records/reference.json

{
 "body_sha256": "3756dc2aa873ff5ca6f8732bbb0c547fb3aa4db6f629b91b2528e8d05b19399f",
 "caveats": {
  "direct": "These packs' walks were not retained: a 100,000-walker walk of 2,677 saves is 8 GB per fixture. The direct numbers are therefore the ones the walk that produced each pack recorded, in this family's own records/build.json, measured by the estimator cross_engine_parity.floors owns. Nothing is re-walked to obtain a number a record already holds.",
  "disimpy-cylinder": "**HELD on [dmipy-sim#488](https://github.com/dmrai-lab/dmipy-sim/issues/488), and this gate cannot see why.** The gate below compares ONE declared measurement -- the last of the fixture's own protocol, its highest b -- and there this pack agrees with the MISST reference to 1.8e-5, so it passes. The evidence for the hold is a VECTOR: over the protocol's 100 measurements the worst is 1.97e-3, 7.27 sigma, with 31 of them outside their own 3-sigma band. A `ReferenceQuantity` is a scalar, so no scalar comparison this protocol can express will fail this pack; [dmipy-sim#493](https://github.com/dmrai-lab/dmipy-sim/issues/493) is the item that would let the gate carry a vector and therefore see it.\n\nWhat #488 is about: since #483 fixed the pickle's face winding the mesh walks correctly, and yet at 100,000 walkers neither the mesh NOR the analytic cylinder of the same radius reproduces this MISST reference to the Monte-Carlo floor (1.97e-3 and 1.31e-3, 7.27 and 5.66 sigma). The measured faceting term -- mesh against the analytic cylinder at the same N, seed and waveform, MISST not involved -- is 8.00e-4 and sits inside its own band, so the two geometries agree with each other and both disagree with MISST. #488 is whether that is MISST's own truncation or ours. Until it says, read this pack as a faithful record of a walk whose reference is in question, and see `records/pre-protocol/gate.json` for the vector comparison that held it.",
  "reduction": "A pack's signal is the MODULUS of the weighted ensemble mean, and both references sum cosines. Comparing one convention against the other read 1.2e-5 at b = 1925 s/mm^2 and 3.1e-3 at 13190 and filed #484 against the engine; the per-walker phases are identical across the routes. Every comparison on this card reduces both sides the same way.",
  "reproduces": "The `reproduces-` comparison is DEGENERATE for this family and its 1e-9 sigma should be read as such. Our direct number is measured on the pack's own decoded positions, because the walk that produced the pack was not retained, so `reproduces-` and `served-equals-decoded` are two readings of one channel and differ only by the reduction. The comparison that carries information here is `published-`, against the other engine's released array."
 },
 "crossref": {
  "agency": "crossref",
  "container": "Frontiers in Neuroinformatics",
  "doi": "10.3389/fninf.2020.00008",
  "resolved": "2026-09-27T02:19:20Z",
  "title": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results",
  "type": "journal-article"
 },
 "description": "Two other simulators' own released substrates, walked here on their surfaces at their acquisitions: the cross-ENGINE check that this engine is not being compared with itself.",
 "doi": "10.3389/fninf.2020.00008",
 "geometry_derivation": "the geometry is the released one when no free parameter marked 'ours' changes it; none of this family's parameters does",
 "geometry_parameters_ours": [],
 "inputs_sha256": "5c9c5b258ce7702f6f8f59b5396d25f74626d29e9d09a813731330192f648a9b",
 "licence_note": "the packs are the source's licence: LGPL-2.1 for the MC/DC fixtures, MIT for the Disimpy one; neither source's bytes are redistributed",
 "parameters": [
  {
   "changes_geometry": false,
   "how": "read by io.mcdc.read_scheme, which puts TE, every Delta, every delta and every pad on a sample and refuses what the format leaves ambiguous",
   "name": "TE / delta / Delta / G",
   "unit": "-",
   "value": "ActiveAxG140_PM.scheme",
   "where": "their scheme file",
   "whose": "theirs"
