Blind test
Hardware or CA-72?
In each example, one clip is a Behringer MODEL D and the other is CA-72 0.3.0 playing the same notes with matched settings. Which is A and which is B is drawn at random for every example, and drawn again each time you start. Identities are revealed when you finish.
Recorded hardware noise is added to every CA-72 clip in this test, so the noise floor gives nothing away.
About the clips
Both clips of a pair have the same length and are played at the levels they were matched to, through the same path, centred. Without the added noise, the CA-72’s silent background would give it away at the start and end of a clip; the noise is real background noise from the hardware recording setup, at the level of each hardware clip’s own quiet floor. Both clips also get a very low-level random dither, different for every attempt and far below that noise floor, so neither copy is byte-identical to a published file. Nothing else is changed: no EQ, filtering, compression or timing edits beyond one fixed gain and start point per source. The three noise examples are short excerpts of a longer held sound, so both of their clips end with the same 5 ms fade.
The bass and lead examples are their six-note phrases; the separate held notes recorded for measurement aren’t used.
When you finish, your answers are saved anonymously, without any name, cookie or address, to see which examples are hardest to tell apart. See the privacy policy.