BitBOINC

From BOINC Projects
Jump to navigation Jump to search



BitBOINC
Project
StatusActive
CategoryCryptography, astrophysics
ComputeCPU & GPU
Development
DeveloperAlperen Yavuz
AuthorAlperen Yavuz
MaintainerAlperen Yavuz
Initial releaseSeptember 28, 2026  (0 years ago)
CompletedNo
Repositoryhttps://github.com/alplix/kampylos (Kampylos)
Software
Written inC, C++
Operating systemWindows, Linux, macOS, Android
BOINC statistics
Stats as ofOctober 10, 2026  (0 years ago)
Performance142.337 TFLOPS
Active users101
Total users101
Active hosts454
Total hosts458
Metadata
Websitehttps://bitboinc.athena.org.tr/

BitBOINC is an active BOINC-based volunteer computing project that uses donated computer time to run several independent research efforts on the same pool of volunteer CPU and GPU resources.[1] It is listed among the projects documented by this wiki's BOINC projects registry. The project launched as a search for the private keys associated with selected, long-unclaimed Bitcoin puzzle addresses, and subsequently added an application targeting the Certicom ECC Challenge, a gravitational microlensing modelling application, a search for the home stars of known interstellar objects, and a search for the birth clusters of runaway stars and pulsars.[2][3] The project is part of the Athena network and is operated by Alperen Yavuz.[1]

BitBOINC distinguishes its Bitcoin-related work from Bitcoin mining. It does not create new coins, mine blocks, or contribute to the operation or security of the Bitcoin network. Instead, it searches a fixed set of publicly known addresses whose private keys were deliberately placed within specified numerical ranges as a public challenge.[2]

History

BitBOINC was publicly announced on 28 September 2026. The project initially described its applications as being rolled out in stages while applications for different operating systems, processor types, and graphics processors were compiled and tested.[4] On the same day, the project reported that it had been delisted by the third-party statistics site stats.free-dc.org, stating that it was exporting statistics hourly in the standard BOINC format and had been unable to reach the site's operators.[5]

On 29 September 2026, the project announced that both of its original search applications were active. It also announced support for Windows, Linux, macOS, and Android on CPUs, together with GPU support using NVIDIA CUDA, OpenCL, and Apple Silicon Metal.[6]

Later on 29 September 2026, the project announced a third application, Sphinx, which applies the same kangaroo search used by Keraunos to the Certicom ECC Challenge rather than to Bitcoin addresses.[7] On 30 September 2026, a fourth application, Kampylos, was announced as being in development; unlike the cryptographic applications it performs astrophysical modelling rather than a key search.[8]

On 2 October 2026 the operator reported scheduler and validator problems caused by Cloudflare caching together with filled scheduler-log and data-compression servers, and described a large volume of automated requests that led the project to rate-limit incoming traffic. The same notice stated that the server had supported IPv6 from the day it was set up and asked volunteers to enable IPv6 for the fastest connection.[1] On 3 October 2026 the project noted that BOINC Synergy had published a wiki article about BitBOINC.[1]

A later revision of the Kampylos announcement, last modified on 6 October 2026, stated that Kampylos had left its beta period and was processing real archival photometry, with completed grid-search datasets published on Zenodo.[8] On 7 October 2026 the project invited volunteers to suggest names for two Kampylos objects (a 2002 MOA alert identified as a flaring dwarf nova, and a 2003 microlensing event) and to opt in to being named in scientific notes.[1]

On 9 October 2026 BitBOINC announced two further astronomy applications, Nostos and Phygas. Both use Gaia measurements of tens of millions of stars and integrate orbits backward through the Milky Way. The announcement described them as CPU applications for Windows, Linux, macOS, and Android, each with its own badge series, and stated that all results including raw data would be published openly.[3]

Research and computation

Bitcoin puzzle addresses

The project works on a series of Bitcoin addresses created as a public cryptographic challenge in 2015. According to independent trackers of the challenge, an anonymous creator funded a batch of puzzle outputs in a transaction confirmed on 15 January 2015 in block 339085; the series later became known as a sequence of 160 addresses whose private keys are restricted to successively larger bit ranges.[9][10] According to BitBOINC, the puzzles were numbered sequentially, with the private key for puzzle number n placed somewhere in an n-bit range. The series extends to puzzle number 160, although many of the larger puzzles remain unsolved.[2]

Bitcoin transactions are authorised with elliptic-curve cryptography. In simplified form, a candidate private key k is transformed into a public key using elliptic-curve multiplication:[11]

P=kG

where G is the generator point of the elliptic curve used by Bitcoin. The resulting public key is then processed to derive an address. A candidate is successful only when the derived address matches the target address.