  },
  {
   "changes_geometry": false,
   "how": "stated; 2e-9 for the Disimpy fixture, from their own test",
   "name": "diffusivity",
   "unit": "m^2/s",
   "value": 6e-10,
   "where": "their .conf",
   "whose": "theirs"
  },
  {
   "changes_geometry": false,
   "how": "read cyclically as MC/DC reads it, which is why the spec's seeding rule is `explicit` and not a uniform draw over the lumen",
   "name": "seed positions",
   "unit": "-",
   "value": "their released initial-walker list",
   "where": "*_ini_points.txt",
   "whose": "theirs"
  },
  {
   "changes_geometry": false,
   "how": "theirs is a fixed 5,000 steps over TE, recorded for comparison; a converged walk's signal does not depend on either",
   "name": "sub-step rule",
   "unit": "-",
   "value": "the engine's own",
   "where": "physics.resolve_sub_steps",
   "whose": "ours"
  },
  {
   "changes_geometry": false,
   "how": "none: both references' walls are purely reflecting",
   "name": "surface relaxivity",
   "unit": "m/s",
   "value": 0.0,
   "where": "this family's walk",
   "whose": "ours"
  }
 ],
 "previous_sha256": "805552c9e9801b936118bb0c3609c07d4788634657c0f3a029128f081aba5d85",
 "published_kind": "data",
 "quantities": [
  {
   "direct": {
    "grid": {
     "index": -1,
     "measurement": "the last measurement of the fixture's own released protocol (its highest b)",
     "n_t": 2677,
     "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)"
    },
    "n_walkers": 100000,
    "se": 0.005437714343851,
    "se_derivation": "the delta method on the modulus of the ensemble mean over the DECODED walk's own walkers: each walker's phasor projected on the direction the modulus is taken along, sd / sqrt(N) / |S|. No fold, no split half -- the one-permutation split half this family used before swung the worst measurement between 2.09 and 3.51 sigma over 200 seeds, and the family-wise check failed for 49.5 % of them",
    "se_kind": "delta_method",
    "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)",
    "source": "examples/substrate_bank/build_parity_fixtures.py (direct_at): the pack's positions channel decoded and read with the engine's own gradient phase, no band contraction in the reduction. The walk that produced the pack was not retained (8 GB per fixture) and is not repeated",
    "unit": "-",
    "value": 0.35845581349963046
   },
   "documents": {
    "10.3389/fninf.2020.00008": {
     "agency": "crossref",
     "container": "Frontiers in Neuroinformatics",
     "doi": "10.3389/fninf.2020.00008",
     "resolved": "2026-09-27T02:19:20Z",
     "title": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results",
     "type": "journal-article"
    }
   },
   "name": "signal_at_top_b",
   "published": {
    "data_sha256": "c0f8fa05ab65ae97569fead36343c4e5b3e4fd6877abbb42414714ebdcada8f1",
    "data_url": "https://github.com/jonhrafe/Robust-Monte-Carlo-Simulations",
    "document": "10.3389/fninf.2020.00008",
    "locator": "the last measurement of ActiveAxG140_PM.scheme in uAxon_d_1.0_amp_0.2_wL_32.0_DWI.bfloat, an unnormalised sum over walkers divided by the count its b = 0 rows state",
    "printed_in": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results (Rafael-Patino et al. 2020), whose released Experiments-raw-signals archive the array is read from",
    "uncertainty": 0.008735906142570006,
    "uncertainty_is": "their released file is ONE Monte-Carlo realisation of 50,000 walkers and states no spread, so its standard error is the per-walker spread measured on OUR walkers at THEIR count (cross_engine_parity.floors), which is what makes this comparison failable",
    "unit": "-",
    "value": 0.361631796875,
    "verbatim": ""
   },
   "substrate": "mcdc-0.2-32.0"
  },
  {
   "direct": {
    "grid": {
     "index": -1,
     "measurement": "the last measurement of the fixture's own released protocol (its highest b)",
     "n_t": 2677,
     "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)"
    },
    "n_walkers": 100000,
    "se": 0.004114417659177954,
