer Reference, Block Chain, checked September 8, 2026. Used for the 6,930,000 height boundary and reward definitions. Its historically stale current-subsidy example is not reused.
- Satoshi Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, section 6. Used for the proposed transition from subsidy to transaction-fee incentives, not as proof of future fee sufficiency.
Risk and ecosystem disclosure. The 2140 date is an estimate based on the current block-height schedule and nominal target spacing. This article does not predict that Bitcoin, its fee market, its security budget or any mining business will operate in a particular form through 2140. Mining revenue, difficulty, fees, energy costs, hardware performance and regulation can change. Historical protocol rules are not financial forecasts. ROI ASIC is part of the VNISH ecosystem. No firmware compatibility, hashrate, efficiency, uptime, revenue, savings or investment result is claimed or guaranteed. This article is educational information, not financial advice.