On listening at lower volumes.

Discussion in 'General Audio Discussion' started by yotacowboy, Dec 9, 2021.

  1. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,696
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    So, I'd been dropping little hints here and there when I was discussing the sound of my Elekit tu-8600, that listening at lower volumes had become a "thing," if not a quite more pleasurable "thing", not due to any absolute restriction on volume levels (aside from living in a condo and being respectful of my neighbors). But, I don't typically listen louder than 85dB, sustained.

    I'm finding myself listening more often at 65-70dB and am surprised that it's a completely compelling experience.

    So what gives here? what within an amplifier design/speaker parameters makes something like this work (better)?

    Personally if think this "enhanced" low volume listening is a combination of two things, basically:
    • greater than 92-ish dB/W speaker efficiency (and not dicking around with 2.83v/W kinda fake "efficiency" numbers. I'm also not interested in the differences between efficiency and sensitivity in this use case, either)
    • high plate voltages found typically in tube based topologies
    So school me. What's making this recipe work so well, given my experience?

    My thinking is with proper iron and a good high voltage plate supply, the pint-sized versions of macrodynamic contrasts become something that no (ehh, perhaps very few) solid state amps can accomplish.

    I'm no EE, so i'm probably, no, 100% likely talking out of my ass, but can anyone help me understand from an engineering perspective how my 8W 300B amp and sort of sensitive speakers sounds about 37 times more compelling at 65dB than my old Schiit Aegir monos?

    But the bigger picture is this: how much of an audio experience is purely sheer volume based? Is there any research to support what's actually going on when people anecdotally report that "louder is perceived to be better" when citing some never actually cited science literature?

    What makes us turn up the volume? Is that same stimulation possible if the effective dynamic range is increased while peak SPL is decreased?
     
  2. Jinxy245

    Jinxy245 Vegan Puss

    Pyrate
    Joined:
    Nov 17, 2016
    Likes Received:
    5,731
    Trophy Points:
    113
    Location:
    Montgomery, New York
    I'm curious what people come up with as well. I have read that the mind does compensate after a while of listening at lower volumes to "fill in" the gaps as it were. It sounds like kind of a reverse effect of the ears adjusting to playing at louder volumes, but for the life of me I can't find the article I swear I read.
     
  3. k4rstar

    k4rstar Britney fan club president

    Pyrate BWC
    Joined:
    Jun 11, 2016
    Likes Received:
    6,942
    Trophy Points:
    113
    Location:
    Toronto, Canada
    An engineering answer to such questions will be unsatisfactory at best and misleading at worst. You can talk about the relationship between power and speaker sensitivity in hard terms and while it may have something to do with your sensory stimulation from music, I think it really has little to do with your emotional and subconscious engagement with music.

    Let's keep in mind but put aside that studio-produced recordings have critical volume levels at which they are mixed and therefore have the most 'correct' instrument balance. Generally the more grossly compressed the recording it is, the easier and more desirable it is to find the critical volume level when listening at home. Avril Lavigne just doesn't hit the same at 65dB. Neither do most symphonic works where large contrasts in dynamics are integral to the aesthetic of the musical work. It's not really surprising then that most 'low volume' listeners, either by choice or force, gravitate towards vocal, jazz, and chamber music.

    From an engineering perspective, what there is to consider is the acoustic power of the system (a function of the amplifier's output capability into a given load before clipping, the electrical sensitivity of the acoustic system), the boom radius (two systems with similar sensitivities but different acoustic dispersion and displacement characteristics fill a room differently), and the room area/volume. But this only tells us the story of what peak SPLs we can expect without distortion. I'm sure we have all heard systems that play loudly and cleanly and yet are not interesting to listen to.

    So this brings us to the subjective and esoteric. Large contrasts in dynamics mostly reside in the listener's first level of perception and can even be considered as semantic or detached from the aesthetic of a given musical work. Dynamic shading however plays with the listener's emotional perception and doesn't have much to do with how loud a system can play (or how loud we listen). It's a completely natural instinct to turn up the volume when such shading is missing in an unsuccessful attempt to compensate. I suppose you get more of the latter in your new system. You can also get it with inefficient speakers and sand amplifiers but it's rarer and also more challenging on the listener's perception.

    Even if you zoom out from the topic of sound reproduction and ask why we as human beings like it when the drummer sometimes plays soft and then plays loud you'll find that science has no real answers.
     
    Last edited: Dec 9, 2021
  4. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Lower listening levels translate to lower electro-acoustic diaphragm excursions and typically lower distortion, especially in the bass region. SET amplifiers are likely have higher linearity for less deviation from zero signal set point. Once one has acclimated to lower levels the benefits are many. As mentioned, lower distortion, less wear and tear on the stereocilia in the basilar membrane, and usually lower fatigue.
     
  5. ColdsnapBry

    ColdsnapBry Almost "Made"

    Joined:
    Feb 4, 2021
    Likes Received:
    640
    Trophy Points:
    93
    Location:
    Earth
    I was just thinking about this as my headphone setup has some seirously tempting volume creep going on. Unless I keep a db meter around I'll go beyond 80db. I then set it back to 65-75 and think it sounds pretty boring and flat.
     
