LEVELAPI2

Compliance results

MEASURED AGAINST THE
REFERENCE FILES THE
STANDARDS BODIES PUBLISH

Every row below is one measurement of one published reference file, graded against the value and the tolerance printed in that standard's own document. The document and clause are named beside each result, and the measured figure is shown rather than a tick, so any claim here can be checked against the same file.

Reference sets in this run: EBU Tech 3341, EBU Tech 3342, Report ITU-R BS.2217-2, ARIB TR-B32, plus the shipped preset assertions and the in-house checks.

161Pass
0Fail
9Measured, not graded
3Unsupported
173Rows

WHAT THIS PAGE IS, AND WHAT IT IS NOT

This is not a certification. Nothing here has been assessed, audited or approved by the EBU, the ITU, ATSC, ARIB or Free TV Australia. These are measurements made in-house, using reference files those bodies publish for exactly this purpose. No accreditation is claimed or implied.

The reference audio is freely available from the standards bodies, every expected value is cited to the clause it comes from, and the tolerance is printed beside the measurement rather than folded into a verdict. An engineer who disagrees can download the same files and run the same test.

The verification claim, stated exactly

All ten shipped presets are verified to configure exactly what their own standard's document specifies. Three are additionally verified against that body's own published reference audio: EBU R 128, ITU-R BS.1770 and ARIB TR-B32.

The other seven have no published reference set to measure against: only some standards bodies release conformance audio. The measurement path is the same either way. SetPresetStandard() is a handful of setter calls with different constants, and there is no per-standard measurement code anywhere, so the three verified standards exercise the engine all ten of them use.

Where the gaps are

How to read a class

ClassGraded?Meaning
PASS/FAILyesThe standard's own document publishes a value and a tolerance. The citation is in the row.
CONSISTENCYyes, in-houseA check that can genuinely fail, graded against an in-house criterion rather than a standard's. Every such row says so.
CHARACTERISEDnoThe library produces a number and no standard defines a reference for it. Measured and reported, never graded.
UNSUPPORTEDnoThe reference file exceeds what the library accepts.

No row is marked pass against a criterion invented here. A measured number with no published reference is shown as measured, and an in-house check is labelled as one.

THE BUILD THESE NUMBERS CAME FROM

PropertyValue
Library revision202609080
Header revision202609080
BinaryLevelAPI2-x64.dll
Configurationx64 Release, code-signed
Reference material136 published files from the EBU, ITU and ARIB sets
Platforms runWindows and Linux. macOS not run

Windows and Linux measure the same thing. Same revision, and the same value in every row to three decimals, across two compilers, two C++ standard libraries and two operating systems. The two reports come out byte for byte the same apart from the binary's file name.

Build configuration is not a variable in these results either. The table was produced once from a debug build and again from the shipped release binary, and the two agree to three decimals in every row.

RESULTS, ROW BY ROW

Grouped by reference set. Open a group for its rows, the tolerance applied to each and the document each expected value comes from.

