{{Short description|None}} {{pp-semi|small=yes}} {{main|Fork (blockchain)}} {{More citations needed|date=September 2021}} Bitcoin forks are defined variantly as changes in the protocol of the Bitcoin network or as the situations that occur "when two or more blocks have the same block height".<ref>{{cite book|last1=Antonopoulos|first1=Andreas|title=Mastering Bitcoin: Programming the Open Blockchain|date=2017|publisher=O' Reilly media, inc.|location=USA|isbn=978-1491954386|page=Glossary|edition=2}}</ref> A fork influences the validity of the rules. Forks are typically conducted in order to add new features to a blockchain, to reverse the effects of hacking or catastrophic bugs. Forks require consensus to be resolved or else a permanent split emerges.
According to {{abbrv|BIP|Bitcoin Improvement Proposal}} 123 specifications, there are 3 main categories of Bitcoin forks:<ref name="BIP123">{{cite web |last1=Lombrozo |first1=Eric |id=BIP 123 |title=BIP Classification |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0123.mediawiki |date=2016-11-30 |lang=en |via=Github |access-date=2026-05-20}}</ref>
* Soft Forks within Bitcoin: These don't split the network. These are consensus rule changes where some previously-valid blocks are now invalid. * Hard Forks within Bitcoin: These cause a permanent chain split between the 2 versions. These are consensus rule changes where some previously-invalid blocks are now valid. * Hard Forks that created alternative blockchains and cryptocurrencies.
There is also another subcategory of consensus rule changes that had a chance to cause a hard fork and chain split. However, because they were never exploited, they technically never resulted in a hard fork.
==Soft Forks within Bitcoin== These are consensus rule changes that did not cause a split in the network. These are the result of consensus rule changes where previously-valid blocks are now invalid in newer clients.<ref name="BIP123" /> Client nodes both with and without the update can still build on the same chain. Two of the most well-known Bitcoin soft forks are Segwit and Taproot. ===Segwit=== {{Excerpt|Segwit|paragraphs=1,2|only=paragraphs}} ===Taproot=== Taproot is an agreed soft fork in the transaction format. The fork adds support for Schnorr signatures, and improves functionality of smart contracts and the Lightning Network. The fork was installed in November 2021.<ref>{{Cite news|last=Sigalos|first=MacKenzie|date=2021-06-09|title=Bitcoin just got its first makeover in four years|language=en-US|work=CNBC|url=https://www.cnbc.com/2021/06/12/bitcoin-taproot-upgrade-what-it-means.html|access-date=2021-06-15|issn=}}</ref> The upgrade adds privacy features.<ref>{{Cite news|last= Kharif|first=Olga|date=2021-06-15|title=Bitcoin to get more privacy features in Taproot update, making it harder to trace payments|language=en-US|work=Bloomberg|url=https://www.bloomberg.com/news/articles/2021-06-15/bitcoin-network-approves-privacy-update-as-scrutiny-increases|access-date=2021-06-23|issn=}}</ref><ref>{{Cite news|last= Locke|first=Taylor|date=2021-06-14|title=7 key things that happened in crypto over the past week|language=en-US|work=CNBC|url=https://www.cnbc.com/2021/06/14/7-key-things-that-happened-in-crypto-over-the-past-week.html|access-date=2021-06-23|issn=}}</ref><ref name="btcoptech-taproot">{{cite web |url=https://bitcoinops.org/en/preparing-for-taproot/#is-taproot-even-worth-it-for-single-sig |title=Is taproot even worth it for single-sig? |lang=en |date=2021-06-23 |access-date=2026-05-20 |website=Bitcoin Optech |archive-url=https://web.archive.org/web/20260123111127/https://bitcoinops.org/en/preparing-for-taproot/#is-taproot-even-worth-it-for-single-sig |archive-date=2026-01-23 |url-status=live}}</ref><ref