    "se_derivation": "the delta method on the modulus of the ensemble mean over the DECODED walk's own walkers: each walker's phasor projected on the direction the modulus is taken along, sd / sqrt(N) / |S|. No fold, no split half -- the one-permutation split half this family used before swung the worst measurement between 2.09 and 3.51 sigma over 200 seeds, and the family-wise check failed for 49.5 % of them",
    "se_kind": "delta_method",
    "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)",
    "source": "examples/substrate_bank/build_parity_fixtures.py (direct_at): the pack's positions channel decoded and read with the engine's own gradient phase, no band contraction in the reduction. The walk that produced the pack was not retained (8 GB per fixture) and is not repeated",
    "unit": "-",
    "value": 0.4414517416701021
   },
   "documents": {
    "10.3389/fninf.2020.00008": {
     "agency": "crossref",
     "container": "Frontiers in Neuroinformatics",
     "doi": "10.3389/fninf.2020.00008",
     "resolved": "2026-09-27T02:19:20Z",
     "title": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results",
     "type": "journal-article"
    }
   },
   "name": "signal_at_top_b",
   "published": {
    "data_sha256": "254d70cbc0a20cfefd3e2be23d1c3b7f402d5bb9b9451e11c087c8c4c9a751bb",
    "data_url": "https://github.com/jonhrafe/Robust-Monte-Carlo-Simulations",
    "document": "10.3389/fninf.2020.00008",
    "locator": "the last measurement of ActiveAxG140_PM.scheme in uAxon_d_1.0_amp_1.0_wL_12.0_DWI.bfloat, an unnormalised sum over walkers divided by the count its b = 0 rows state",
    "printed_in": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results (Rafael-Patino et al. 2020), whose released Experiments-raw-signals archive the array is read from",
    "uncertainty": 0.007233017845072881,
    "uncertainty_is": "their released file is ONE Monte-Carlo realisation of 50,000 walkers and states no spread, so its standard error is the per-walker spread measured on OUR walkers at THEIR count (cross_engine_parity.floors), which is what makes this comparison failable",
    "unit": "-",
    "value": 0.4439753125,
    "verbatim": ""
   },
   "substrate": "mcdc-1.0-12.0"
  },
  {
   "direct": {
    "grid": {
     "index": -1,
     "measurement": "the last measurement of the fixture's own released protocol (its highest b)",
     "n_t": 2677,
     "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)"
    },
    "n_walkers": 100000,
    "se": 0.0006726927177553796,
    "se_derivation": "the delta method on the modulus of the ensemble mean over the DECODED walk's own walkers: each walker's phasor projected on the direction the modulus is taken along, sd / sqrt(N) / |S|. No fold, no split half -- the one-permutation split half this family used before swung the worst measurement between 2.09 and 3.51 sigma over 200 seeds, and the family-wise check failed for 49.5 % of them",
    "se_kind": "delta_method",
    "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)",
    "source": "examples/substrate_bank/build_parity_fixtures.py (direct_at): the pack's positions channel decoded and read with the engine's own gradient phase, no band contraction in the reduction. The walk that produced the pack was not retained (8 GB per fixture) and is not repeated",
    "unit": "-",
    "value": 0.864485547792327
   },
   "documents": {
    "10.3389/fninf.2020.00008": {
     "agency": "crossref",
     "container": "Frontiers in Neuroinformatics",
     "doi": "10.3389/fninf.2020.00008",
     "resolved": "2026-09-27T02:19:21Z",
     "title": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results",
     "type": "journal-article"
    }
   },
   "name": "signal_at_top_b",
   "published": {
    "data_sha256": "9b60772159b6319228bec95950284d9df5e152f310efdf86ade5421b77e7cae6",
    "data_url": "https://github.com/jonhrafe/Robust-Monte-Carlo-Simulations",
    "document": "10.3389/fninf.2020.00008",
    "locator": "the last measurement of ActiveAxG140_PM.scheme in uAxon_d_1.0_amp_2.6_wL_4.0_DWI.bfloat, an unnormalised sum over walkers divided by the count its b = 0 rows state",