EBU Tech 334175 rows · 75 of 75 graded pass

EBU Tech 3341 (2023), Table 1 - the EBU Mode minimum requirement set.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
3341-1-IPASS/FAILLUFS-22.991-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 1
3341-1-LUPASS/FAILLU0.0090.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 1
3341-1-MPASS/FAILLUFS_Max_M-22.991-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 1
3341-1-SPASS/FAILLUFS_Max_S-22.991-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 1
3341-2-IPASS/FAILLUFS-32.998-33.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 2
3341-2-LUPASS/FAILLU-9.998-10.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 2
3341-2-MPASS/FAILLUFS_Max_M-32.997-33.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 2
3341-2-SPASS/FAILLUFS_Max_S-32.997-33.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 2
3341-3-IPASS/FAILLUFS-23.053-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 3
3341-3-LUPASS/FAILLU-0.0530.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 3
3341-4-IPASS/FAILLUFS-23.053-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 4
3341-4-LUPASS/FAILLU-0.0530.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 4
3341-5-IPASS/FAILLUFS-23.017-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 5
3341-5-LUPASS/FAILLU-0.0170.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 5
3341-6a-IPASS/FAILLUFS-23.055-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 6
3341-6a-LUPASS/FAILLU-0.0550.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 6
3341-6b-IPASS/FAILLUFS-23.055-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 6
3341-6b-LUPASS/FAILLU-0.0550.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 6
3341-7-IPASS/FAILLUFS-23.028-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 7
3341-7-LUPASS/FAILLU-0.0280.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 7
3341-8-IPASS/FAILLUFS-23.040-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 8
3341-8-LUPASS/FAILLU-0.0400.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 8
3341-9-SPASS/FAILLUFS_Max_S-23.024-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 9
3341-10-1PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-2PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-3PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-4PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-5PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-6PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-7PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-8PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-9PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-10PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-11PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-12PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-13PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-14PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-15PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-16PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-17PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-18PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-19PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-10-20PASS/FAILLUFS_Max_S-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 10
3341-11PASS/FAILLUFS_Max_S20.00020.000–PASSEBU Tech 3341 (2023) Table 1, case 11
3341-12-MPASS/FAILLUFS_Max_M-22.998-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 12
3341-13-1PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-2PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-3PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-4PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-5PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-6PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-7PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-8PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-9PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-10PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-11PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-12PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-13PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-14PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-15PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-16PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-17PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-18PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-19PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-13-20PASS/FAILLUFS_Max_M-23.031-23.000+0.10/-0.10PASSEBU Tech 3341 (2023) Table 1, case 13
3341-14PASS/FAILLUFS_Max_M20.00020.000–PASSEBU Tech 3341 (2023) Table 1, case 14
3341-15-TPPASS/FAILTruePeak-6.000-6.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 15