name="btccom-taproot">{{cite web |url=https://www.bitcoin.com/get-started/bitcoin/technical/what-is-bitcoin-taproot/ |title=Bitcoin Taproot Explained: How the Upgrade Works |orig-date=2026-03-06 |date=2026-04-28 |lang=en |first=Neill |last=Velardo |website=Bitcoin.com |access-date=2026-05-20 |archive-url=https://web.archive.org/web/20260520113350/https://www.bitcoin.com/get-started/bitcoin/technical/what-is-bitcoin-taproot/ |archive-date=2026-05-20 |url-status=live}}</ref> Taproot includes BIP numbers 340,<ref name="bip340">{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0340.mediawiki |id=BIP 340 |title=Schnorr Signatures for secp256k1 |date=2023-02-19 |first1=Pieter |last1=Wuille |first2=Jonas |last2=Nick |first3=Tim |last3=Ruffing |lang=en |via=GitHub}}</ref> 341,<ref name="bip341">{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0340.mediawiki |id=BIP 340 |title=Taproot: SegWit version 1 spending rules |date=2023-02-19 |first1=Pieter |last1=Wuille |first2=Jonas |last2=Nick |first3=Anthony |last3=Towns |lang=en |via=GitHub}}</ref> 342.<ref name="bip342">{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0342.mediawiki |id=BIP 342 |title=Validation of Taproot Scripts |date=2023-02-19 |first1=Pieter |last1=Wuille |first2=Jonas |last2=Nick |first3=Anthony |last3=Towns |lang=en |via=GitHub}}</ref>
Advantages:<ref name="btcoptech-taproot"/><ref name="btccom-taproot"/>
* Complex transactions, such as those requiring multiple signatures or those with delayed release, are indistinguishable from simple transactions in terms of on-chain data. * Reduced transaction costs: The data size of complex Bitcoin transactions is reduced, which leads to lower transaction fees. * Support for more complicated conditions for a transaction via Schnorr signatures. * Benefits for the Lightning Network: More flexibility, privacy enhancement, lower costs.
==Hard Forks within Bitcoin== These are Hard Forks within Bitcoin that caused a chain split but did not create an alternative network or cryptocurrency. These are consensus rule changes where some previously-invalid blocks are now valid in newer clients.<ref name="BIP123" />
* v0.3.6: The addition of OP_NOP functions eventually caused a chain split. Clients prior to 0.3.6 can no longer sync with the current Bitcoin blockchain after block 163,685.{{Citation needed|date=December 2025}} * March 2013 Chain Fork. The migration from BerkeleyDB to LevelDB in v0.8 caused a chain split. Miners reorged 24-blocks by switching back to v0.7.2. At least one double-spend occurred during this event. This bug was later fixed in v0.8.1.<ref>{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0050.mediawiki |first=Gavin |last=Andresen |id=BIP 50 |title=March 2013 Chain Fork Post-Mortem |lang=en |via=GitHub |date=2016-02-01 |access-date=2026-05-20}}</ref>
==Hard Forks that created alternative blockchains and cryptocurrencies== These are Hard Forks created via consensus rule changes that created an alternative blockchain and cryptocurrency. They share a transaction history with Bitcoin up to a certain time and date. The first hard fork splitting bitcoin happened on 1 August 2017, resulting in the creation of Bitcoin Cash.
The following is a list of notable hard forks splitting bitcoin by date and/or block:
* Bitcoin Cash: Forked at block 478558, 1 August 2017, for each bitcoin (BTC), an owner got 1 Bitcoin Cash (BCH) * Bitcoin Gold: Forked at block 491407, 24 October 2017, for each bitcoin (BTC), an owner got 1 Bitcoin Gold (BTG) * Bitcoin Satoshi Vision: Forked at block 556766, 15 November 2018, for each Bitcoin Cash (BCH), an owner got 1 Bitcoin SV (BSV). * eCash: Forked at block 661648, 15 November 2020, for each Bitcoin Cash (BCH), an owner got 1,000,000 eCash (XEC).
