What's worse: a resistor, a capacitor or an inductor?

[serengetiplains]

I'd be interested hearing views on what people consider the hierarchy of electronics building blocks least detrimental to the sound of an amplifying circuit. Does a resistor sound worse, for instance, than an inductor, all other things being equal? I know the "all other things being equal" part is to a degree unrealistic, as replacing a resistor with an inductor in a given design will require other circuit changes, complicating a direct comparison. My question might therefore be too abstract. Perhaps those adept at building RIAA compensation networks might have direct experience??

Cheers!

[tiroth]

I would say that what you are asking is, "to what degree do real versions of these components approach the ideal version?"

My reply to that would be first resistors, then capacitors, and finally inductors. At extremes of current, voltage, and frequency, though this may no longer hold true.

[markp]

I think a carbon comp resistor is the worst as far as noise goes. A good metal film is the way to go.

[serengetiplains]

Originally posted by tiroth
I would say that what you are asking is, "to what degree do real versions of these components approach the ideal version?"

That's one great angle of approach, and probably a version of the question I was asking, being: given the ability to swap one component type for another, which has the least detrimental impact on the sound?

[dmfraser]

These parts are not really interchangeable. As a designer, I use whichever is necessary to do the job as each as distinctly diferent functions.

However, in a design, we will try to use capacitors instead of inductors if we can because capacitors are cheaper. However, I use a passive inductor/capacitor network for the EMI filters on my inputs as it is more effective than an RC network.

The proper question I suggest you ask is which is the best resistor, capacitor or inductor for each place you need a resistor, capacitor or inductor.

For example, carbon composition resistors have the most noise and the least accuracy but they have the lowest self inductance and some say they were a contributing factor to the warmth of tube amplifiers of yesteryear.

The best part within a category depends what you are doing.

[Sch3mat1c]

I view it in terms of coupling... you have RC, LC and direct coupling, among a few other possibilities... direct is the most blameless but RC is best if you have to have a DC shift to keep things stable (as in tube amps). Anything else, like interstage transformers, are impossible to get a good loop of NFB around if you have just a few of them.

Tim