5 ms·

That transfer function argument is within an ace of saying all amplifiers sound alike, and it's shaky. OTOH we have ABX data that cannot disprove the propositio

That transfer function argument is within an ace of saying all amplifiers sound alike, and it's shaky. OTOH we have ABX data that cannot disprove the proposition. However, only one transfer function admits of that possibility even according to Carver (well, indirectly via Stereophile) [1]). Audiophilia is wonderful because it is full of set-piece arguments that promote the drinking of lots of Scotch. [1] https://www.stereophile.com/content/carver-silver-seven-t-monoblock-power-amplifier https://www.stereophile.com/content/carver-silver-seven-t-mo...

analog31 6y ago
In my view, nobody's saying that all amplifiers sound alike. Indeed, you can design an amplifier to have a pronounced sonic signature, which is common in the guitar amplifier world. A more precise statement might be that two amplifiers with the same transfer function should sound alike, but this needs additional clarification, namely a specified margin of error. Finding out that margin of error has been a pursuit of acoustics research, and is the original motivation for the ABX test. For instance, I read that when Sony developed their MiniDisc recorder, the data compression scheme was subjected to extensive ABX testing.
- TylerE 6y ago
  It's not even that simple... for instance some circuits are more susceptible to thermal drift, or dirty power.
  - analog31 6y ago
    Indeed, and this can be a problem with "audiophile" gear that is sometimes built with exceptional craftsmanship but not necessarily designed with the best electrical performance in a mainstream sense. For instance if someone believes they must experience a feedback-free Class-A amplifier, well, yes they're going to have some thermal drift and poor power supply noise rejection. that can be hard to replicate. ;-) Good mainstream designs take component tolerances into account, and it's worth remembering that a 1% variation is a tiny fraction of a dB.
- lb1lf 6y ago
  -I don't think 'all amplifiers sound alike' is a widespread opinion, but rather (in engineering circles, at any rate) that any competently designed amplifier, run within its limits, sound like any other competently designed amplifier run within its limits. Audio amplifiers are pretty much a solved problem, to the extent that in order to create a 'house sound' or whatever the preferred audiophile term is nowadays, the manufacturer needs to introduce distortion to differentiate their product from the others. Sigh. (The above obviously only applies to amplifiers intended for faithful reproduction; guitar amps and the like is a whole different beast - where distortion serves a valid artistic/creative purpose!)

analog31 6y ago

In my view, nobody's saying that all amplifiers sound alike. Indeed, you can design an amplifier to have a pronounced sonic signature, which is common in the guitar amplifier world. A more precise statement might be that two amplifiers with the same transfer function should sound alike, but this needs additional clarification, namely a specified margin of error. Finding out that margin of error has been a pursuit of acoustics research, and is the original motivation for the ABX test. For instance, I read that when Sony developed their MiniDisc recorder, the data compression scheme was subjected to extensive ABX testing.

TylerE 6y ago
It's not even that simple... for instance some circuits are more susceptible to thermal drift, or dirty power.
- analog31 6y ago
  Indeed, and this can be a problem with "audiophile" gear that is sometimes built with exceptional craftsmanship but not necessarily designed with the best electrical performance in a mainstream sense. For instance if someone believes they must experience a feedback-free Class-A amplifier, well, yes they're going to have some thermal drift and poor power supply noise rejection. that can be hard to replicate. ;-) Good mainstream designs take component tolerances into account, and it's worth remembering that a 1% variation is a tiny fraction of a dB.
lb1lf 6y ago
-I don't think 'all amplifiers sound alike' is a widespread opinion, but rather (in engineering circles, at any rate) that any competently designed amplifier, run within its limits, sound like any other competently designed amplifier run within its limits. Audio amplifiers are pretty much a solved problem, to the extent that in order to create a 'house sound' or whatever the preferred audiophile term is nowadays, the manufacturer needs to introduce distortion to differentiate their product from the others. Sigh. (The above obviously only applies to amplifiers intended for faithful reproduction; guitar amps and the like is a whole different beast - where distortion serves a valid artistic/creative purpose!)