Why 711 and B&K 5128 measurements can't be compared

Headphone measurements are made by playing a sweep into an artificial ear. The two common standards are the IEC 60318-4 coupler (universally nicknamed '711') and the newer Brüel & Kjær 5128 head-and-torso simulator. They model the human ear canal and eardrum impedance differently, so the same physical headphone produces visibly different curves on each, especially above roughly 1 kHz, where ear-canal resonances dominate.

The 711 coupler has a well-known standing-wave resonance around 8 kHz that exaggerates or masks treble features, and its simplified canal makes measurements above about 10 kHz unreliable. The 5128 models eardrum impedance across frequency far more realistically, which shifts both the bass coupling and the entire treble signature relative to a 711 curve.

This is why overlaying a 711 measurement of one IEM against a 5128 measurement of another tells you nothing about how the two compare. The rig difference can be larger than the difference between the products. It also means a target curve is rig-specific: 'the Harman target' as usually plotted is defined for 711-class rigs, and 5128 measurements need their own reference.

qfiAudio's graph tool enforces this: each chart is locked to a single rig, the rig is labeled on every chart, and switching rigs clears the overlay rather than silently mixing incompatible data.