The project does not publish the precise subranges already searched. It instead publishes aggregate progress information, because revealing the searched locations could allow unrelated searches to avoid the same regions.[2] Aggregate figures are published on a puzzle progress page, which reports distinguished-point counts for Keraunos and completed-chunk counts for Potamos, along with the on-chain balances of the queued target addresses, refreshed hourly from a public block explorer.[12]

As of 10 October 2026, the project listed two puzzles as actively being searched and a queue of a further 75 puzzles, of which four were reported to have an exposed public key and therefore to be eligible for Keraunos. The combined balance of the remaining targets was given as 903.0168 BTC.[12]

Applications

BitBOINC provides six applications. Three are cryptographic searches and three are astronomy applications.[13]

Application Task Status (as of 10 October 2026)
Potamos Sequential keyspace scan of Bitcoin puzzle ranges Active
Keraunos Bounded discrete-logarithm search (kangaroo) on Bitcoin puzzles Active
Sphinx Discrete-logarithm search against the Certicom ECC Challenge Active
Kampylos Binary-lens gravitational microlensing grid search Active
Nostos Backward Galactic tracing of interstellar objects against Gaia stars Active
Phygas Birth-cluster search for runaway stars and pulsars Active

Potamos

Potamos performs a sequential search, testing private keys one at a time. The project describes this method as exhaustive but practical mainly for smaller key ranges, noting that it does not require the target's public key to be known.[2][12]

Keraunos

Keraunos uses Pollard's kangaroo algorithm, a generic discrete logarithm method introduced by John M. Pollard in 1978. The algorithm searches for an unknown exponent x satisfying

αx=β

inside a known interval of width w. Under standard randomness assumptions its expected running time is

O(w)

group operations, a square-root improvement over a linear scan of the same interval.[14][15] BitBOINC states that Keraunos can be used only when the public key for a target has already been exposed on the Bitcoin blockchain, such as when the address has previously sent a transaction; most unclaimed puzzles do not meet this condition.[2][12]

Keraunos is described as a collaborative rather than merely parallel search: each host runs its own randomised walk over the same range and periodically reports distinguished points to a shared table, and a collision between points reported by any two hosts allows the private key to be computed directly.[2] In the classical formulation, a collision between a "tame" walk that starts from a known point and a "wild" walk that starts from the target yields

x≡b+d−dj(modn)

where d and dj are the distances travelled by the two walks.[14]

Only one target is processed at a time for each application, and a given puzzle may be worked by one application, the other, or both simultaneously.[12] According to the project, a collision is detected automatically, upon which the server stops generating further work for that puzzle and activates the next queued target; if an outside party claims a target's coins first, the switch occurs after the project operator's next check of the target's on-chain balance.[2]

As of 29 September 2026, BitBOINC reported that Keraunos was working on puzzle number 140 and Potamos was working on puzzle number 71.[6] Both assignments were still current on 10 October 2026. On that date Keraunos reported about 5.01 million distinguished points from 368 hosts on puzzle 140, and Potamos reported 112,466 completed chunks from 285 hosts on puzzle 71.[12]

Sphinx

Sphinx, announced on 29 September 2026, targets the Certicom ECC Challenge, a set of elliptic-curve discrete-logarithm problems published by Certicom in 1997 with cash prizes attached. The application reuses the collaborative kangaroo search mechanism developed for Keraunos, applied to a non-Bitcoin elliptic curve; the project emphasises that no coins, wallets, or addresses are involved and that the fund-handling procedures described for the Bitcoin puzzles therefore do not apply to it.[2][7]

The project states that Sphinx's first target is ECCp-131, with ECCp-163 and ECCp-191 queued next, and ECCp-239 and ECCp-359 planned subject to wider-precision arithmetic support.[7] The ECCp-131 problem has been open since the challenge's release in November 1997 and carries a stated prize of US$20,000.[16] BitBOINC states that it has not independently confirmed that the prizes are still honoured and advises participants to treat a solution as a cryptographic achievement rather than as a guaranteed payout.[2][7]

Kampylos

Animation of gravitational microlensing: a foreground mass bends and temporarily brightens the light of a more distant star. Kampylos searches the more complicated case in which the lens is a pair of masses.

