Conflux
Two execution spaces, hybrid consensus and an ongoing argument about the economics of useful adoption.
Conflux combines proof-of-work block production with proof-of-stake finality. Core Space and Ethereum-compatible eSpace maintain separate account state on shared underlying infrastructure. CFX supports fees, staking and governance, while network upgrades and supporting services introduce important differences between older explanations and the current system.
この読み物は現在、英語で提供されています。画面の操作部分には、選択した言語を使用しています。
英語の原文を読む →ブラウザーの読み上げ対応を確認しています…
One network has two deliberately different execution environments
Core Space is Conflux's original execution environment; eSpace was introduced to reduce the effort of bringing Ethereum applications and tools into the ecosystem. Their transactions, accounts and contracts are logically separate even though they share an underlying ledger. This is not the relationship between an Ethereum rollup and Ethereum settlement. An eSpace application inherits Conflux's network assumptions. The familiar Ethereum-style interface helps software integration, but it does not mean the application is executing on Ethereum or using ETH as the native currency.
Conflux's economic introduction separates founding builders, network maintainers and ecosystem contributors. Its incentives attempt to support both infrastructure operation and application development, with CFX as the common unit. That explains the project's official broad ambition better than reducing it to a geographic nickname. Fees, storage resources, block rewards and staking connect the token to several activities. The introduction's dated numerical examples are historical snapshots, however, and should not be republished as today's circulation, reward rate or expected income.
Mining history precedes the two-space architecture
F2pool's November 2020 operational update records the Tethys mainnet launch on October 29 and the pool's introduction of CFX mining. This contemporaneous operator account anchors the network's early proof-of-work period. It also helps distinguish a blockchain launch from later exchange listings and application announcements. The history matters because someone reading an older description may encounter Conflux before its proof-of-stake finality layer and eSpace existed.
Those descriptions are not necessarily false about their own period, but they are incomplete accounts of the present architecture.
The Hydra documentation records the 2022 introduction of proof-of-stake finality and eSpace, together with new internal contracts and RPC fields. Core Space retained its own interfaces while eSpace exposed Ethereum-style methods. The upgrade therefore combined consensus changes with a developer-access strategy. It was more than a new marketing name. Integrators gained another execution environment, while operators needed to understand new PoS configuration and finalization information rather than assuming that an ordinary software replacement preserved every previous operational detail.
Mining and finality answer different questions
Conflux's consensus documentation assigns block production and Tree-Graph ordering to proof-of-work miners, while stake-selected PoS participants sign pivot blocks to finalize them. Mining has not simply disappeared because staking exists. The design adds a finality decision to the work-based history rather than converting every role into one validator function. This division helps readers interpret explorer fields: a recently observed block and a PoS-finalized point are different observations. It also means evaluating participation requires looking at both kinds of operators.
The practical staking guide describes a staged lock and exit process, with solo participation and pool participation offering different operational arrangements. It distinguishes an offline validator's forced retirement from malicious behavior. Its older permanent-slashing description conflicts with the later protocol change, so it must be read alongside the current specification. The useful durable lesson is that staking has timing, infrastructure and custody choices. A displayed annual rate is not a fixed bank deposit rate, and a pool adds its own implementation and operator dependencies.
The penalty model changed, and then needed a correction
Final CIP-156 changes punishment for disputed double voting from permanent forfeiture to a six-month stake lock. Its authors, Peilun Li and Chenxing Li, explicitly discuss honest operators accidentally keeping two voting instances active. The intention is to preserve time for investigation while reducing the irreversible consequence of an operational mistake. This does not make misbehavior costless: liquidity and interest can be lost during the lock. Nor does it authorize treating every accusation of double voting as an already proven attack.
The 3.1 documentation includes CIP-173, which fixes dispute-evidence verification and extends the lock to a validator whose stake is already entirely unlocking. The same package contains canonical transaction-encoding and cross-space delegation fixes. The specification records repairs to verification and transaction handling. It does not report an exploited loss, and none is inferred here.
