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

... When I think of “warmth,” I tend to think of a sound that is “intimate” or “immediate.” In other words, I think of a high ratio of direct to indirect sound, which would typically correspond to listening to a live event from a relatively close position. But that may be an idiosyncratic association on my part. Like you, Hifibri seems to have the OPPOSITE association.... This makes me wonder whether you, or Hifibri, would say that studio recordings (i.e. those with few or no ambient cues) can’t sound warm?

Others can undoubtedly speak to that more knowledgeably than I can, in part because most of my listening is to classical music that has been recorded in halls. But I would say that depending on the instrument, and on what is being played and how it is being played, warmth can in many cases certainly be captured and reproduced via up close miking in a studio. While at the same time it can often be better captured in a hall via more distant miking.

The key to that apparent paradox, it seems to me, is that warmth is a multi-faceted concept, as this thread makes clear. “Woodiness,” “body,” etc. are for example certainly important aspects of warmth, and their successful reproduction involves capturing the fine detail and harmonic balance of the instrument. That in turn can be expected to be compromised as distance increases. On the other hand, massed strings, to cite another example, can sound overly bright at close distances. As was noted above, high frequencies will be attenuated more rapidly as a function of increasing distance than low and mid frequencies. Also, reflected energy will be subject to frequency response contouring as a result of both the greater distance it travels before reaching the listener or the mic’s, and the acoustic properties of the reflecting surface. Finally, and perhaps most significantly, summation of reflected energy with directly captured sound will result in comb filtering effects.

So it would seem that optimal reproduction of recordings that are produced in a concert hall, at least, would involve drawing a balance between preservation of detail and harmonic balance on the one hand, and either reproducing hall ambience and distance effects correctly, if present in the recording, or enhancing them, if they are not present but should be. There would seem to be no easy answers ....

Best regards,
-- Al

02-06-11: Almarg
“Woodiness,” “body,” etc. are for example certainly important aspects of warmth, and their successful reproduction involves capturing the fine detail and harmonic balance of the instrument. That in turn can be expected to be compromised as distance increases. On the other hand, massed strings, to cite another example, can sound overly bright at close distances. As was noted above, high frequencies will be attenuated more rapidly as a function of increasing distance than low and mid frequencies.

This is a good point, Al, and it makes perfect sense of the idea that, in at least some cases, warmth increases as the distance of the listener to the live event increases. That would explain the correlation you observed between recordings with ambient cues and those with warmth. On the other hand, it also seems to suggest that, for smaller scale performances, warmth might DECREASE as the distance of the listener to the live event increases. So maybe there is reason to believe that a common characteristic of warm recordings is that the distance of the microphone to the live event is “proportional” to the scale of the event being recorded. That idea makes a lot of sense to me.

Also, reflected energy will be subject to frequency response contouring as a result of both the greater distance it travels before reaching the listener or the mic’s, and the acoustic properties of the reflecting surface. Finally, and perhaps most significantly, summation of reflected energy with directly captured sound will result in comb filtering effects.

The first of these points is more or less what I meant in the OP by item #5, which I called “room resonance,” but would probably more accurately be called simply “room acoustics.” Having said that, it never occurred to me that comb filtering might be a detriment to a system’s warmth. That strikes me as a plausible idea, since the destructive interference of comb filtering effects could conceivably result in a kind of "harmonic thinning,” which might very well be experienced as a lack of warmth. An insightful observation, Al.

Bryon

Once again, more proof of the need for a better audiophile vocabulary; one which, if it is to have any real meaning, needs concensus. A good place to start is, ironically, outside the audiophile world; in real world vocabulary. How often do we say something like :"Joe Toob is a warm person", or "John Fett is not a very warm person, he is cold". When we say these things, are we referring to anything having to do do with frequency response? Probably not. We are probably referring to wether a person is amiable or not. I think the term warm, in the audio context, is being misused, and has more to do with a components ability to draw you in, and let you get involved with the music. Frequency response certainly plays a role, but I think more impotant in determinng a component's "amiability" is the texture of the sound it makes. This can be independent of perceived brightness, or darkness. I have heard components that sound way too dark, and thick through the midrange, but are also very dry sounding. Conversely, I have heard some that I would consider way too bright sounding, but are also very liquid sounding. Texture. Micro dynamics ability is the other key factor. That is where a component's ability to convey a performer's expressive qualities lies; the warmth of the performance.

Music that approaches the sound of a live performance is, by defintion, warm. The sound of live music runs the gamut as far as wether it sounds shrill, bright, dark, etc. It is the human element that makes it warm, or not. What gives music that hard to describe texture, and dimensionality is the human body attached to it.

Consensus in audio ???

You're right. What was I thinking?!?