Kampylos (from the Greek καμπύλος, "bent" or "curved") is an astrophysics application that exhaustively models binary-lens gravitational microlensing events, in which light from a background star is bent by a foreground pair of masses such as a binary star or a star with a planet.[17] Microlensing is the same physical effect as strong gravitational lensing, but the lens is of stellar or planetary mass, so the multiple images are not resolved and the observable is a transient brightening of the source.[18] Surveys including the Optical Gravitational Lensing Experiment (OGLE), which has operated from Las Campanas Observatory in Chile since 1992, have used this method both to detect exoplanets and to compile large catalogues of variable stars.[19]

Fitting a binary-lens event requires a grid search over lens parameters including mass ratio q, projected separation s (in units of the Einstein radius), and source trajectory angle. The project notes that the resulting χ2 surface is frequently multimodal, so that several distinct parameter combinations may fit the same light curve almost equally well, and that published analyses commonly report the first solution found rather than a complete accounting of the parameter space.[2][17] A well-known example is the close/wide degeneracy, in which a lens with separation s and one with separation 1/s can produce similar light curves.[8] Each task evaluates the forward-modelled magnification curve against real photometry at every grid point and returns a full χ2 map rather than a single best fit. The binary-lens solver is built on the VBMicrolensing library. According to the project, a typical grid is about 10 by 10 in (log⁡s,log⁡q), and at every cell a Nelder-Mead local optimisation with multiple random restarts is run over the remaining parameters, with source and blend fluxes solved in closed form.[8][17]

Input photometry is drawn from the public archives of the OGLE Early Warning System, the MOA alert archive, and the KMTNet data archive, which together have alerted on tens of thousands of events since the 1990s.[17] The project states that results, including complete solution maps, are published openly on Zenodo, with notable findings submitted to Research Notes of the AAS.[8][17] As of 10 October 2026 the published-dataset list contained 141 Zenodo records (99 from OGLE-IV and 42 from MOA), with KMTNet events described as queued next.[20] The project also maintains a public list of events that pass an automatic anomaly threshold. One strong-looking candidate, OGLE-2011-BLG-0952, was later classified after human review as a systematic artifact rather than a planet.[8]

The project cites the Nancy Grace Roman Space Telescope, launched on 30 August 2026 with microlensing science operations expected from around 2027, as a future source of a substantially larger event stream.[2][17][21]

Kampylos was first listed as beta software. On 30 September 2026 the project described it as in development and not yet sending work while the solver underwent correctness testing, and the repository likewise described the effort as early-stage.[8][17] Test work units were nevertheless distributed, and on 1 October 2026 a participant reported that all returned Kampylos tasks had been marked invalid. The project operator responded that Kampylos was a test application whose core code was not yet fully configured and that returned results were being used to support development.[8] The exchange prompted criticism from one participant, who argued that a project distributing work should grant credit for correct results, and support from others, who noted that only large projects typically maintain separate beta-test sites and that beta work units are normal in a newly launched project.[8] The operator also stated at that time that the GPU version of Kampylos had not been released because of an unresolved defect in its FP32 implementation.[8]

By 6 October 2026 the same thread described Kampylos as fully in production on real archival data, with CPU, CUDA, and OpenCL backends running the same algorithm across Linux, Windows, macOS (Intel and Apple Silicon), and Android.[8] The public repository later listed GPU backends including CUDA, OpenCL, Apple Metal, and MUSA, and described the solver as live.[17] As of 10 October 2026 the project's status page reported 20,148 valid Kampylos results in the preceding seven days, together with invalid and error counts still characteristic of a young application.[13]

Nostos

1I/'Oumuamua, the first confirmed interstellar object, imaged on 28 October 2017. Nostos traces its inbound path backward against tens of millions of Gaia stars.

