Be careful which listening tests you are quoting
@lowendtheory · · 1mo
Whenever the question of how capture methods compare comes up, the same listening test gets linked — a large one, over a thousand participants and a hundred thousand ratings, with a clear winner.
It is a real test and the sample size is genuinely large. It was also run by a party with a stake in the outcome, and nobody has independently replicated it. Both things are true at once.
I am not saying the result is wrong. I am saying "the only numbers in circulation come from an interested party" is the honest summary of where the evidence stands, and that is worth knowing before repeating a number as settled. If the technology is as good as it seems, an independent replication only helps it.
2 replies
@roomtone · 1mo
Thank you for phrasing it that way. The number gets repeated constantly with no mention of who ran it. I would add that a large listening test is genuinely hard to run well, and running one at all is more than most parties do. "Interested party, unreplicated" is a caveat, not an accusation.@quietamp · 1mo
Bench comparison is the one I trust, for what little it is worth: amp in the room, capture on the desk, same cab, same mic, switch between them. It is not a controlled test and I am not claiming it is. But I have done it on maybe thirty amps now and the gap has closed a lot in a year.
Sign in to reply. Reading needs no account.