The scientific foundation
Bring relativity, quantum systems and precision studies into a common research workflow. Inspect the model, execute the calculation and follow the evidence behind each result.
Scientific workflows
Explore SR and static-GR timing, clock histories, photon propagation and PF constructions within declared configurations.
Inspect protocol and component models, density matrices and performance measures with explicit assumptions.
Trace mission, optical and timing inputs into observables and companion studies.
Live satellite tracking
PF retrieves published CelesTrak orbital elements and propagates selected satellites at the current UTC time. Moving-endpoint photon calculations connect predicted trajectories to visibility, release timing and declared quantum-channel models.
Use SGP4 with published GP or supported TLE inputs. Preserve the element epoch, retrieval time and source hashes, or replay a frozen input set.
Compare SGP4 with configured numerical two-body or J2 orbit models. Inspect prediction differences without treating either branch as measured ground truth.
Inspect photon interception, Earth visibility, timing, conditional quantum states and collection-rate predictions. Optional QKD studies use declared optics, detectors and background assumptions.
Live means current-time model updates. Orbital elements refresh periodically; the default refresh is two hours. Solver runtime determines achievable update cadence. This is not streaming spacecraft telemetry, measured quantum-link performance or an operational navigation service.
Qiskit modeling
PF’s Qiskit/Aer routes execute local density-matrix circuits, component benchmarks and seeded measurement sampling. Change a declared noise model, compare quantum-state metrics and inspect the state behind the headline result.
Separate exact fidelity and purity from finite-shot Pauli measurements. The Gallery compares two Bell encodings under a shared, explicitly configured Gaussian phase-noise model.
With Qiskit Experiments installed, reconstruct two-qubit states or one-qubit processes. Eligible saved mission states can seed new Aer measurements with retained parent lineage.
These supported routes are local simulations. They do not execute IBM quantum hardware; resampling a saved state does not turn it into a new physical measurement or identify its channel.
From computation to assessment
Configurations, numerical arrays, logs, source references, plots and report exports make a run inspectable. Resolution comparisons and reference checks help distinguish numerical agreement from a research interpretation.
The AI assistant proposes a private candidate. PF executes baseline and candidate stacks and records comparisons. You decide whether to activate a passing layer; an assistant suggestion does not silently become a shared-core update.
Numerical agreement supports the stated model and test domain. It does not by itself establish experimental validity, a PF operational advantage or universal precision.
Scientific software connections
Use supported library operations, scientific workers and recorded-input adapters in focused studies. Where compatible parent fields exist, a companion consumes the actual saved state, orbit, clock history or ray data and records its own assumptions and results.
Compare hypotheses under controlled settings, inspect numerical agreement across implementations and retain the inputs and methods needed for reproduction.
Explore design alternatives, identify sensitive parameters and prioritize what to measure next. Private layers preserve your implementation choices and their validation history.
These are workflow benefits supported by the implemented controls and artifacts. Simulation results do not establish measured mission accuracy, hardware certification, quantified cost savings or a PF-specific advantage.
A research platform that develops with you
r34 uses curated scientific references, versioned source catalogs and explicit numerical reference checks. Software optimizations, precision improvements and integrations enter the shared core through operator-controlled releases. Your private subsystem stays distinct from those releases.
When adopting a new core release, rebase and revalidate private layers against that source version. This preserves the connection between an implementation, its validation and the research that uses it.
The longer-term vision is to bring peer-reviewed literature into a traceable pipeline for model validation, precision enhancements and future core updates. Automatic continuous literature ingestion and real-time core self-updating are not features of r34; the current foundation is curated references and explicit, reviewable checks.
Researcher control
Combine the scientific foundation with your hypotheses, configurations and permitted implementation changes. Keep the history of what changed and test its consequences before using it in new research.
Explore private stacks