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

One thing I’ve noticed about the various characteristics that go by the term ‘warmth’ in the context of playback is that most or all of them seem like ADDITIONS to the signal. For example, ADDITIONAL low order harmonics, ADDITIONAL lower midrange/upper bass, ADDITIONAL ambience provided by the listening room, and so on.

Strictly speaking, any additions to the signal (other than gain) are deviations from accuracy. For that reason, I think many audiophiles, myself included, are tempted to eschew them. But lately I've been having second thoughts about that attitude. I’m starting to wonder about the relative merits of the following two characteristics:

1. Accuracy to the recording
2. Accuracy to the recorded event

The relative merits of accuracy-to-the-recording vs. accuracy-to-the-recorded event has periodically occurred to me ever since, on the neutrality thread, Al wrote this:

12-02-09: Almarg
A perfectly accurate system…would be one that resolves everything that is fed into it, and reproduces what it resolves with complete neutrality. Another way of saying that is perhaps that what is reproduced at the listener's ears corresponds precisely to what is fed into the system.

Which does not necessarily make that system optimal in terms of transparency. Since the source material will essentially always deviate to some degree and in some manner from being precisely accurate relative to the original event, then it can be expected that some deviation from accuracy in the system may in many cases be complementary to the inaccuracies of the recording (at least subjectively), resulting in a greater transparency into the music than a more precisely accurate system would provide.

Which does not mean that the goals of accuracy and transparency are necessarily inconsistent or in conflict. It simply means, as I see it, that the correlation between them, although substantial, is less than perfect.

Al makes his point about accuracy in terms of neutrality vs. transparency, in keeping with the nomenclature of that thread, but it is essentially the same distinction as accuracy-to-the-recording vs. accuracy-to-the-recorded-event, or Recording Accuracy vs. Event Accuracy, for short.

I agree with Al that Recording Accuracy correlates with Event Accuracy, but not perfectly so. In other words, I now believe that efforts to maximize Recording Accuracy sometimes come at the expense of Event Accuracy, which is a viewpoint that, I suspect, more experienced audiophiles tend to adopt, but has taken me some time to appreciate. A turning point for me was an observation that Albert Porter made in an old digital vs. analog thread, which I read only recently:

09-12-08: Albertporter
The digital (or analog) master tape is not the issue here, the CD format is.

If any of you could hear a master digital tape (or hard drive) and compare that to CD or LP, you would realize how much we've been screwed. The problem with digital is when that great master is "moved" for public distribution…

Moving that master digital signal from one place to another and from one sample rate to another does it so much harm it cannot be repaired. Then to make matters worse, our only choice is an outdated format that's too low a sample rate to replicate what was on the master…

With CD, you get a severely downsampled format that's only a shadow of what could be if the format had evolved this last 25 years.

This observation resonated with me, as I have had the experience of recording, editing, and mixing with high quality professional equipment, to create a master recording I was proud of. I then watched - dismayed - as my master recording was compressed, downrez'd, and finally, transferred to its delivery format. Even on a very high quality playback system, the delivery format's recording was a shadow of its former self. Albert Porter’s observations about CD recordings undergoing this process of diminishment as a matter of routine procedure highlights the many respects in which the recordings available to consumers deviate dramatically from their master recordings, to say nothing of how the master recordings themselves deviate from the recorded events. Taken together, both deviations create a gulf between the live event and its consumer playback, a gulf that some audiophiles try to fill with ADDITIVE measures. And that brings me back to the point of this post...

It now seems to me that the use of ADDITIVE measures can be a means of filling, to whatever extent possible, the gulf between the live event and the (in many cases) extremely diminished recordings available to consumers. IMO, that provides a plausible rationale for sacrificing a small measure of Recording Accuracy for the sake of potentially greater Event Accuracy. Put another way, it provides a rationale for the ADDITIVE approach to playback.

Just which types of additions are the right ones is another matter entirely.

Bryon

02-08-11: Mrtennis
it would facilitate communication if there were "standard" definitions for terms frquently used in discussion so that each of us would not have to qualify or explain the denotation or connotation of terms used in a sentence...

this is especially the case when some one asks for advice regarding an aspect of sound, such as "bright" and the posters would understand to a large extent, what was meant by the term.

I agree with you, Mrtennis, and I think you've identified another term that is often in need of clarification, namely 'bright.' Some people use it exclusively to refer to frequency response, but other people seem to use it more loosely to include anything they don't like about high frequencies, like shrillness, grain, glare, etch, etc..