    "printed_in": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results (Rafael-Patino et al. 2020), whose released Experiments-raw-signals archive the array is read from",
    "uncertainty": 0.0031435529533804596,
    "uncertainty_is": "their released file is ONE Monte-Carlo realisation of 50,000 walkers and states no spread, so its standard error is the per-walker spread measured on OUR walkers at THEIR count (cross_engine_parity.floors), which is what makes this comparison failable",
    "unit": "-",
    "value": 0.8642859375,
    "verbatim": ""
   },
   "substrate": "mcdc-2.6-4.0"
  },
  {
   "direct": {
    "grid": {
     "index": -1,
     "measurement": "the last measurement of the fixture's own released protocol (its highest b)",
     "n_t": 1401,
     "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)"
    },
    "n_walkers": 100000,
    "se": 0.0005647347397067077,
    "se_derivation": "the delta method on the modulus of the ensemble mean over the DECODED walk's own walkers: each walker's phasor projected on the direction the modulus is taken along, sd / sqrt(N) / |S|. No fold, no split half -- the one-permutation split half this family used before swung the worst measurement between 2.09 and 3.51 sigma over 200 seeds, and the family-wise check failed for 49.5 % of them",
    "se_kind": "delta_method",
    "solver": "the ensemble modulus |mean exp(i phi)| over the pack's walkers, the reduction ReplayPack.replay returns (cross_engine_parity.ensemble_modulus)",
    "source": "examples/substrate_bank/build_parity_fixtures.py (direct_at): the pack's positions channel decoded and read with the engine's own gradient phase, no band contraction in the reduction. The walk that produced the pack was not retained (8 GB per fixture) and is not repeated",
    "unit": "-",
    "value": 0.8762886919187907
   },
   "documents": {
    "10.21105/joss.02527": {
     "agency": "crossref",
     "container": "Journal of Open Source Software",
     "doi": "10.21105/joss.02527",
     "resolved": "2026-09-27T02:19:21Z",
     "title": "Disimpy: A massively parallel Monte Carlo simulator for generating diffusion-weighted MRI data in Python",
     "type": "journal-article"
    }
   },
   "name": "signal_at_top_b",
   "published": {
    "data_sha256": "f8af9348eef1941cd9ec930ccf154ae004313e326bcf9a5a5b79b92f03b7f041",
    "data_url": "https://github.com/kerkelae/disimpy",
    "document": "10.21105/joss.02527",
    "locator": "the last of the 100 normalised values of misst_cylinder_signal_smalldelta_30ms_bigdelta_40ms_radius_5um.txt",
    "printed_in": "Disimpy: A massively parallel Monte Carlo simulator for generating diffusion-weighted MRI data in Python (Kerkelae et al. 2020), whose tests/ fixtures the array is read from",
    "uncertainty": 0.0,
    "uncertainty_is": "none: MISST is an exact eigenfunction solution, not a Monte-Carlo estimate. What the file does limit is its own resolution -- it is distributed to seven decimal places -- and that quantisation is a term of the design's tolerance rather than an uncertainty of the reference",
    "unit": "-",
    "value": 0.8763042,
    "verbatim": ""
   },
   "substrate": "disimpy-cylinder"
  }
 ],
 "same_released_geometry": true,
 "sample": "the same objects: MC/DC's own released undulating-axon surfaces and Disimpy's own released cylinder mesh, each walked here on THEIR surface at THEIR acquisition and diffusivity, the MC/DC ones seeded from their own released initial-walker list",
 "sample_relation": "the same object",
 "source_note": "MC/DC (Rafael-Patino et al. 2020, Front. Neuroinform. 14:8, LGPL-2.1) and Disimpy (Kerkelae et al. 2020, JOSS 5(52):2527, MIT), whose MISST reference is Drobnjak, Zhang, Hall and Alexander's exact eigenfunction solution.",
 "stage": "reference",
 "title": "Robust Monte-Carlo Simulations in Diffusion-MRI: Effect of the Substrate Complexity and Parameter Choice on the Reproducibility of Results",
 "title_matches": true
}