3341-16-TPPASS/FAILTruePeak-6.000-6.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 16
3341-17-TPPASS/FAILTruePeak-6.000-6.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 17
3341-18-TPPASS/FAILTruePeak-6.000-6.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 18
3341-19-TPPASS/FAILTruePeak3.0103.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 19
3341-20-TPPASS/FAILTruePeak-0.1310.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 20
3341-21-TPPASS/FAILTruePeak-0.1310.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 21
3341-22-TPPASS/FAILTruePeak-0.1310.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 22
3341-23-TPPASS/FAILTruePeak-0.1310.000+0.20/-0.40PASSEBU Tech 3341 (2023) Table 1, case 23
  • 3341-5-I (ebu/seq-3341-5-16bit-v02.wav) – the relative gate case: -26/-20/-26 dBFS integrates to -23 only when gated
  • 3341-6a-I (ebu/seq-3341-6-5channels-16bit.wav) – 5.0, WAVE_FORMAT_PCM - the same test as the 5.1 file below
  • 3341-6b-I (ebu/seq-3341-6-6channels-WAVEEX-16bit.wav) – 5.1 with silent LFE, WAVE_FORMAT_EXTENSIBLE
  • 3341-7-I (ebu/seq-3341-7_seq-3342-5-24bit.wav) – authentic programme 1, narrow loudness range
  • 3341-8-I (ebu/seq-3341-2011-8_seq-3342-6-24bit-v02.wav) – authentic programme 2, wide loudness range
  • 3341-9-S (ebu/seq-3341-9-24bit.wav) – S must be constant at -23.0 after 3 s
  • 3341-10-1 (ebu/seq-3341-10-1-24bit.wav) – file-based dual of case 11
  • 3341-10-2 (ebu/seq-3341-10-2-24bit.wav) – file-based dual of case 11
  • 3341-10-3 (ebu/seq-3341-10-3-24bit.wav) – file-based dual of case 11
  • 3341-10-4 (ebu/seq-3341-10-4-24bit.wav) – file-based dual of case 11
  • 3341-10-5 (ebu/seq-3341-10-5-24bit.wav) – file-based dual of case 11
  • 3341-10-6 (ebu/seq-3341-10-6-24bit.wav) – file-based dual of case 11
  • 3341-10-7 (ebu/seq-3341-10-7-24bit.wav) – file-based dual of case 11
  • 3341-10-8 (ebu/seq-3341-10-8-24bit.wav) – file-based dual of case 11
  • 3341-10-9 (ebu/seq-3341-10-9-24bit.wav) – file-based dual of case 11
  • 3341-10-10 (ebu/seq-3341-10-10-24bit.wav) – file-based dual of case 11
  • 3341-10-11 (ebu/seq-3341-10-11-24bit.wav) – file-based dual of case 11
  • 3341-10-12 (ebu/seq-3341-10-12-24bit.wav) – file-based dual of case 11
  • 3341-10-13 (ebu/seq-3341-10-13-24bit.wav) – file-based dual of case 11
  • 3341-10-14 (ebu/seq-3341-10-14-24bit.wav) – file-based dual of case 11
  • 3341-10-15 (ebu/seq-3341-10-15-24bit.wav) – file-based dual of case 11
  • 3341-10-16 (ebu/seq-3341-10-16-24bit.wav) – file-based dual of case 11
  • 3341-10-17 (ebu/seq-3341-10-17-24bit.wav) – file-based dual of case 11
  • 3341-10-18 (ebu/seq-3341-10-18-24bit.wav) – file-based dual of case 11
  • 3341-10-19 (ebu/seq-3341-10-19-24bit.wav) – file-based dual of case 11
  • 3341-10-20 (ebu/seq-3341-10-20-24bit.wav) – file-based dual of case 11
  • 3341-11 (ebu/seq-3341-11-24bit.wav) – all 20 successive readings within +/-0.1 of -38.0 .. -19.0 – live-meter dual of case 10: 20 successive Max S values, -38.0 rising to -19.0 in 1 LU steps
  • 3341-12-M (ebu/seq-3341-12-24bit.wav) – M must be constant at -23.0 after 1 s
  • 3341-13-1 (ebu/seq-3341-13-1-24bit.wav) – file-based dual of case 14
  • 3341-13-2 (ebu/seq-3341-13-2-24bit.wav) – file-based dual of case 14
  • 3341-13-3 (ebu/seq-3341-13-3-24bit.wav) – file-based dual of case 14
  • 3341-13-4 (ebu/seq-3341-13-4-24bit.wav) – file-based dual of case 14
  • 3341-13-5 (ebu/seq-3341-13-5-24bit.wav) – file-based dual of case 14
  • 3341-13-6 (ebu/seq-3341-13-6-24bit.wav) – file-based dual of case 14
  • 3341-13-7 (ebu/seq-3341-13-7-24bit.wav) – file-based dual of case 14
  • 3341-13-8 (ebu/seq-3341-13-8-24bit.wav) – file-based dual of case 14
  • 3341-13-9 (ebu/seq-3341-13-9-24bit.wav) – file-based dual of case 14
  • 3341-13-10 (ebu/seq-3341-13-10-24bit.wav) – file-based dual of case 14
  • 3341-13-11 (ebu/seq-3341-13-11-24bit.wav) – file-based dual of case 14
  • 3341-13-12 (ebu/seq-3341-13-12-24bit.wav) – file-based dual of case 14
  • 3341-13-13 (ebu/seq-3341-13-13-24bit.wav) – file-based dual of case 14
  • 3341-13-14 (ebu/seq-3341-13-14-24bit.wav) – file-based dual of case 14
  • 3341-13-15 (ebu/seq-3341-13-15-24bit.wav) – file-based dual of case 14
  • 3341-13-16 (ebu/seq-3341-13-16-24bit.wav) – file-based dual of case 14
  • 3341-13-17 (ebu/seq-3341-13-17-24bit.wav) – file-based dual of case 14
  • 3341-13-18 (ebu/seq-3341-13-18-24bit.wav) – file-based dual of case 14
  • 3341-13-19 (ebu/seq-3341-13-19-24bit.wav) – file-based dual of case 14
  • 3341-13-20 (ebu/seq-3341-13-20-24bit.wav) – file-based dual of case 14
  • 3341-14 (ebu/seq-3341-14-24bit.wav) – all 20 successive readings within +/-0.1 of -38.0 .. -19.0 – live-meter dual of case 13: 20 successive Max M values, -38.0 rising to -19.0 in 1 LU steps
EBU Tech 33426 rows · 6 of 6 graded pass