Nostos (Greek for "homecoming") traces known interstellar objects backward through the Milky Way in order to find stars they may have passed closely and slowly, which are treated as candidate birthplaces.[2][22] As of 2025, three such objects have been confirmed inside the Solar System: 1I/'Oumuamua, discovered on 19 October 2017 with the Pan-STARRS 1 telescope in Hawaii; 2I/Borisov, discovered on 29 August 2019 by amateur astronomer Gennadiy Borisov; and 3I/ATLAS, discovered on 1 July 2025 by the ATLAS survey. Their inbound speeds relative to the Sun are about 26.3 km/s, 32.3 km/s, and 58 km/s respectively.[23][24][25][26]

The search uses stars with a full six-dimensional phase-space measurement in Gaia Data Release 3. Gaia DR3, published on 13 June 2022, includes combined radial velocities for about 33.8 million stars.[27] BitBOINC states that Nostos applies the official Gaia parallax zero-point correction of Lindegren and colleagues before integrating orbits.[2][28] Galactic motion is integrated in the published mass model of Allen and Santillán (1991), which represents the Milky Way as a bulge, a flattened disk, and a dark-matter halo, with a timestep of one thousand years over a look-back of 100 million years.[22][29]

Each task follows one interstellar object against a block of Gaia stars. The project about page and the 9 October 2026 technical announcement describe 1,500 stars per task; the live Nostos progress page, accessed the following day, states that a task covers 500 stars on a CPU or 20,000 stars on a GPU.[2][22][30] For every star the code converts Gaia measurements into a Galactic position and velocity, integrates the star and the object backward together, and records the closest approach. The object's inbound velocity is drawn 100 times from its published uncertainty, including correlations among components, so that a robust encounter is one that remains close across those copies. Only stars that come within 50 parsecs (about 160 light years) are kept. The project estimates about 2.4 CPU-years of computing per object for a full 33-million-star search.[22] On 9 October 2026 the operator wrote that a modern CPU task was expected to take about 20 minutes and that a GPU version, then described as forthcoming, would cover about 27 times as many stars and take 50 to 70 minutes.[22] Returned close encounters are rechecked on the server while also varying the star's own Gaia errors. The project treats encounters older than about 50 million years as lower confidence, because small velocity errors grow with time.[22]

Nostos is intended to extend earlier Gaia-based origin searches. Bailer-Jones and colleagues (2018) used Gaia DR2 to identify candidate home stars of 'Oumuamua, including HIP 3757 and HD 292249; a 2020 study of 2I/Borisov, using about 7.4 million Gaia DR2 stars, found a closest encounter with the M dwarf Ross 573 about 910,000 years ago at a median separation of 0.068 pc.[31][32] BitBOINC states that, before any volunteer task was issued, its pipeline had to reproduce those published encounters of 'Oumuamua with HIP 3757 and HD 292249 and of Borisov with Ross 573 to within the published uncertainty ranges. Two bugs found during that check (use of the outgoing rather than incoming asymptote, and a starting speed 2 to 4 percent too high) were corrected before launch.[22] The project notes that a star the object merely flew past at tens of kilometres per second is a poor birthplace candidate, because an object ejected from a planetary system typically leaves at only a few kilometres per second.[22]

As of 10 October 2026 Nostos was searching 1I/'Oumuamua, with 2I/Borisov and 3I/ATLAS queued. The progress page reported 2,330,500 of about 33,163,000 stars traced (7.03 percent), 15,293 stars within 50 pc, four closer than 1 pc, and a closest approach so far of 0.43 pc (1.4 light years) some 3.04 million years ago. Tasks are computed by two volunteers whose results must agree.[30] The project cites the Vera C. Rubin Observatory as a likely source of additional interstellar objects that can be added to the same search.[22][23] Related Galactic-dynamics work with Gaia data is also carried out by the separate BOINC project Gaia@home.

Phygas

Artist's impression of the European Space Agency Gaia spacecraft, whose Data Release 3 supplies the 33 million stars with full 3D velocities used by both Nostos and Phygas.

Phygas (Greek for "fugitive") traces fast-moving stars and pulsars backward in order to identify which young star cluster they may have left, when they left, and how fast they were ejected.[2][33] Two established channels produce such fugitives: a supernova in a binary, which can unbind the companion as a runaway star and impart a natal kick of hundreds of kilometres per second to the leftover neutron star; and close encounters in the crowded core of a young cluster.[33][34] Pulsars are rotating magnetised neutron stars whose beams of radiation are observed as pulses when they sweep across the Earth; their proper motions are compiled in catalogues such as the ATNF pulsar catalogue.[34] Related pulsar searches on BOINC include Einstein@Home.

