Each recording has one right volume level.


This started from a reply by mijostyn, but I wanted to create a new distinct topic since it is critical, but misunderstood.

I think most people here will be familiar with Fletcher-Munson equal loudness curves? In summary, how that applies to audio listening, is that the perceived tonal balance is heavily dependent on listening volume.

At a basic level, if you want to recreate the tonal balance of the original recording, then you need to recreate the volume of the live music, or the volume used in mixing and mastering. If you don’t, then you are not listening as intended.

One way that applies to audiophilia is when we are trying to compare components and any number of "tweaks". It is critical to maintain constant volume when making comparisons or the perceived tonal differences can swamp out any component differences, leaving a proper choice impossible, though you may blame it on a component.

A second way, which mijostyn raised, that applies to audiophilia is perhaps this concept of "flat frequency response" is flawed w.r.t. recreating a musical experience at any given volume. To the post title, "Each recording has one right volume level". What if we are not at that volume level? If we are not, then arguably we should be equalizing such that the perceived tonal balance matches closer to the tonal balance at the intended listening volume.

Enjoy the discussion and keep the mud to a minimum.

mijostyn1,269 posts11-01-2019 2:11pm Without loudness compensation each recording has one right volume level.

atdavid

Showing 6 responses by atdavid

Most classical / jazz / and similar are not usually played at ear destroying volumes. Same would be true for everything outside of pop / rock / probably dance.

But .. and maybe some of the recording engineers could pop in on this one, we rarely listen to live rock and pop music at home, we listen to studio music and that is not mixed/recorded at front row concert levels but something more akin to what you may listen to at home.
Elizabeth, nope, you are exactly the target audiophile audience. How can we adapt the signal getting to your speakers to maximize the experience ... this holy grail of "flat" may be ideal when comparing speaker to speaker or amp to amp, but not listening experience to listening experience.
In the home theater world, these techniques are already being used to maximize the perceived experience. Why not audiophiles?
Today you could embed that in metadata or provide an online database where that information is available. I expect mostly it is not.
If by graphic you mean finite impulse response filters in DSP with a graphic user interface and song/genre presets .... You may be right :-)
While I think I know what you are trying to communicate at this point, I am not sure I agree completely. If the recording is "flat" or at least as intended, then the playback should be "flat" as well at least to sound as intended at the volume level intended and you are going to get closer to that with a meter than with your ears. In theory, at least at the recording level, all vagaries of the Fletcher Munson curve have already been taking into account (within the limits of the recording engineers listening equipment).
If you are moving away from the intended playback level, and of course that would vary in level from recording to recording, then no, a flat response would not be ideal if you wish to experience the tonal balance of what the recording engineer/artist intended. The problem is, by ears or by meter, you are always going to be wrong to some level.
millercarbon1,869 posts11-02-2019 1:53am
I think most people here will be familiar with Fletcher-Munson equal loudness curves? In summary, how that applies to audio listening, is that the perceived tonal balance is heavily dependent on listening volume.

But that’s only the first part of the equation. The second is what happens with the system and room when audiophiles enter the picture. Because the same Fletcher-Munson perception applies as well in your listening room as their recording studio.

This is where too any audiophiles screw up. Using meters, which are not ears and so do not "perceive" let alone experience Fletcher-Munson perceived level that varies with volume, they set things up to be "flat" and not by ear but by meter. They never consider what sounds flat only sounds flat at that one volume level.


Millercarbon,
I understand, quite well. the Fletcher-Munson curves thank you.

However, there is a difference between understanding them, and applying them correctly. I disagree with how you attempted to apply them w.r.t. my post.

By thinking what I wrote is wrong, you are perhaps illustrating your lack of understanding at a whole, beginning to end, system level, the application of the Fletcher-Munson curvers, not my lack of understanding of Fletcher-Munson curves.

1) When the recording engineer makes the final mix, his impression of the tonal balance, is influenced, as indicated by the Fletcher-Munson curves, by the volume level that he is listening at when he makes the final mix.

(Side note, the engineers playback system is likely close to flat.)

2) IF the listener wants to experience the music, approximately as the recording engineer intended, with the same approximate tonal balance, they must both listen at the same volume AND their system must be METER flat. If you change the volume, or your system is not flat, then the tonal balance will not be the same as what was intended at the time of mixing and mastering (ignoring minor differences between hearing of the recording engineer and listener -- note on that later).

3) If the listener is not listening at the same volume as the engineer, then equalizing the system, using the Fletcher-Munson curves and how they vary between the two volume levels, could restore at some level, the intended tonal balance ... if your goal is to hear what the recording engineer intended.
Euphonically, we all do have different preferences, so you can adjust however, you like, but that is personal preference, not specifically aspects of Fletcher-Munson. If I understood the data correctly, while there are of course variances from listener to listener in their own personal Fletcher-Munson curves (what is published are averages), the differences are not large.


I think your comment, "volume for tonality, it's volume for dynamic range" misses the point of my post. My point is that if you want to hear the recording as intended you either need to listen to it at the volume it was mixed and mastered at, or you need to equalize with the Fletcher Monson curves as your guide between the intended volume and your listening volume. That won't be perfect, but would be "better"
Dynamic compression is a completely different topic. That is a choice made at the mixing/mastering stage, which is still done at a specific volume.



geoffkait17,910 posts11-02-2019 6:14amBasically the volume level of a given CD is a function of its dynamic range. That’s why one is inclined to turn the volume knob higher (or lower) for some CDs relative to the one you just played. The louder the CD sounds at a given volume setting the lower the dynamic range. Example, Mercury Living Presence classical CDs generally have high dynamic range. So when you first play one you’ll notice the volume is rather low for when the volume knob is set at. That is to account for the large dynamic swings that come later, you know, so you won’t blow up your tweeters or woofers.

So, there is a happy medium for CDs with high dynamic range, where you can select the right volume and get the full dynamics of the recording. It’s subjective to some extent, obviously. That’s what the Loudness War is all about - providing high volume level but overly compressed dynamic range. Yuk!
So, I’d say the trade-off is not volume for tonality, it’s volume for dynamic range.