A mapped address is not merely a different spelling
Core Space uses Conflux's base32 address format, while eSpace follows an Ethereum-style hexadecimal account model. The mapped-address documentation warns that the special address used for returning funds across spaces is derived by hashing, not by simply displaying the Core address in hexadecimal. Confusing those operations can send funds somewhere unintended. Using the same private key in both ordinary spaces also does not merge their balances. The source is particularly valuable because it explains a concrete user hazard hidden by otherwise familiar wallet interfaces.
CrossSpaceCall is an internal contract accessed from Core Space. Its interface supports transferring CFX, creating or calling eSpace contracts and withdrawing from a mapped account. These are network-defined operations with particular calling rules. They should not be described as arbitrary communication among unrelated blockchains. The existence of this internal path also does not imply that any externally bridged token can be moved through it under the same assumptions. Contract developers still need to identify the asset and direction supported by each operation.
An application can pay without making execution free
Core Space's SponsorWhitelistControl tracks separate sponsors and balances for gas and storage collateral. Eligibility is controlled through a contract whitelist, with an explicit option to cover all accounts. Sponsors also set a per-transaction gas limit for subsidy purposes. This can let a newcomer call an application before acquiring native currency, but somebody still funds the operation. The mechanism therefore serves an onboarding goal without proving that every contract, every account or every future transaction will receive an unlimited subsidy.
The v2.3 specification explains storage points: part of sponsored collateral is burned and converted into nontransferable capacity that can cover storage requirements. Those points do not earn storage interest. This is more specific than saying that Conflux burns tokens whenever data is deleted. The distinction matters because an early discussion of an idea can differ from its final implementation. It also separates a resource-accounting mechanism from a forecast that aggregate token supply must fall or that any holder will earn a particular return.
A partial burn and a governance decision are not fixed scarcity
Conflux's base-fee design adapts the Ethereum idea to a Tree-Graph ledger and deferred execution. Part of the base fee is burned, part can go to miners, and the priority fee also rewards miners. The burn share is governed rather than universally fixed at the initial setting. Inclusion and execution use related but distinct checks because a transaction can appear in a non-pivot block. These details are why an Ethereum fee explanation should not be copied wholesale into a Conflux guide.
In March 2026, the team published the results of the twentieth parameter vote, including a planned reduction of the PoW reward from 0.8 to 0.4 CFX per block and a higher storage-point ratio. The notice gave an estimated April effective date tied to block height. It is evidence of a specific governance result, not a permanent September parameter snapshot. Rewards and burns remain adjustable, so an investor's scarcity thesis must be checked against enacted settings and actual activity rather than an old introductory percentage.
Voting power carries a lock commitment
The governance guide ties voting power to both locked CFX and the remaining commitment period. Its illustrated choices range from a quarter to a year, with longer commitments producing greater weight. Once committed, holders cannot simply shorten the existing lock to withdraw early. This makes governance participation an economic choice as well as an expression of opinion. It does not create one-person-one-vote representation. A forum discussion, a locked voting balance and an executed protocol change are separate stages of influence.
A 2022 discussion of emissions makes the competing interests explicit. GLM favored increasing miner rewards, obaska wanted a smaller increase than the proposed doubling, and jhassiel_alcarraz argued for controlling supply while improving utility and professional economic oversight. These are named positions in an original discussion, not a synthetic description of community consensus. The disagreement is informative because the same issuance can look like security funding to an operator and dilution to a holder.
Current status requires more than a scheduled date
The May 2026 maintenance announcement recommends v3.0.3-fix to address a possible node panic. It does not describe stolen assets or provide a detailed exploit narrative. Publishing that bounded description is more accurate than either omitting the security work or labeling it a hack. The announcement also distinguishes replacing a running node's executable from starting a new node. Maintenance of this kind concerns service stability even when it brings no new consumer-facing feature or token-economic announcement.
The August 2026 mainnet announcement sets the 3.1 upgrade boundary at epoch 155,140,000 and a separate PoS threshold for CIP-173. A read-only query of the published mainnet RPC returned a finalized epoch beyond that boundary and a committed PoS height beyond the second threshold. The boundary epoch itself is timestamped August 25, 2026. Together, the specification and observed chain records support treating those scheduled boundaries as passed. This does not independently test every instruction or certify every hosted node's software.