Illustration of the lighthouse effect of a pulsar. Phygas traces pulsars with measured proper motions back against more than 4,000 young star clusters.

The stellar sample is the same Gaia DR3 set of about 33 million stars with full 3D velocities used by Nostos. The pulsar sample is described as all 702 pulsars in the ATNF catalogue that have a measured proper motion and a distance, of which 356 have a measured parallax; otherwise distance comes from the dispersion measure. Because a pulsar's radial velocity is not observed, it is drawn from the known spread of neutron-star birth velocities (Hobbs and colleagues, 2005, about 265 km/s in each direction).[33][35] The cluster sample is 4,087 star clusters and moving groups that can be younger than 100 million years, taken from the Gaia-based catalogue of Hunt and Reffert (2024). Older clusters are omitted because an escape more than 100 million years ago cannot be traced reliably. For 878 clusters without a measured mean radial velocity, that component is drawn around the value predicted by Galactic rotation, plus or minus 10 km/s.[33][36] The Galactic potential is the same Allen and Santillán (1991) model used by Nostos. Each pair is followed back for as long as the cluster could have existed, at most 100 million years.[33]

There are two kinds of task. Star tasks take 20,000 stars each and compare every star with all 4,087 clusters using a faster linear screening method that traces best-estimate orbits together with six slightly shifted orbits, so that the size and shape of the uncertainty cloud are known at every time. Pairs with a probability of at least 0.1 percent that the star was inside the cluster are kept. The project estimates that one star against all clusters takes about 0.15 seconds, so a task runs roughly 50 minutes, and that the whole stellar catalogue is about 1,660 tasks.[33] Pulsar tasks take one pulsar each and use a full Monte Carlo of 1,000 copies of the pulsar and 1,000 copies of each nearby cluster, because pulsar distance and velocity uncertainties are too large for the shortcut. Integration uses 50,000-year steps, with closest approach refined between steps.[33]

The project emphasises that most raw cluster crossings are coincidences. In a control run that shuffled stellar velocities, about 90 percent as many "candidates" appeared as in the real run. An excess remained for slow pairs with relative speed under 20 km/s (758 against 454 in the shuffled test). Phygas therefore compares every cluster's real counts with shuffled control runs and counts only the excess.[33] A velocity error of 1 km/s moves a star by about 1 parsec per million years, so distant pairs at tens of millions of years have large uncertainties and correspondingly low probabilities.[33]

Before launch, the pulsar pipeline was required to reproduce published associations: PSR B1259-63 (also designated J1302-6350) ranked UBC 525 first among all 4,087 clusters; for the double pulsar J0737-3039, the moving group Theia 58 appeared in the top four, above the cluster OC 0450, matching a 2026 study cited by the project; and a 2014 link of J0826+2637 to Stock 7 was found to be weak (7.5 percent of copies) after Gaia distances were used.[33]

As of 10 October 2026 the Phygas progress page reported 13,860,000 of 33,162,595 stars traced (41.79 percent) and 399 of 702 pulsars (56.84 percent), with 21 volunteers contributing. Pulsar tasks were described as going out first, then the stars. Every task is computed by two volunteers whose results must agree, and the project states that all raw results will be published openly.[37] Other BOINC projects that model the Milky Way or its stellar populations include MilkyWay@home, Universe@Home, and Gaia@home.

Volunteer computing

BitBOINC distributes work through BOINC, the same middleware used by projects such as SETI@home, Einstein@Home, and World Community Grid. A volunteer installs the BOINC client, attaches it to the BitBOINC project at https://bitboinc.athena.org.tr/, and allows the client to run tasks according to the volunteer's resource preferences. The project states that any computer with a supported CPU or GPU can participate, and that BitBOINC may not yet appear in BOINC Manager's public project list, in which case the URL is pasted under Grid Computing.[1]

The project reports that returned results are independently checked before being accepted and credited. Nostos and Phygas additionally require that each task be computed by two different volunteers whose results agree.[30][37] The project also announced that its credit calculation had been changed to use a native BOINC-style calculation based on the work performed by the participant's hardware.[6]