EBU Tech 3342 - the loudness range minimum requirement set.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
3342-1PASS/FAILLRA10.00110.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 1
3342-2PASS/FAILLRA4.9995.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 2
3342-3PASS/FAILLRA19.99520.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 3
3342-4PASS/FAILLRA14.99915.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 4
3342-5PASS/FAILLRA5.0065.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 5
3342-6PASS/FAILLRA14.99315.000+1.00/-1.00PASSEBU Tech 3342 Table 1, case 6
Report ITU-R BS.2217-247 rows · 45 of 45 graded pass · 2 not graded

Report ITU-R BS.2217-2 (10/2016) - the ITU compliance material for BS.1770.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
2217-relgatePASS/FAILLUFS-10.067-10.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-absgatePASS/FAILLUFS-69.490-69.500+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-sweepPASS/FAILLUFS-18.029-18.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-25HzPASS/FAILLUFS-22.998-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-100HzPASS/FAILLUFS-23.032-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-500HzPASS/FAILLUFS-23.036-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-1000HzPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-2000HzPASS/FAILLUFS-23.042-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-10000HzPASS/FAILLUFS-23.038-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-summingPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkLeftPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkRightPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkCentrePASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkLsPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkRsPASS/FAILLUFS-23.031-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-23LKFS-checkLFEPASS/FAILLUFS-999.000-999.000+899.00/-0.00PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-25HzPASS/FAILLUFS-23.998-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-100HzPASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-500HzPASS/FAILLUFS-24.036-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-1000HzPASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-2000HzPASS/FAILLUFS-24.043-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-10000HzPASS/FAILLUFS-24.038-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-summingPASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkLeftPASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkRightPASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkCentrePASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkLsPASS/FAILLUFS-24.033-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkRsPASS/FAILLUFS-24.033-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24LKFS-checkLFEPASS/FAILLUFS-999.000-999.000+899.00/-0.00PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinCntr-23PASS/FAILLUFS-23.024-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinLR-23PASS/FAILLUFS-23.032-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinLRC-23PASS/FAILLUFS-23.029-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-StereoVinLR-23PASS/FAILLUFS-23.023-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-MonoVoiceMusic-23PASS/FAILLUFS-23.034-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinCntr-24PASS/FAILLUFS-24.020-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinLR-24PASS/FAILLUFS-24.031-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-6chVinLRC-24PASS/FAILLUFS-24.019-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-StereoVinLR-24PASS/FAILLUFS-24.020-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-conf-MonoVoiceMusic-24PASS/FAILLUFS-24.037-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-8ch-23PASS/FAILLUFS-23.038-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-10ch-23PASS/FAILLUFS-23.039-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-12ch-23PASS/FAILLUFS-23.039-23.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24ch-23UNSUPPORTEDLUFS–-23.000+0.10/-0.10SKIPPEDReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-8ch-24PASS/FAILLUFS-24.038-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-10ch-24PASS/FAILLUFS-24.039-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-12ch-24PASS/FAILLUFS-24.039-24.000+0.10/-0.10PASSReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
2217-24ch-24UNSUPPORTEDLUFS–-24.000+0.10/-0.10SKIPPEDReport ITU-R BS.2217-2, compliance table (tolerance +/-0.1 LKFS)
  • 2217-relgate (ITU/1770-2_Comp_RelGateTest.wav) – relative gate. ~-8.0 LKFS means an -8 LU gate is in force
  • 2217-absgate (ITU/1770-2_Comp_AbsGateTest.wav) – absolute gate. -73 LKFS means no absolute gate at all
  • 2217-sweep (ITU/1770-2_Comp_18LKFS_FrequencySweep.wav) – K-weighting sweep: 1 dB of filter error in a 1/3 octave band shows as ~0.5 LU on a 3 s meter
  • 2217-23LKFS-summing (ITU/1770-2_Comp_23LKFS_SummingTest.wav) – channel gains and summing
  • 2217-23LKFS-checkLeft (ITU/1770-2_Comp_23LKFS_ChannelCheckLeft.wav) – per-channel gain check
  • 2217-23LKFS-checkRight (ITU/1770-2_Comp_23LKFS_ChannelCheckRight.wav) – per-channel gain check
  • 2217-23LKFS-checkCentre (ITU/1770-2_Comp_23LKFS_ChannelCheckCentre.wav) – per-channel gain check
  • 2217-23LKFS-checkLs (ITU/1770-2_Comp_23LKFS_ChannelCheckLs.wav) – per-channel gain check
  • 2217-23LKFS-checkRs (ITU/1770-2_Comp_23LKFS_ChannelCheckRs.wav) – per-channel gain check
  • 2217-23LKFS-checkLFE (ITU/1770-2_Comp_23LKFS_ChannelCheckLFE.wav) – read -999.000; a compliant meter excludes LFE entirely, so anything at or below -100.0 LKFS counts. That window is set here, not by the Report. – LFE is excluded from BS.1770, so a compliant meter reads -infinity. The pass window (anything at or below -100 LKFS) is set here, not by the Report, which says 'the lowest resolvable value, or -infinity' - and a float API cannot return -inf
  • 2217-24LKFS-summing (ITU/1770-2_Comp_24LKFS_SummingTest.wav) – channel gains and summing
  • 2217-24LKFS-checkLeft (ITU/1770-2_Comp_24LKFS_ChannelCheckLeft.wav) – per-channel gain check
  • 2217-24LKFS-checkRight (ITU/1770-2_Comp_24LKFS_ChannelCheckRight.wav) – per-channel gain check
  • 2217-24LKFS-checkCentre (ITU/1770-2_Comp_24LKFS_ChannelCheckCentre.wav) – per-channel gain check
  • 2217-24LKFS-checkLs (ITU/1770-2_Comp_24LKFS_ChannelCheckLs.wav) – per-channel gain check
  • 2217-24LKFS-checkRs (ITU/1770-2_Comp_24LKFS_ChannelCheckRs.wav) – per-channel gain check
  • 2217-24LKFS-checkLFE (ITU/1770-2_Comp_24LKFS_ChannelCheckLFE.wav) – read -999.000; a compliant meter excludes LFE entirely, so anything at or below -100.0 LKFS counts. That window is set here, not by the Report. – LFE is excluded from BS.1770, so a compliant meter reads -infinity. The pass window (anything at or below -100 LKFS) is set here, not by the Report, which says 'the lowest resolvable value, or -infinity' - and a float API cannot return -inf
  • 2217-conf-6chVinCntr-23 (ITU/1770-2 Conf 6ch VinCntr-23LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-6chVinLR-23 (ITU/1770-2 Conf 6ch VinL+R-23LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-6chVinLRC-23 (ITU/1770-2 Conf 6ch VinL-R-C-23LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-StereoVinLR-23 (ITU/1770-2 Conf Stereo VinL+R-23LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-MonoVoiceMusic-23 (ITU/1770-2 Conf Mono Voice+Music-23LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-6chVinCntr-24 (ITU/1770-2 Conf 6ch VinCntr-24LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-6chVinLR-24 (ITU/1770-2 Conf 6ch VinL+R-24LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-6chVinLRC-24 (ITU/1770-2 Conf 6ch VinL-R-C-24LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-StereoVinLR-24 (ITU/1770-2 Conf Stereo VinL+R-24LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-conf-MonoVoiceMusic-24 (ITU/1770-2 Conf Mono Voice+Music-24LKFS.wav) – authentic programme. An underscore-named byte-identical copy also exists on disk; it is an alias and is deliberately not counted twice
  • 2217-8ch-23 (ITU/1770Conf-23LKFS-8channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-10ch-23 (ITU/1770Conf-23LKFS-10channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-12ch-23 (ITU/1770Conf-23LKFS-12channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-24ch-23 (ITU/1770Conf-23LKFS-24channel.wav) – 24 channels at 48000 Hz - the library's max_channels is 16 – 24 channels; the library's max_channels is 16
  • 2217-8ch-24 (ITU/1770Conf-24LKFS-8channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-10ch-24 (ITU/1770Conf-24LKFS-10channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-12ch-24 (ITU/1770Conf-24LKFS-12channel.wav) – channel gain check. Weights are set per channel from the Report's own channel-ordering table - SetSurroundOrder must NOT be used above 5.1
  • 2217-24ch-24 (ITU/1770Conf-24LKFS-24channel.wav) – 24 channels at 48000 Hz - the library's max_channels is 16 – 24 channels; the library's max_channels is 16
ARIB TR-B3210 rows · 9 of 9 graded pass · 1 not graded