Applications depend on services that can be replaced
The Foundation's November 2023 announcement said ShuttleFlow's operation would close and its technology and asset-bridging operations would move to Zero Gravity. That dated notice supersedes generic promotional pages that still present ShuttleFlow as an unchanged Foundation-operated product. The announcement said users who had completed destination claims needed no additional migration operation. This is a historical publisher statement, not an assurance that an arbitrary bridge route remains supported today or that every later interface shares the original service's controls.
Current oracle documentation says Pyth's Conflux eSpace support ended on July 31, 2026 and describes a replacement maintained for the Conflux community. The replacement uses an upgradeable proxy and authorized updaters, with roughly hourly prices. Its spot-reading interface is compatible with selected Pyth calls, but it does not preserve Pyth's publisher model or all functions. In particular, EMA reads and the old pull-update flow are not equivalent. An application must assess freshness and authority, not infer identical security from familiar method names.
Burns, games and physical products inspire different supporters
Oneofall's early CIP-107 discussion welcomed a potential burn mechanism as beneficial to holders. Other participants asked which operations released collateral and whether a zero-burn option made sense. The original thread therefore captures both scarcity enthusiasm and demands for understandable mechanics. It is not the final storage-point specification and cannot prove a price effect. The subsequent implementation should be read separately, while the discussion remains useful evidence of how investor hopes shaped attention to a technical proposal.
In a 2021 Reddit debate, german_bruce_lee predicted a prominent role in blockchain gaming, connecting technical capacity with an expected market opportunity in China. Sick-of-this123 asked for actual applications, and the original author acknowledged not having extensively investigated existing games. That exchange documents an investment narrative and its evidentiary weakness in the participants' own words. It does not establish a government-guaranteed monopoly, current regulatory permission or present gaming-market share.
The aspiration should remain attached to its author and historical setting.
Jhassiel Alcarraz's fashion-marketplace grant application proposes linking physical garments to NFTs and redemption. The attraction is ordinary commerce and authenticity, not only another financial pool. In April 2026, Mate_Conflux asked for stronger evidence about shipping, audits, impact and sustainability; a later reply promised a testnet prototype. The exchange remains a reviewed proposal and prototype commitment, with fulfillment still to demonstrate. The proposed physical business still requires off-chain obligations that an ownership token cannot perform by itself.
ここまでの道のり。
- 2020-10-29
Tethys mainnet launches
F2pool's contemporaneous operator update records the launch before its own CFX pool announcement.
- 2022-10-08
An emissions debate opens
Oneofall starts a public discussion of CIP-94 choices and the tradeoff between rewards and supply.
- 2023-02-07
Holders discuss a proposed burn mechanism
Oneofall opens the CIP-107 discussion, followed by questions about collateral and implementation.
- 2023-11-07
ShuttleFlow's service transition is announced
The Foundation publishes a closure and transfer notice for the following day.
- 2025-03-12
CIP-156 is created
The proposal records a move from permanent PoS forfeiture toward a temporary lock, later marked Final.
- 2026-03-26
Parameter-vote results are published
The team announces the twentieth round's reward and storage-point changes with height-dependent implementation timing.
- 2026-05-20
A node security update is announced
Operators are urged to install v3.0.3-fix for a possible panic condition.
- 2026-08-25
The 3.1 epoch boundary is recorded
The public chain record timestamps epoch 155,140,000 on this date; the reviewed finalized history had passed it.
信念、目標、未解決の問い。
これらは出所を明記した見解であり、賛同を示すものではありません。各証拠ファイルを開き、裏付けの記録と、そこから分かることの限界を確認してください。
議論のある解釈Reward security work or constrain dilution?
証拠ファイルを開く
GLM, obaska and jhassiel_alcarraz express different preferences about emissions and economic management.
物語の出所
The original October 2022 CIP-94 community discussion.
記録が裏付けること
- Named participants argue for larger rewards, smaller increases and supply restraint.
証明できないこと
- Their preferences are neither a vote result nor a September 2026 parameter snapshot.
注目する点
- Compare enacted rewards with participation and application demand.
将来の可能性A burn becomes a holder's hope
証拠ファイルを開く
Oneofall expects a proposed storage-related burn to benefit long-term holders.
物語の出所
The February 2023 CIP-107 thread.