Supported CPU platforms include Windows, Linux, macOS, and Android. GPU acceleration is available through NVIDIA CUDA, OpenCL implementations, and Metal on Apple Silicon, subject to application and hardware support.[6] The 9 October 2026 announcement of Nostos and Phygas described both as CPU applications that volunteers receive automatically unless they have changed project preferences.[3] As of 10 October 2026 the Nostos progress page also described GPU tasks.[30] On the same date the status page reported 135 active hosts with NVIDIA CUDA GPUs, 159 with AMD or Intel GPUs using OpenCL, and 4 with Apple GPUs using Metal.[13] Combined recent throughput was given as 142,337.01 GFLOPS (142.337 TFLOPS), with 101 users and 454 hosts showing recent credit.[13]

In the seven days preceding 10 October 2026 the status page reported completed-task counts of 100,594 for Potamos, 89,784 for Keraunos, 21,913 for Sphinx, 19,342 for Kampylos, 9,552 for Nostos, and 1,878 for Phygas.[13]

Security and handling of results

BitBOINC states that its applications read the work packet supplied by the project and write their results within the BOINC slot directory. It also states that the applications do not open their own network connections, with communications handled by the BOINC client.[2]

If a private key is found, the project says that the result is encrypted immediately, within the same process that discovered it, using RSA-4096 with OAEP padding, and that the plaintext key is never written to disk or to logs. According to the project, only the encrypted result is stored by the server, in the database, logs, and a notification email, while the corresponding decryption key is kept offline by the project operator and has never been present on the server.[2]

The project further states that it has no automated code path for transferring funds. Its published procedure is for a person to retrieve and decrypt the result offline, independently verify that the key corresponds to the target address, and then manually transfer any funds to a hardware wallet.[2]

BitBOINC describes a proposed distribution of recovered funds among platform operation, science projects, nature and environmental protection, and participating volunteers. The volunteer share is described as a weighted random drawing among recent contributors to the particular puzzle rather than as an equal payment to every participant.[2]

These procedures apply only to the Bitcoin puzzle applications. The project states that no funds or custody arrangements are involved in Sphinx, Kampylos, Nostos, or Phygas.[2][3]

Community

BitBOINC provides project support through the Libera.Chat IRC network. Its listed channel is #bitboinc, reachable at irc.libera.chat:6697 over TLS, and the project recommends its own Rhizome client for users who do not already use an IRC client. The channel is shared by all applications on the project and by the wider Athena project family.[1]

The project also provides account registration, discussion forums, progress information, achievement badges, and user levels through its website. A badge system comprising 100 badges across ten categories, together with an open-ended level and title system, was announced on 29 September 2026.[1][6] Nostos tasks count toward a separate badge series described as running from Wanderer to Odysseus, and Phygas has its own badges.[30][3] Volunteers may optionally add an official name and country on an account page if they wish to be credited in Kampylos scientific notes; otherwise they remain anonymous and are only counted.[1]

