So, how would we go about reviewing audio processors?

[dfuller]

Say, an equalizer or compressor. Obviously these are devices that aren't intended to be linear (figure a DAC, ADC, preamp, power amp are all supposed to not add or subtract anything from the signal going through them; an equalizer is designed to do so). What do we measure? EQ curves? Phase shift? Distortion with or without the processing engaged? With compression, what do we measure - attack and release times? transfer function "knee"?

Interested to hear your thoughts.

 

[sergeauckland]

Conventional frequency response and distortion figures are still relevant, together with noise depending on whether the processor includes a noise gate and at some defined gain.
Then, indeed, compression ratios, attack and release times and overshoot are all important. A comment on whether the controls have excessive interaction and how accurate is any metering would be part of a review.

A full mastering processor will largely be evaluated subjectively as a creative tool, but nevertheless, there's scope for technical measurements, especially if they highlight any problems.
S

 

[dfuller]

Conventional frequency response and distortion figures are still relevant, together with noise depending on whether the processor includes a noise gate and at some defined gain.

Yes, though I think it would be useful to not pass judgement w/r/t whether that's good or not.

Then, indeed, compression ratios, attack and release times and overshoot are all important. A comment on whether the controls have excessive interaction and how accurate is any metering would be part of a review.

This, to me, is crucial stuff.

A full mastering processor will largely be evaluated subjectively as a creative tool, but nevertheless, there's scope for technical measurements, especially if they highlight any problems.

Right. I'm just thinking about how I'd go about doing analysis later on down the line, for reasons of understanding why certain equipment has the effect it does. Think "Neves make everything round and fat", what makes it that way?

 

[sergeauckland]

Yes, though I think it would be useful to not pass judgement w/r/t whether that's good or not.
This, to me, is crucial stuff.

Right. I'm just thinking about how I'd go about doing analysis later on down the line, for reasons of understanding why certain equipment has the effect it does. Think "Neves make everything round and fat", what makes it that way?

Having watched in some awe at creative experts manipulating sounds on a processor, I think operating a mastering or on-air processor is akin to playing an instrument. It needs a deep knowledge of what each control does, and more importantly, how the various controls interact to create a particular sound. Another analogy is a skilled cook that can take lots of disparate ingredients and create something that tastes wonderful. Analysing the individual ingredients / musical notes / controls is fine, insofar as it allows an understanding of what each one is / does, but it doesn't tell you anything about what the subjective results will be.

In the words of the much loved and lamented Eric Morecambe when told he was playing all the wrong notes, replied he's playing all the right notes, but not necessarily in the right order.

S

 

[pozz]

Right. I'm just thinking about how I'd go about doing analysis later on down the line, for reasons of understanding why certain equipment has the effect it does. Think "Neves make everything round and fat", what makes it that way?

This will go a long way. Some people hold the idea that the audio path in certain consoles just produces better results. I'm sure that's not the case. Same goes for hardware compressors and limiters.

We could develop a few typical settings for each and test the output, which I predict will be more similar than expected in kind, but the level/gain controls will never have exactly the same action. This is probably the cause of the majority if comments: roughly the same settings set by ear or by interface, but not exactly the same, causing differences in output and perceived loudness.

 

[dfuller]

Having watched in some awe at creative experts manipulating sounds on a processor, I think operating a mastering or on-air processor is akin to playing an instrument. It needs a deep knowledge of what each control does, and more importantly, how the various controls interact to create a particular sound. Another analogy is a skilled cook that can take lots of disparate ingredients and create something that tastes wonderful. Analysing the individual ingredients / musical notes / controls is fine, insofar as it allows an understanding of what each one is / does, but it doesn't tell you anything about what the subjective results will be.

I work in this field, and yes - mixing on a large format console is very much akin to playing an instrument. Not super different at all.