Core vs. Knots: The Great Bitcoin Debate Over 83 Bytes and Its Implications for the Community

In the spring of 2025, one of the largest conflicts in the Bitcoin community erupted in recent years. A technical proposal to remove the limit on data in OP_RETURN escalated into a significant debate among developers and users.

Proponents of abolishing the 83-byte limit emerged from Bitcoin Core, while opponents coalesced around the alternative client Bitcoin Knots. Together with the Bitcoin mixer Mixer.Money, we delve into the technical aspects of the dispute and its potential implications for the future of the first cryptocurrency.

OP_RETURN is an opcode (or command) in Bitcoin’s stack-based Script language that renders the output of a transaction unspendable, allowing the recording of up to 83 bytes of arbitrary data. This is the standard method for adding metadata to the blockchain without expanding the size of the UTXO set — the database of all unspent transaction outputs.

Introduced in 2014 with Bitcoin Core 0.9.0, OP_RETURN was never intended to encourage data storage in the blockchain but served as a compromise to prevent more harmful practices.

«This change does not endorse data storage in the blockchain. OP_RETURN creates a provably unspendable output to avoid data storage schemes where arbitrary information like images would be saved as forever unspent transaction outputs, bloating the UTXO database. Storing arbitrary data in the blockchain is still a poor idea; it’s far cheaper and more efficient to store non-financial data elsewhere,» stated the Bitcoin Core documentation.

According to Bitcoin Core developer Gregory Maxwell, prior to the introduction of OP_RETURN, users found ways to embed data by sending bitcoins to «fake» addresses, which were supposedly unspendable. They created outputs that appeared spendable but were not, leading to a continuous bloating of the UTXO set.

«While OP_RETURN outputs do add to the overall blockchain size, they are not included in the UTXO set, meaning they don’t need to be constantly tracked. Nodes can ignore and delete these data post initial verification,» comments Mixer.Money.

Initially, the limit was set at 40 bytes but was increased to 80 bytes in 2015 and then to 83 bytes in 2016. These limits were deliberately chosen to encourage the use of hashes instead of raw data and to emphasize that Bitcoin’s blockchain should primarily be utilized for financial transactions.

It’s important to note that the OP_RETURN limit is a standardness rule (or mempool policy) rather than a consensus rule. This means Bitcoin Core nodes by default do not relay transactions with OP_RETURN data exceeding 83 bytes, although the Bitcoin protocol does not prohibit them.

Miners can include «nonstandard» transactions in blocks, especially if the accompanying fees make them profitable. If such a transaction is included in a valid block, all Bitcoin nodes will accept it, regardless of their policies regarding OP_RETURN.

On April 27, 2025, Peter Todd submitted Pull Request (PR) #32359 to the Bitcoin Core repository. The proposal aimed for a complete abolition of the OP_RETURN output size limit and was a formalization of the initiative by Antoine Poinsot of Chaincode Labs.

The idea arose after Poinsot learned that the ZK-rollup on Bitcoin, Citrea, decided to publish data on the blockchain using one OP_RETURN and two UTXOs due to the existing OP_RETURN standard. These two unspent outputs would accumulate, while one of the goals of Bitcoin Core’s development is to restrict the growth of the UTXO set.

«Theoretically, removing the OP_RETURN limit would allow projects like Citrea to implement technical solutions without increasing the UTXO set. This could also influence miners who accept transactions through private mempools rather than organically through the node network,» comments Mixer.Money.

Todd’s proposal did not require a soft fork or hard fork; it merely expanded the types of transactions that would propagate across the network.

Some commentators supported the removal of what they considered unnecessary code. Others, while acknowledging the inefficiency of the OP_RETURN limit, opposed its removal, arguing that lifting restrictions could lead to spam proliferation on the network.

Specifically, developer Luke Dash Jr. urged for the retention of the limits, suggesting users stick with older versions of Core or switch to Bitcoin Knots, which has a stricter OP_RETURN limit of 42 bytes.

On May 5, Bitcoin Core developer Greg Sanders announced plans to include this change in the next release, further intensifying the community debate. Many users began publicly stating their intention to switch to Bitcoin Knots, and a number of bloggers (such as BTC Sessions and Southern Bitcoiner) provided instructions on YouTube.

By May 2025, the share of Bitcoin Knots nodes approached 10%. In comparison, in January 2023, its share was only 0.3%.

Supporters of the change present several arguments:

One argument has led opponents to hypothesize financial motives behind the actions of Bitcoin Core. Developer Jameson Lopp, who supported Todd’s proposal, was accused of a conflict of interest due to his investments in Citrea, which was the original impetus for the proposal. In response, Lopp stated that his investments (including Citrea) are well known and disclosed on his website.

Opposition members, also referred to as «filterors,» view Bitcoin primarily as a monetary network and do not wish to «turn it into a universal database.» Luke Dash Jr. labeled the proposal as «sheer insanity,» while Jason Hughes of Ocean Mining believes it could render Bitcoin a «useless altcoin.»

«Sidechain developers should not influence Bitcoin Core. At its core, Bitcoin is money, and it should focus solely on that,» asserted Willem Shroe, founder of Botanix Labs.

Critics worry that unlimited OP_RETURN could pave the way for a surge of non-financial data, potentially overloading the network. Although OP_RETURN data can be removed from the UTXO set, they still occupy space in blocks.

Moreover, opponents argue that current limits are effective as they serve as a deterrent. Some interpret the fact that «spammers» resort to more costly methods as evidence that the limits are effective, making undesirable behavior more expensive.

Additionally, critics point to the active migration to Bitcoin Knots as a form of «node voting» — a decentralized protest where node operators express discontent with the direction of Bitcoin Core’s development by opting for alternative software.

Finally, «filterors» like Bitcoin Mechanic accuse Bitcoin Core of promoting changes without achieving widespread consensus, as well as poor communication regarding the decision-making process.

On May 12, 2025, PR #32359 was closed despite assertions that it would be included in the next version (30.0), slated for release in early October.

The conflict has highlighted key governance issues within the Bitcoin ecosystem: who makes the decisions about its development — developers, miners, or the broader user community?

Many observers compare the situation to the block size wars of 2015–2017, but a crucial distinction exists: changing the OP_RETURN limit is a client policy alteration, not a consensus rules change. Therefore, a hard fork is virtually excluded in this case.

The rising popularity of Bitcoin Knots shows that users are willing to «vote with their nodes,» choosing alternative implementations of the protocol when they disagree with the developmental direction dictated by Bitcoin Core.

«The debate over OP_RETURN transcends a purely technical issue. It represents a fundamental clash of visions about what Bitcoin should be: a pragmatically evolving system or a strictly defined monetary network,» comment representatives from Mixer.Money.

Experts further note that the dispute does not encompass privacy issues, and ensuring anonymity within the Bitcoin network will remain a pressing concern for all community members regardless of their stance in this conflict.

The Bitcoin mixer Mixer.Money has been active in the market since 2016. The service mixes coins in three modes: «Mixer,» «Exact Payment,» and «Complete Anonymity.»

In the «Complete Anonymity» and «Exact Payment» modes, Mixer.Money sends users bitcoins that have been obtained directly from major exchanges and retain their corresponding status.