; Bitcoin Classic: In its first 8 months, Bitcoin Classic promoted a single increase of the maximum block size from one megabyte to two megabytes.<ref name="paulvigna">{{cite web|url=https://www.wsj.com/articles/is-bitcoin-breaking-up-1453044493|title=Is Bitcoin Breaking Up?|author=Paul Vigna|date=17 January 2016|work=Wall Street Journal}}</ref><ref name=":0" /> In November 2016 this changed and the project moved to a solution that moved the limit out of the software rules into the hands of the miners and nodes.<ref>{{cite web |url=https://zander.github.io/posts/Classic%20is%20Back/ |title=Classic is Back |access-date=April 28, 2017 |archive-url=https://web.archive.org/web/20170202055402/https://zander.github.io/posts/Classic%20is%20Back/ |archive-date=2017-02-02 |url-status=dead}}</ref> ; Bitcoin Unlimited All three software clients attempt to increase transaction capacity of the network. None achieved a majority of the hash power.<ref name="tbs">{{cite book |title=The Bitcoin Standard: The Decentralized Alternative to Central Banking |last=Ammous |first=Saifedean |year=2018 |publisher=John Wiley & Sons |isbn=9781119473893 |pages=227, 228 |url=https://play.google.com/store/books/details?id=Pw5TDwAAQBAJ |access-date=23 April 2018}}</ref><ref name=":0" />
==Software forks of Bitcoin Core without a blockchain== The following are proposed software forks of the Bitcoin Core software client that did not successfully launch a blockchain: ; Bitcoin XT : A fork initiated by Mike Hearn. The current reference implementation for bitcoin contains a computational bottleneck.<ref name="inth">{{Cite news |url=http://www.newyorker.com/business/currency/inside-the-fight-over-bitcoins-future |title=Inside the Fight Over Bitcoin's Future |author=Maria Bustillos |access-date=7 January 2017 |date=25 August 2015 |newspaper=The New Yorker |publisher=Conde Naste }}</ref> The actual fork was preceded by Mike Hearn publishing BIP 64 on June 10, 2014, calling for the addition of "a small P2P protocol extension that performs UTXO lookups given a set of outpoints."<ref name="bip64">{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0064.mediawiki |id=BIP 64 |title=getutxo message |first=Mike |last=Hearn |lang=en |date=2014-06-10 |via=GitHub}}</ref> On December 27, 2014 Hearn released version 0.10 of the forked client XT, with the BIP 64 changes.<ref group=src>{{cite web|url=https://github.com/bitcoinxt/bitcoinxt/releases/tag/v0.10|title=bitcoinxt/bitcoinxt|website=GitHub}}</ref> It achieved significant attention within the bitcoin community in mid-2015 amid a contentious debate among core developers over increasing the block size cap.<ref name="Hern-2015">{{cite news|url=https://www.theguardian.com/technology/2015/aug/17/bitcoin-xt-alternative-cryptocurrency-chief-scientist|title=Bitcoin's forked: chief scientist launches alternative proposal for the currency|author=Alex Hern|date=17 August 2015|newspaper=the Guardian|access-date=20 August 2015}}</ref><ref name=":0">{{Cite web |date=2022-04-29 |title=A list of Bitcoin forks and how they have changed the network - CNBC TV18 |url=https://www.cnbctv18.com/cryptocurrency/a-list-of-bitcoin-forks-and-how-they-have-changed-the-network-13318902.htm |access-date=2024-09-09 |website=CNBCTV18 |language=en}}</ref>
:On June 22, 2015, Gavin Andresen published BIP 101 calling for an increase in the maximum block size. The changes would activate a fork allowing eight MB blocks (doubling in size every two years) once 75% of a stretch of 1,000 mined blocks is achieved after the beginning of 2016.<ref name="bip101">{{cite web |url=https://github.com/bitcoin/bips/blob/16e81515be7814b91dc304f87c1a4ee40a3798f8/bip-0101.mediawiki |id=BIP 101 |title=Increase maximum block size |first=Gavin |last=Andresen |lang=en |date=2015-06-22 |via=GitHub}}</ref> The new maximum transaction rate under XT would have been 24 transactions per second.<ref name="edov">{{Cite news |url=http://www.pcworld.com/article/2974339/bitcoin-xt-debate-overshadowing-growth-opportunities.html |title=Bitcoin XT debate overshadowing growth opportunities |author=Tim Hornyak |access-date=7 January 2017 |date=21 August 2015 |work=PC World |publisher=IDG }}</ref>