ARIB TR-B32 v1.5-E1, Table 6-1 - the Japanese broadcast set.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
gate-diagnosisPASS/FAILLUFS-20.257-20.200+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Tables 6-1 and 6-8
trb32-A01PASS/FAILLUFS-24.032-24.000+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A02PASS/FAILLUFS-24.038-24.000+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A03PASS/FAILLUFS-24.035-24.000+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A03cPASS/FAILLUFS-24.034-24.000+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A03hUNSUPPORTEDLUFS–-24.000+0.10/-0.10SKIPPEDARIB TR-B32 v1.5-E1 Table 6-1
trb32-A04PASS/FAILLUFS-69.782-69.700+0.20/-0.20PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A05PASS/FAILLUFS-17.031-17.000+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A06PASS/FAILLUFS-20.257-20.200+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
trb32-A07PASS/FAILLUFS-22.462-22.400+0.10/-0.10PASSARIB TR-B32 v1.5-E1 Table 6-1
  • gate-diagnosis (TRB32/TRB32_A05_16.wav + A06) – A05 read -17.031 (expect -17.0), A06 read -20.257 (expect -20.2): relative gate -10 LU - correct for TR-B32, BS.1770-2 and later – A05 and A06 judged together. Neither file alone can tell -10 LU from -8 LU or from no gate at all.
  • trb32-A01 (TRB32/TRB32_A01_16.wav) – K-weighting accuracy at 1000 Hz
  • trb32-A02 (TRB32/TRB32_A02_16.wav) – K-weighting across frequency
  • trb32-A03 (TRB32/TRB32_A03_16.wav) – channel weighting, 5.1
  • trb32-A03c (TRB32/TRB32_A03c_16.wav) – channel weighting, 7.1. TR-B32 Table 6-4 gives its own channel order; if this fails, suspect the channel order before the measurement
  • trb32-A03h (TRB32/TRB32_A03h_16.wav) – 24 channels at 48000 Hz - the library's max_channels is 16 – 22.2 - 24 channels; the library's max_channels is 16
  • trb32-A04 (TRB32/TRB32_A04_L10-H15_60-10-60.wav) – absolute gating. Tolerance is +/-0.2 by TR-B32 6.1: a 16-bit low-level signal, true value -69.7414, so interval and rounding error add
  • trb32-A05 (TRB32/TRB32_A05_16.wav) – relative gating - JUDGE WITH A06. Alone this value cannot tell -10 LU from -8 LU
  • trb32-A06 (TRB32/TRB32_A06_16.wav) – relative gating - JUDGE WITH A05. Alone this value cannot tell -10 LU from no gate
  • trb32-A07 (TRB32/TRB32_A07_16.wav) – 400 ms gating block with 75% overlap. The +/-0.1 is INFERRED from TR-B32 6.1, not printed beside this row. -27.0 = no gating block, -20.0 = no overlap, -21.8 = 50% overlap
Standards presets10 rows · 10 of 10 graded pass