記録が裏付けること
- Replies ask how collateral release works and question the available choices.
証明できないこと
- The early discussion differs from the final storage-point design and proves no price effect.
注目する点
- Read the final implementation and observed issuance alongside burn totals.
議論のある解釈A large market still needs actual games
証拠ファイルを開く
German_bruce_lee anticipates gaming leadership; sick-of-this123 asks for deployed applications.
物語の出所
An August 2021 Reddit discussion.
記録が裏付けること
- The advocate acknowledges limited research into existing games after being challenged.
証明できないこと
- The post does not prove regulatory privilege, monopoly or later adoption.
注目する点
- Follow maintained games and recurring users rather than repeating market-size analogies.
記録に残る信念Fashion ownership with a real redemption obligation
証拠ファイルを開く
Jhassiel Alcarraz proposes using Conflux for garment authenticity and secondary trading.
物語の出所
The Phygital Marketplace grant thread, with review continuing in 2026.
記録が裏付けること
- Mate_Conflux requests clearer shipping, security and sustainability evidence.
証明できないこと
- An application and promised prototype are not verified production fulfillment.
注目する点
- Look for audited contracts, completed redemptions and continuing operations.
出典ライブラリ。
一次資料は仕組みや決定を説明し、コミュニティの記録は参加者が何を信じていたかを示します。以下の日付はリンクの確認日です。外部のページは変更される場合があります。
- Conflux Spaces ↗Conflux documentation · primary · 確認日 2026-09-30
- Conflux economics ↗Conflux documentation · primary · 確認日 2026-09-30
- Conflux pool launch and Tethys history ↗F2pool · primary · 公開日 2020-11-09 · 確認日 2026-09-30
- Hydra v2.0 hardfork ↗Conflux documentation · primary · 確認日 2026-09-30
- Conflux consensus ↗Conflux documentation · primary · 確認日 2026-09-30
- PoS staking overview ↗Conflux documentation · primary · 確認日 2026-09-30
- CIP-156: change the PoS malicious penalty ↗Peilun Li and Chenxing Li · primary · 公開日 2025-03-12 · 確認日 2026-09-30
- Conflux v3.1 hardfork specification ↗Conflux documentation · primary · 確認日 2026-09-30
- eSpace mapped addresses ↗Conflux documentation · primary · 確認日 2026-09-30
- CrossSpaceCall internal contract ↗Conflux documentation · primary · 確認日 2026-09-30
- SponsorWhitelistControl ↗Conflux documentation · primary · 確認日 2026-09-30
- Conflux v2.3 storage-point specification ↗Conflux documentation · primary · 確認日 2026-09-30
- Conflux base fee ↗Conflux documentation · primary · 確認日 2026-09-30
- Key parameters round 20 results ↗ConFi-Conflux · primary · 公開日 2026-03-26 · 確認日 2026-09-30
- Conflux governance overview ↗Conflux documentation · primary · 確認日 2026-09-30
- How will you vote: CFX emission voting with CIP94 ↗Oneofall and Conflux forum participants · community · 公開日 2022-10-08 · 確認日 2026-09-30
- Conflux v3.0.3-fix upgrade announcement ↗ConFi-Conflux · primary · 公開日 2026-05-20 · 確認日 2026-09-30
- Conflux v3.1.0 mainnet hardfork announcement ↗ConFi-Conflux · primary · 公開日 2026-08-20 · 確認日 2026-09-30
- Conflux public mainnet chain records ↗Conflux public RPC · primary · 確認日 2026-09-30
- ShuttleFlow closure and service-transfer notice ↗ConFi-Conflux · primary · 公開日 2023-11-07 · 確認日 2026-09-30
- Conflux price oracle and Pyth migration ↗Conflux documentation · primary · 確認日 2026-09-30
- Community discussion of CIP-107 ↗Oneofall and Conflux forum participants · community · 公開日 2023-02-07 · 確認日 2026-09-30
- A Conflux gaming thesis and its critics ↗german_bruce_lee and Reddit participants · community · 公開日 2021-08-22 · 確認日 2026-09-30
- Phygital Marketplace grant proposal ↗Jhassiel Alcarraz and Conflux grants participants · community · 公開日 2025-10-20 · 確認日 2026-09-30