:On August 6, 2015 Andresen's BIP 101 proposal was merged into the XT codebase.<ref group=src>{{cite web|url=https://github.com/bitcoinxt/bitcoinxt/commit/946e3ba8c7806a66c2b834d3817ff0c986c0811b|title=Implement hard fork to allow bigger blocks · bitcoinxt/bitcoinxt@946e3ba|website=GitHub}}</ref><ref group=src>{{cite web|url=https://github.com/bitcoinxt/bitcoinxt/releases/tag/v0.11A|title=bitcoinxt/bitcoinxt|website=GitHub}}</ref> BIP 101 was reverted<ref group=src>{{cite web|url=https://github.com/bitcoinxt/bitcoinxt/pull/117|title=2MB block size bump by dgenr8 · Pull Request #117 · bitcoinxt/bitcoinxt|website=GitHub}}</ref> and the 2-MB block size bump of Bitcoin Classic was applied instead.{{cn|date=February 2021}}
:The August 2015 release of XT received widespread media coverage. ''The Guardian'' wrote that "bitcoin is facing civil war".<ref name="Hern-2015" />
:''Wired'' wrote that "Bitcoin XT exposes the extremely social—extremely democratic—underpinnings of the open source idea, an approach that makes open source so much more powerful than technology controlled by any one person or organization."<ref>{{cite magazine|url=https://www.wired.com/2015/08/bitcoin-schism-shows-genius-open-source/|title=The Bitcoin Schism Shows the Genius of Open Source|date=19 August 2015|magazine=WIRED|author=Cade Metz}}</ref> Developer Adam Back was critical of the 75% activation threshold being too low and that some of the changes were insecure.<ref name="bisp">{{Cite news |url=https://www.cnbc.com/2015/08/20/bitcoin-splits-will-it-break-or-be-better-than-ever.html |title=Bitcoin splits: Will it break, or be better than ever? |author=Everett Rosenfeld |access-date=5 January 2017 |date=20 August 2015 |publisher=CNBC}}</ref>
:On August 25, 2017, Bitcoin XT published ''Release G'', which was a Bitcoin Cash client by default.<ref name="releases" group=src>{{cite web|url=https://github.com/bitcoinxt/bitcoinxt/releases|title=Bitcoin XT Releases|website=GitHub |access-date=17 June 2018}}</ref> Subsequently, ''Release H'' was published, which supported the November 2017 Bitcoin Cash protocol upgrade, followed by ''Release I'', which supported the May 2018 Bitcoin Cash protocol upgrade.{{cn|date=February 2021}}
==Potential hard forks that were never exploited==
These are client consensus updates that could have potentially caused a hard fork. They were either never exploited before being patched, or were reverted by a reorg before they were patched.
* CVE-2010-5139: There was a value overflow bug that was exploited when a transaction minted 184B bitcoins. Miners reorged the blockchain to remove the minting transaction, causing a chain split. Because the reorg occurred before wxBitcoin v0.3.11 was released to fix the bug, it technically was not a hard fork.<ref>{{cite web |title=NVD - CVE-2010-5139 |url=https://nvd.nist.gov/vuln/detail/CVE-2010-5139 |publisher=NIST National Vulnerability Database |access-date=2 December 2025}}</ref><ref>{{cite journal |last1=Quamara |first1=Sidharth |last2=Singh |first2=Awadhesh |title=A systematic survey on security concerns in cryptocurrencies: State-of-the-art and perspectives |journal= Computers & Security|date=22 February 2022 |volume=113 |issue=102548 |article-number=102548 |doi=10.1016/j.cose.2021.102548 |url=https://www.sciencedirect.com/science/article/abs/pii/S0167404821003722 |access-date=2 December 2025|url-access=subscription }}</ref> * CVE-2018-17144: Bitcoin 0.15 had a critical inflation bug that allowed double spending certain inputs in the same block. This was not exploited before being patched.<ref>{{cite web |title=NVD - cve-2018-17144 |url=https://nvd.nist.gov/vuln/detail/cve-2018-17144 |publisher=NIST National Vulnerability Database |access-date=2 December 2025}}</ref>
==References== ===Source code=== {{Reflist|group=src}}
===Other references=== {{Reflist|30em}}
{{Bitcoin}}
{{DEFAULTSORT:Bitcoin forks}} * Category:Computing-related lists Category:Clients (computing)