The BitBOINC banner
The BitBOINC banner

See also

External links

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 BitBOINC. Athena. Retrieved 2026-10-10.
  2. ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 2.11 2.12 2.13 2.14 2.15 2.16 2.17 2.18 2.19 2.20 About BitBOINC. Athena. Retrieved 2026-10-10.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 (2026-10-09).Two new research projects: Nostos and Phygas. BitBOINC. Retrieved 2026-10-10.
  4. ↑ (2026-09-28).Welcome. BitBOINC. Retrieved 2026-09-30.
  5. ↑ (2026-09-28).The problem with statistics websites. BitBOINC. Retrieved 2026-10-02.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 (2026-09-29).BitBOINC is fully up and running. BitBOINC. Retrieved 2026-09-30.
  7. ↑ 7.0 7.1 7.2 7.3 (2026-09-29).SPHINX. BitBOINC. Retrieved 2026-10-02.
  8. ↑ 8.00 8.01 8.02 8.03 8.04 8.05 8.06 8.07 8.08 8.09 8.10 (2026-09-30).KAMPYLOS. BitBOINC. Retrieved 2026-10-10.
  9. ↑ (2026-06-23).Key Hunters Eye $58.87M Bitcoin Puzzle as 916 BTC Sits Unsolved in 78 Addresses. Cryptonews. Retrieved 2026-10-10.
  10. ↑ (2025-02-21).Puzzle 67 is taken. Reddit. Retrieved 2026-10-10.
  11. ↑ Bitcoin. Wikipedia. Retrieved 2026-10-10.
  12. ↑ 12.0 12.1 12.2 12.3 12.4 12.5 Bitcoin puzzle progress. BitBOINC. Retrieved 2026-10-10.
  13. ↑ 13.0 13.1 13.2 13.3 13.4 Project status. BitBOINC. Retrieved 2026-10-10.
  14. ↑ 14.0 14.1 Pollard's kangaroo algorithm. Wikipedia. Retrieved 2026-10-10.
  15. ↑ Pollard, John M..(July 1978).Monte Carlo Methods for Index Computation (mod p). Mathematics of Computation. pp. 918-924. Retrieved 2026-10-10.
  16. ↑ Certicom ECC Challenge. Certicom. Retrieved 2026-10-02.
  17. ↑ 17.0 17.1 17.2 17.3 17.4 17.5 17.6 17.7 alplix/kampylos. GitHub. Retrieved 2026-10-10.
  18. ↑ Gravitational microlensing. Wikipedia. Retrieved 2026-10-10.
  19. ↑ Optical Gravitational Lensing Experiment. Wikipedia. Retrieved 2026-10-10.
  20. ↑ Kampylos: published datasets. BitBOINC. Retrieved 2026-10-10.
  21. ↑ Nancy Grace Roman Space Telescope. Wikipedia. Retrieved 2026-10-10.
  22. ↑ 22.0 22.1 22.2 22.3 22.4 22.5 22.6 22.7 22.8 (2026-10-09).NOSTOS. BitBOINC. Retrieved 2026-10-10.
  23. ↑ 23.0 23.1 Interstellar object. Wikipedia. Retrieved 2026-10-10.
  24. ↑ 1I/'Oumuamua. Wikipedia. Retrieved 2026-10-10.
  25. ↑ 2I/Borisov. Wikipedia. Retrieved 2026-10-10.
  26. ↑ 3I/ATLAS. Wikipedia. Retrieved 2026-10-10.
  27. ↑ (June 2023).Gaia Data Release 3. Properties and validation of the radial velocities. Astronomy & Astrophysics. pp. A5. Retrieved 2026-10-10.
  28. ↑ (May 2021).Gaia Early Data Release 3. Parallax bias versus magnitude, colour, and position. Astronomy & Astrophysics. pp. A4. Retrieved 2026-10-10.
  29. ↑ (October 1991).An improved model of the galactic mass distribution for orbit computations. Revista Mexicana de Astronomia y Astrofisica. pp. 255-263. Retrieved 2026-10-10.
  30. ↑ 30.0 30.1 30.2 30.3 30.4 Nostos: where did the interstellar objects come from?. BitBOINC. Retrieved 2026-10-10.
  31. ↑ (October 2018).Plausible Home Stars of the Interstellar Object 'Oumuamua Found in Gaia DR2. The Astronomical Journal. pp. 205. DOI: 10.3847/1538-3881/aae3eb. Retrieved 2026-10-10.
  32. ↑ (February 2020).A search for the origin of the interstellar comet 2I/Borisov. Astronomy & Astrophysics. pp. A14. DOI: 10.1051/0004-6361/201937231. Retrieved 2026-10-10.
  33. ↑ 33.00 33.01 33.02 33.03 33.04 33.05 33.06 33.07 33.08 33.09 (2026-10-09).PHYGAS. BitBOINC. Retrieved 2026-10-10.
  34. ↑ 34.0 34.1 Pulsar. Wikipedia. Retrieved 2026-10-10.
  35. ↑ (July 2005).A statistical study of 233 pulsar proper motions. Monthly Notices of the Royal Astronomical Society. pp. 974-992. DOI: 10.1111/j.1365-2966.2005.09087.x. Retrieved 2026-10-10.
  36. ↑ (2024).Improving the open cluster census. III. Using cluster masses, radii, and dynamics to create a cleaned open cluster catalogue. Astronomy & Astrophysics. pp. A42. DOI: 10.1051/0004-6361/202348662. Retrieved 2026-10-10.
  37. ↑ 37.0 37.1 Phygas: where were the runaway stars born?. BitBOINC. Retrieved 2026-10-10.
Retrieved from "https://boincsynergy.ca/wiki/index.php?title=BitBOINC&oldid=1874"