Of course, all of these terms are imperfect descriptors of what is actually heard. But the subtle differences among them are often significant, as they can suggest different diagnoses of the problem, and therefore different remedies.

FWIW, what I find particularly valuable is any effort to correlate subjective terms like 'warm' or 'bright' with the objective characteristics to which those terms could refer, which was part of my motivation for beginning this thread. Correlating subjective descriptions with objective characteristics not only might help facilitate communication among audiophiles, but it might also contribute to our understanding of why some systems sound more like real music than others.

Bryon

I would like to comment on one thing that Hifibri said - while I agree with pretty much all of the rest of the post, I would disagree with this part: "Recording studios can and usually are 'warm', dead maybe, designed to lack reverberations to control the sound, but not usually characterized as cold. Great pains are taken to control studio acoustics. "

Great pains are usually taken indeed, but almost never to make it "warm." In fact, quite the opposite - the engineers want the room to be as dead as possible, as they want to totally control the sound not only of the room itself, but even more importantly (and objectionably, to us musicians), of the actual instruments/voices. This is true not only of small studios, but also of the big studios in Hollywood and London. Some very famous musicians truly detest what some engineers do to their sounds in the studio, including in the top movie studios.

By the way, this does not necessarily mean that the resulting sound is bad; but although it may have cool effects, and the recording itself made and edited and mixed very well, it usually has very little to do with what the musician actually sounded like (although for the vast majority of pop singers, for instance, this is actually a very good thing, and they love it).

Also, this is not to say that the studio cannot be made to sound more like a real performing venue - once I had the pleasure of playing with an orchestra I was in with Georg Solti in Abbey Road studios (it was a one-off rehearsal in a training orchestra), and most of the deadening treatments in the room were pretty much removed for the purpose. And this is sometimes done for big budget films where the score is an even more than usually important part of the film. But what I am saying is that this is never done for the overwhelming majority of studio recordings - if they wanted it to sound like a concert hall, they would record in one. Usually, the room sounds so cold and dead that it is actually hard to hear your fellow musicians - the sound dies almost as soon as it leaves the instrument. Of course, normally there is a click-track going on in earphones you are wearing anyway, so there is very little sense of ensemble in any case. And of course, it is usually a much smaller ensemble than a full orchestra, but that just gives the engineer that much more control over his production.

Bryon, some of this also relates to your Recording Accuracy/Event Accuracy thing. Obviously, I am almost always much more interested in the latter than the former, with the type of music I perform/listen to. But if one listens either entirely or at least primarily to electronically produced music, then all of the above is nowhere near as big of a deal (if not practically irrelevant!). It is certainly a hell of alot easier for engineers to manipulate the sounds of electronic instruments exactly how they want to.

I like the contrinuum someone put down between analytical and warm and syrupy, though in truth, you could use other terms at each end. Like many respondents on this thread, I like a sound on the warm side of neutral too. I take it to mean richer harmonics, with a full rendition of a note, not the leading edge alone.
I think I get this from tubes, somewhere in the system. I can't think of an all solid state system I have really enjoyed.
A subject I can't remember being discussed, although it's fundamental, is just what should a perfect, audiophile sound be? Yes I know one talks of being nearest to a live sound, but a system that reproduces one type of live music well, may not for another. Do systems always reproduce a jazz quartet and symphony orchestra optimally.
If you like a warmer than neutral sound, are you betraying audiophile ideals. Should you proverbially have your badges of rank torn off and sword broken.
Ultimately of course, it does'nt matter one bit what others think. If it sounds good to you, it is good.
If you want an example of a fundamentalist approach to this hobby, it is Arthur Salvatore. I enjoy his website enormously, but reading it, you realise that for some, there is only one path to enlightenment

hi byron:

my idea of establishing definitions of audiophile terms is to define them using objective terms.

in this case, the word "bright", refers to an elevation in frequency in the upper mids/lower treble. so i would say a n audible peak (2 db +?) in the range 1000hz to 5000 hz might suffice. it could be narrowed down and defined more precisely by selecting different deflection points.

the point is my definition, even if it is inaccurate is an attempt to clarify, using terms which are replicable, what is experienced by a listener.

perhaps the range should be 3000 hz to 5000 hz. certainly frequency response can be measured in a room.

it would interesting to see under what circumstances, i.e., what frequencies, correspond to reports of brightness. while definitions could be objective, perceptions may differ as to the occurrence of brightness, creating a problem.