Every shipped preset is set, then read back and compared to the value its own document specifies.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
preset-100PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-101PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-102PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-103PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-200PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-201PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-300PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-400PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-500PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
preset-501PASS/FAILconfiguration0.0000.000–PASSdocs/notes/loudness-standards.md, every row cited to its own document
  • preset-100 – EBU R 128: target -23.0 LUFS, relative gate -10.0, dialogue gate off, ceiling -1.0 dBTP
  • preset-101 – EBU R 128 (2010 edition): target -23.0 LUFS, relative gate -8.0, dialogue gate off, ceiling -1.0 dBTP
  • preset-102 – EBU R 128 s2 - streaming: target -18.0 LUFS, relative gate -10.0, dialogue gate off, ceiling -1.0 dBTP
  • preset-103 – EBU R 128 s4 - cinematic, anchor-based: target -23.0 LUFS, relative gate -10.0, dialogue gate on, ceiling -1.0 dBTP
  • preset-200 – ATSC A/85 - long form: target -24.0 LKFS, relative gate 0.0, dialogue gate on, ceiling -2.0 dBTP
  • preset-201 – ATSC A/85 - short form: target -24.0 LKFS, relative gate -10.0, dialogue gate off, ceiling -2.0 dBTP
  • preset-300 – ARIB TR-B32: target -24.0 LKFS, relative gate -10.0, dialogue gate off, ceiling -1.0 dBTP
  • preset-400 – Free TV Australia OP-59: target -24.0 LKFS, relative gate -10.0, dialogue gate off, ceiling -2.0 dBTP – Free TV OP-59
  • preset-500 – ITU-R BS.1770-1 (ungated): target -23.0 LUFS, relative gate 0.0, dialogue gate off, ceiling -1.0 dBTP
  • preset-501 – ITU-R BS.1770-4 (gated): target -23.0 LUFS, relative gate -10.0, dialogue gate off, ceiling -1.0 dBTP
Consistency13 rows · 13 of 13 graded pass

