Instrument · front door
The Virtual Collider
The accelerator you don't have to build.
Three particle-scale quantities — the proton-to-electron mass ratio, the gravitational constant, and the neutron–proton mass difference — each predicted by a closed-form, parameter-free expression from G's Ø Predictions (Parts I–III), each confronted with the value the precision-measurement community actually produced (CODATA 2018/2022), and each judged at G's own published tolerance — outside tolerance means that Part is dead, with no appeal. The single measured dimensionless input is α, the fine-structure constant; Part II, whose output carries units, additionally takes the electron mass me from CODATA (ħ and c are exact by SI definition). On the live ladder your browser computes every number at page-load; the identical Python twin runs the same arithmetic offline, self-verifying, in under a second.
What this is not
This instrument does not simulate particle collisions, and it has no affiliation with CERN or any accelerator programme. Nothing here models a beam, a detector, or an event. The "measured" column is CODATA — decades of real accelerator and precision work, distilled — not raw events.
And the deeper boundary, printed by the proof itself on every run: reproducing the arithmetic is internal consistency, not proof that nature agrees. This page establishes that G's published numbers follow from G's published formulas. Whether those closed forms are physics or numerology that happens to land is the open question, and it belongs to specialists.
1The three readings
verify.py
(Appendix B). α = 7.2973525693×10⁻³ (CODATA 2018) — the single measured
dimensionless input; Part II also takes me from CODATA (ħ and c are
exact by SI definition).
| Part | Predicted | Measured (CODATA) | Residual | G's tolerance | Verdict |
|---|---|---|---|---|---|
| I — mp/me | 1836.152673444 | 1836.152673426 | 0.010 ppb | 0.017 ppb | PASS |
| II — G | 6.7206×10⁻¹¹ N·m²/kg² | 6.6743×10⁻¹¹ | 0.69 % | 1.0 % | PASS |
| III — (mn−mp)/me | 2.5309939330 | 2.5309883000 | 2.23 ppm | 5.0 ppm | PASS |
Numbers verbatim from the self-verifying run below — this front door hardcodes nothing that run did not print. On the live ladder the same three formulas run in your browser and the PASS is never typed in; it is the live result of |residual| ≤ tolerance. Part II, at 0.69 % of a 1.0 % tolerance, is the coarsest of the three and the first place a sceptic should push.
2Run it yourself — offline, one command
Nothing here asks for trust. The Python twin is standard library only, needs no network, runs in under a second, and exits non-zero if any residual falls outside G's stated tolerance. From the repo root:
python3 the420code/proofs/particle-bench/demo.py # Parts I–III, self-verifying gate
Source: the420code/proofs/particle-bench/demo.py
· proof README. Run output,
reproduced verbatim:
particle-bench — G's particle-scale predictions vs measurement (no collisions simulated; no CERN affiliation; arithmetic vs CODATA) ==================================================================== Part I — proton-to-electron mass ratio m_p/m_e formula : 21^2*4 + 21*3 + 3^2 + a*21*(1-1/(84pi)) + a^2*21*16/1836 predicted: 1836.152673444 measured : 1836.152673426 (CODATA 2022) residual : 0.010 ppb tolerance 0.017 ppb PASS Part II — gravitational constant G formula : a^21 * (1 + 1/pi) * hbar*c / m_e^2 predicted: 6.7206e-11 N*m^2/kg^2 measured : 6.6743e-11 (CODATA 2018/2022) residual : 0.69 % tolerance 1.0 % PASS Part III — neutron-proton mass difference (m_n - m_p)/m_e formula : 3*(1 - 1/(2pi)) + a*(1 + 1/(2pi)) predicted: 2.5309939330 measured : 2.5309883000 (CODATA) residual : 2.23 ppm tolerance 5.0 ppm PASS ==================================================================== PASS — all 3 particle-scale predictions inside G's stated tolerances. (Reproducing the arithmetic is internal consistency, not proof that nature agrees.)
Provenance: the 0.017 ppb tolerance on Part I is the CODATA 2022 relative uncertainty of the measured ratio, 1836.152673426(32).
3Where this sits
- Three quantities, not a theory of QCD. No hadron spectra, no cross-sections, no running couplings. "Where's QCD?" is a fair question — the published falsification registry (549 conditions) holds the kill switches for that line of attack.
- The tolerances are G's own. 0.017 ppb / 1.0 % / 5.0 ppm are the published failure conditions, not ceilings chosen here to guarantee a PASS.
- Parts I–III are the particle-scale face of five predictions from one input. Part IV is the a₀ derivation, tested against 175 real SPARC galaxies — that story, with the independent re-run and the falsification registry, lives at the a₀ research hub.