Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback
E. Brad Meyer and David R. Moran, Journal of the Audio Engineering Society, vol. 55 no. 9, September 2007. The spoke’s first peer-reviewed test of a character claim, and the answer is negative.
The experiment
Double-blind listening over the course of a year. Subjects heard high-resolution playback directly, or the same signal passed through a 16-bit/44.1 kHz “bottleneck” — a CD-standard analog-to-digital and digital-to-analog round trip inserted into the chain and switchable out of it. Playback ran on professional monitors and on high-end systems including electrostatic speakers. Listeners were recording engineers, university students and audiophiles.
The finding, in the authors’ words: the CD-quality A/D/A loop was undetectable at normal-to-loud listening levels, by any of the subjects, on any of the playback systems.
Note what the design controls for. The comparison is not two recordings or two machines; it is the same signal with and without a conversion stage. Everything else is held constant, which is what the spoke’s existing sources never manage.
Why this lands on the analog question
analog-vs-digital frames the divide as continuous circuitry against sampled computation, and records the character claim — warmer, more organic — as the analog side’s standing argument. Every source behind that claim in this wiki is a listening impression: a maker describing an instrument, a reviewer describing a pedal (analog-vs-digital-pedals), an enthusiast describing a rig.
Meyer and Moran do not test the analog claim in general. They test the narrowest version of it: that adding CD-standard sampling to a signal path is audible at all. If the audible cost of digitizing is zero under controlled conditions, then the character difference people report between an analog and a digital box is not coming from the conversion — it is coming from the circuit, the filter design, the nonlinearity, the specific implementation. Which is closer to what analog-vs-digital‘s own through-line already argued, now with a measurement under it rather than an inference.
What it does not settle, and the paper that disagrees
The result is about transparency of a conversion stage, not about whether an analog filter and a digital one sound the same. A discrete ladder filter’s distortion signature is a real, measurable thing; nothing here says otherwise.
And the AES literature is not unanimous. digital-filter-audibility (Stuart and Craven) ran double-blind tests on the filters used in A/D conversion and mastering, and concluded that listeners are sensitive to those alterations. Two peer-reviewed papers, the same society, the same method family, opposite verdicts on adjacent questions. Both are recorded; see synthesis for the tension.
Related
analog-vs-digital · digital-filter-audibility · analog-vs-digital-pedals · synthesizer · synthesis