Checks against in-house criteria rather than a standard's. Graded, because each one can genuinely fail.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
pair-1dB-1CONSISTENCYLUFS delta1.0031.000+0.05/-0.05PASSNot from a document: the two files are the same audio 1 dB apart, verified by direct sample comparison
pair-LRA-1CONSISTENCYLRA delta0.0000.000+0.10/-0.10PASSNot from a document: same audio 1 dB apart
pair-1dB-2CONSISTENCYLUFS delta0.9971.000+0.05/-0.05PASSNot from a document: the two files are the same audio 1 dB apart, verified by direct sample comparison
pair-LRA-2CONSISTENCYLRA delta0.0010.000+0.10/-0.10PASSNot from a document: same audio 1 dB apart
pair-1dB-3CONSISTENCYLUFS delta0.9961.000+0.05/-0.05PASSNot from a document: the two files are the same audio 1 dB apart, verified by direct sample comparison
pair-LRA-3CONSISTENCYLRA delta0.0000.000+0.10/-0.10PASSNot from a document: same audio 1 dB apart
pair-1dB-4CONSISTENCYLUFS delta0.9991.000+0.05/-0.05PASSNot from a document: the two files are the same audio 1 dB apart, verified by direct sample comparison
pair-LRA-4CONSISTENCYLRA delta-0.0000.000+0.10/-0.10PASSNot from a document: same audio 1 dB apart
pair-1dB-5CONSISTENCYLUFS delta0.9901.000+0.05/-0.05PASSNot from a document: the two files are the same audio 1 dB apart, verified by direct sample comparison
pair-LRA-5CONSISTENCYLRA delta0.0000.000+0.10/-0.10PASSNot from a document: same audio 1 dB apart
repeatabilityCONSISTENCYLUFS delta0.0000.000–PASSNot from a document
file-vs-blocksCONSISTENCYLUFS delta0.0000.000+0.01/-0.01PASSNot from a document
dialogue-channel-selectCONSISTENCYLUFS_Dialog delta35.578––PASSNot from a document
  • pair-1dB-1 (ITU/1770-2 Conf Mono Voice+Music-23LKFS.wav) – 1770-2 Conf Mono Voice+Music: -23 file read -23.034, -24 file read -24.037, difference 1.003 – In-house criterion. The integrated loudness must differ by exactly 1.000 between the -23 and -24 members of the pair.
  • pair-LRA-1 (ITU/1770-2 Conf Mono Voice+Music-23LKFS.wav) – 1770-2 Conf Mono Voice+Music: LRA 15.866 against 15.866 – In-house criterion. LRA is level-independent, so the pair must agree. A small deviation is legitimate only if quiet passages sit near the -70 LKFS absolute gate, which these programmes should not.
  • pair-1dB-2 (ITU/1770-2 Conf Stereo VinL+R-23LKFS.wav) – 1770-2 Conf Stereo VinL+R: -23 file read -23.023, -24 file read -24.020, difference 0.997 – In-house criterion. The integrated loudness must differ by exactly 1.000 between the -23 and -24 members of the pair.
  • pair-LRA-2 (ITU/1770-2 Conf Stereo VinL+R-23LKFS.wav) – 1770-2 Conf Stereo VinL+R: LRA 14.599 against 14.598 – In-house criterion. LRA is level-independent, so the pair must agree. A small deviation is legitimate only if quiet passages sit near the -70 LKFS absolute gate, which these programmes should not.
  • pair-1dB-3 (ITU/1770-2 Conf 6ch VinCntr-23LKFS.wav) – 1770-2 Conf 6ch VinCntr: -23 file read -23.024, -24 file read -24.020, difference 0.996 – In-house criterion. The integrated loudness must differ by exactly 1.000 between the -23 and -24 members of the pair.
  • pair-LRA-3 (ITU/1770-2 Conf 6ch VinCntr-23LKFS.wav) – 1770-2 Conf 6ch VinCntr: LRA 10.838 against 10.838 – In-house criterion. LRA is level-independent, so the pair must agree. A small deviation is legitimate only if quiet passages sit near the -70 LKFS absolute gate, which these programmes should not.
  • pair-1dB-4 (ITU/1770-2 Conf 6ch VinL+R-23LKFS.wav) – 1770-2 Conf 6ch VinL+R: -23 file read -23.032, -24 file read -24.031, difference 0.999 – In-house criterion. The integrated loudness must differ by exactly 1.000 between the -23 and -24 members of the pair.
  • pair-LRA-4 (ITU/1770-2 Conf 6ch VinL+R-23LKFS.wav) – 1770-2 Conf 6ch VinL+R: LRA 10.490 against 10.490 – In-house criterion. LRA is level-independent, so the pair must agree. A small deviation is legitimate only if quiet passages sit near the -70 LKFS absolute gate, which these programmes should not.
  • pair-1dB-5 (ITU/1770-2 Conf 6ch VinL-R-C-23LKFS.wav) – 1770-2 Conf 6ch VinL-R-C: -23 file read -23.029, -24 file read -24.019, difference 0.990 – In-house criterion. The integrated loudness must differ by exactly 1.000 between the -23 and -24 members of the pair.
  • pair-LRA-5 (ITU/1770-2 Conf 6ch VinL-R-C-23LKFS.wav) – 1770-2 Conf 6ch VinL-R-C: LRA 13.095 against 13.095 – In-house criterion. LRA is level-independent, so the pair must agree. A small deviation is legitimate only if quiet passages sit near the -70 LKFS absolute gate, which these programmes should not.
  • repeatability (ebu/seq-3341-7_seq-3342-5-24bit.wav) – two runs read -23.028101 and -23.028101 – In-house criterion, and the strictest one here: two runs over the same file must agree exactly.
  • file-vs-blocks (ebu/seq-3341-7_seq-3342-5-24bit.wav) – AnalyseFile read -23.0281, 1024-frame blocks read -23.0281 – In-house criterion. AnalyseFile and a block-fed run from sample zero measure the same audio, so they must agree.
  • dialogue-channel-select (ITU/1770-2 Conf 6ch VinL+R-24LKFS.wav) – L/R/C selected read -29.807, centre alone read -65.385, difference 35.578 – In-house criterion. Speech is in L and R on this file, so selecting the centre channel alone must change the dialogue reading. If it does not, SetDialogChannel is inert.
Adjust round trip3 rows · 3 of 3 graded pass

