What is “warmth” and how do you get it?

Hi Bryon,

... The issue of whether instrument timbre is reducible to harmonic content. I am completely out of my depth here, so will remain agnostic.

Let me make sure it's clear that implicit in my previous post (the one with the lengthy quotes from the "What Is A Sound Spectrum?" writeup) is that I now recognize that timbre is only partially reducible to harmonic content, and other factors also affect timbre significantly. An example of those other factors is the low level broadband component of the flute spectrum that the article discussed. I still suspect, though, that for most instruments harmonic content is typically the most significant determinant of timbre.

IF instrument timbre is EVEN PARTIALLY reducible to harmonic content, then it reveals a flaw in the “additive approach” to playback.... What is ADDED during playback will not be identical to what was SUBTRACTED during the recording process. The reason is because, while the harmonic contents of live events are almost infinitely VARIABLE, the harmonic “additions” of playback equipment are largely CONSTANT, being persistent artifacts of a circuit's more or less fixed parameters.

I'm not sure that the variable/constant distinction is particularly meaningful, even within the limited context of harmonic distortion. The harmonic components resulting from distortion in the electronics will continuously vary as a function of the varying spectral components of the music, and to some extent with the overall signal level. Harmonic distortion introduced by the speakers will vary with signal level as well as with frequency, and will often overshadow the distortion components introduced by the electronics. Throw in intermodulation distortion, transient intermodulation distortion, the complexity of the music, the complexity of the sounds of the instruments themselves, hall and room effects, the vagaries of the recording process, etc., etc., and the “approximation” you spoke of would seem to be all that can be hoped for, regardless of the approach that is chosen.

The upshot of all this is that, in light of my desire for additional warmth, maybe I chose the wrong amp.

On the other hand, as was said earlier, with your speakers an amp having a high output impedance (such as the Prima Luna and probably most other tube amps) will de-emphasize lower frequencies relative to upper mid and high frequencies. Obviously, that would be in the wrong direction with respect to adding warmth.

So I'm not sure where that leaves us, but those are some thoughts.

Best regards,
-- Al

I'm not sure that the variable/constant distinction is particularly meaningful, even within the limited context of harmonic distortion. The harmonic components resulting from distortion in the electronics will continuously vary as a function of the varying spectral components of the music, and to some extent with the overall signal level.

I agree with this, Al, insofar as I recognize that the harmonic content introduced by equipment varies as a function of the spectral content and level of the input signal. So perhaps "variable vs. constant" is not quite the right distinction.

The distinction I was trying to capture is between the harmonic content of the live event vs. the harmonic content of the reproduced event. To the extent that the recording process diminishes the harmonic content of the recording, the additive approach to playback may do something to replace what is missing. But it cannot, so far as I am aware, replace the EXACT missing harmonic content, since the harmonics introduced by equipment are, in essence, artifacts.

The fact that equipment-induced harmonic artifacts vary as a function of the spectral content and level of the input signal does make them "variable," perhaps even as variable as the harmonics of the live event. Nevertheless, equipment-induced harmonic artifacts are not SIMULACRA of the harmonics of the live event. That is the flaw in the additive approach to playback that I was trying to point out.

Bryon

Looking back on this thread, I have another thought...

There has been a lot of discussion about how increasing warmth in a system might involve ADDING something to the signal, like low order harmonics. It occurs to me that increasing warmth might also involve SUBTRACTING something from the signal.

Maybe the most obvious example of how subtracting something from the signal might increase warmth is subtracting treble, which amounts to the same thing as adding midrange/bass. A less obvious example is contained in Al's second post on 2/6, namely subtracting destructive acoustical effects like comb filtering. One final example that I have personally experienced is subtracting (or at least reducing) jitter in digital playback, which to my ears increases the perception of "warmth."

Bryon

“and how do you get it”… I just went to “warm” by changing the footers under my CDP from Stillpoints to HRS Nimbus Couplers. I really liked what the Stillpoints did, but at times my system sounded a little “thin”. Without losing any of the good that the Stillpoints brought, the HRS Nimbus Couplers added what can best described as “weight”.

Some times, it could be as easy as swapping the source I/c. Not to alter freq response but take that electronic tinze/noise surrounding the notes. I have three sets of cables at each source, same corresponding brand for each level of sound i desire- neutral, slightly warm and warmer. You will be surprised how 'listenable' most music would become... true changing cables every once in a while is a chore and one set of cables is always your primary choice. It is no different than some folks trying different tonearms, cart, diif DAC sampling, diff sources...

One characteristic that warmer sound gets you is the image size ( overall- fundamental and harmonics and air) of instruments become life size which I personally consider primary requisite.