Normalise to a target, then re-measure the output. An in-house criterion.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
roundtrip-1CONSISTENCYLUFS-23.000-23.000+0.10/-0.10PASSTarget from docs/notes/loudness-standards.md; the round trip itself is not a published test
roundtrip-2CONSISTENCYLUFS-24.000-24.000+0.10/-0.10PASSTarget from docs/notes/loudness-standards.md; the round trip itself is not a published test
roundtrip-3CONSISTENCYLUFS-24.000-24.000+0.10/-0.10PASSTarget from docs/notes/loudness-standards.md; the round trip itself is not a published test
  • roundtrip-1 (ebu/seq-3341-7_seq-3342-5-24bit.wav) – normalised to -23.0 LUFS and re-measured at -23.000 LUFS (wrote roundtrip-1.wav) – In-house criterion. Normalise to the standard's target, then measure the output: it must land on the target.
  • roundtrip-2 (ebu/seq-3341-7_seq-3342-5-24bit.wav) – normalised to -24.0 LUFS and re-measured at -24.000 LUFS (wrote roundtrip-2.wav) – In-house criterion. Normalise to the standard's target, then measure the output: it must land on the target.
  • roundtrip-3 (ebu/seq-3341-7_seq-3342-5-24bit.wav) – normalised to -24.0 LUFS and re-measured at -24.000 LUFS (wrote roundtrip-3.wav) – In-house criterion. Normalise to the standard's target, then measure the output: it must land on the target.
ATSC A/85 long form8 rows · 8 not graded

Dialogue-gated readings on real programme material. Measured and shown, never graded.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
atsc-longform-1CHARACTERISEDLUFS_Dialog-29.912––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-2CHARACTERISEDLUFS_Dialog-29.808––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-3CHARACTERISEDLUFS_Dialog-31.888––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-4CHARACTERISEDLUFS_Dialog-28.119––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-5CHARACTERISEDLUFS_Dialog-27.129––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-6CHARACTERISEDLUFS_Dialog-24.681––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
atsc-longform-7CHARACTERISEDLUFS_Dialog-23.585––REPORTEDNo published reference: no standards body publishes an expected dialogue-gated value for any test file
dialogue-placementCHARACTERISEDLUFS_Dialog spread2.080––REPORTEDNot from a document
  • atsc-longform-1 (ITU/1770-2 Conf 6ch VinCntr-24LKFS.wav) – dialogue-gated -29.912 LUFS, full mix -24.617 LUFS, dialogue present 49.8% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-2 (ITU/1770-2 Conf 6ch VinL+R-24LKFS.wav) – dialogue-gated -29.807 LUFS, full mix -24.592 LUFS, dialogue present 56.2% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-3 (ITU/1770-2 Conf 6ch VinL-R-C-24LKFS.wav) – dialogue-gated -31.888 LUFS, full mix -24.901 LUFS, dialogue present 58.5% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-4 (ITU/1770-2 Conf Stereo VinL+R-24LKFS.wav) – dialogue-gated -28.119 LUFS, full mix -24.852 LUFS, dialogue present 39.5% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-5 (ITU/1770-2 Conf Mono Voice+Music-24LKFS.wav) – dialogue-gated -27.129 LUFS, full mix -24.976 LUFS, dialogue present 41.1% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-6 (ebu/seq-3341-7_seq-3342-5-24bit.wav) – dialogue-gated -24.681 LUFS, full mix -23.028 LUFS, dialogue present 15.9% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • atsc-longform-7 (ebu/seq-3341-2011-8_seq-3342-6-24bit-v02.wav) – dialogue-gated -23.585 LUFS, full mix -24.020 LUFS, dialogue present 60.6% of the programme – Dialogue-gated loudness under LEVELAPI2_STANDARD_ATSC_A85_LONGFORM. Reported, deliberately not graded.
  • dialogue-placement (ITU/1770-2 Conf 6ch Vin* -24LKFS) – centre-only -29.912, L+R -29.807, L-R-C -31.888: spread 2.080 LU – Reported, never graded. The three files are the ITU's same programme with speech in the centre, in L+R, and in all three front channels. Nothing published anywhere says how far a dialogue-gated reading may move between them, so the spread is measured and shown rather than judged.
RealtimeAdjuster1 rows · 1 not graded

The live loudness stabiliser. Measured and shown, never graded.

CaseClassMeterMeasuredExpectedToleranceVerdictSource
realtime-convergenceCHARACTERISEDLUFS-22.811––REPORTEDNo published reference for a live loudness stabiliser
  • realtime-convergence (ebu/seq-3341-2011-8_seq-3342-6-24bit-v02.wav) – target -23.0 LUFS; ended at -22.811 LUFS with -2.500 dB of gain applied; warm-up window 0.0 s; output SAMPLE peak -1.000 dBFS (sample peak, NOT true peak) – Reported, never graded. Convergence towards the target and the output sample peak, both measured over one pass of an authentic programme.