  6. Vtory

    Vtory Audiophile™

    Pyrate MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    10,831
    Trophy Points:
    113
    Location:
    East Coast
    Interesting topic. But I'm another person who is a little skeptical about engineering/scientific explaining power for the issue you bring up. This is largely because my personal experience was kinda mixed -- I can't clearly prefer one way of listening over the other. Sometimes I find listening at lower level better in some subjective area I value, other days the opposite, etc.

    Also it's not even dichotomy (i.e., loud vs quiet) to me.. rather a preference over a continuous space (what's worse, multidimensional). Apart from sound pressure level, a lot of confounders make this problem super tricky imho.

    That being said, I'm mostly thinking of two different aspects regarding why small and large volume listening differs.

    1. Small signal to environmental noise floor level ratio

    Environmental noise level may vary depending on where we live and how noise generating items are placed around the listening position. FWIW for non-outlier people/places it mostly ranges from 20-ish dbA to 50-ish dbA. In my apartment, quiet night with only computer fans running, noise level is around 30dbA. When the fridge in the same room activated and/or in normal condition during day, the noise floor increases up to 40-50dbA. And note these are all A-weighting. I used it because it's a standard unit in noise research. This can be even higher with wider-spectrum weighting (C, Z, etc).

    On the other hand, many modern dynamically compressed tracks come with the crest factor of 10 or so and zero-ish headroom, which mean there is 10db of power difference between signal peak and rms level. And obviously this is only half of the story as it only tells us about average to upper bound. The opposite is very tricky to assess tho. Why? it's theoretically unbounded in logarithmic scale (i.e., log(0) = -inf) and practically dominated by recording and playback noise floor limitation. Based on my analytic experience with tracks I have (I enjoy running MasVis with favorite music), I am under the impression that I may want to hear at least -30db relative to peak even with crap recordings.

    Suppose I listen to music at X db SPL at peak. SPL range of my interest would be [X-30, X] in db SPL (Z). And I have an environmental noise occurring at [30, 50] db SPL(A). So, at least I can confidently argue listening experience would NOT be level-invariant.

    2. Equal-loudness contour (ELC)

    I believe most of us are already well aware of this. But I further argue we have to interpret it in the multi-dimensional context.

    Assume simple (maybe oversimplified) example of a combination of 0dbFS at 100hz and -20dbFS at 1000hz. And suppose our playback system (Setting A) translates these signals to 100dB SPL and 80db SPL respectively.

    According to ISO 226 (Aside: let's not be confused between ELC and Fletcher-Munson curve.. FM curve is one early implantation of ELC published in 1933. ISO 226 was revised in 2003.), Our perception will be
    • Signal 1: 0dbFS at 100hz (electronic signal) --> 100db SPL (sound pressure) --> 99 phon (human)
    • Signal 2: -20dbFS at 1khz --> 80db SPL --> 80 phon
    Now let's consider a quiet listening playback setting (Setting B). Suppose this system translate input signal to 80 and 60 db SPL respectively. Recalculation of perceived loudness below.
    • Signal 1: 0dbFS at 100hz --> 80db SPL --> 62 phon
    • Signal 2: -20dbFS at 1khz --> 60db SPL --> 60 phon
    Difference between perceived signals (originally differ by 20db)
    • Setting A: +19 phon
    • Setting B: +2 phon
    This must be way simpler than any of real music. Also if I recall correctly, human sensory system would be more complicated with sustaining stimuli. But anyway it will be an interesting hypothetical example to explain why listening experience may not be level-invariant when others being equal.
     
  7. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,696
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    I think there could be a mental model that given greater dynamic gradation, i.e., not greater macrodynamic peaks, or microdynamic resolution above the noise floor, the brain spends less time filing in the dynamic gaps, regardless of volume. Maybe with "better" quality gear, what I'm finding is that rather than trying to find the "correct" instrument balance due to certain recordings being mixed with a critical/specific playback level (like i was used to with the Aegirs), I'm getting a "more correct" instrument balance at more varied volume levels. Like, turning up the volume isn't trying to get lower-level information to come up in the mix, rather, turning up the volume is just raising the upper macrodynamic peaks. In other words, it's not outright dynamic contrast that "sounds better", instead, it's higher resolution of the differences between dynamic content.
     
  8. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,467
    Trophy Points:
    113
    Location:
    Winnipeg
    I'm gonna factor a lot of this on the psych/biological factors. The stapedius reflex (aka acoustic reflex) is a muscle in the ear that is typically tensed all day to restrict the ear canal and protect your eardrum from too much noise. This muscle does not relax quickly, it takes some time. Low level listening in this state will be unsatisfying because the sound literally has difficulty reaching your ear. Oftentimes people will "turn it up" because they want to "feel" more, which in turn tightens the muscle more, etc. You can make reasonable inference that muscles tightening inside your head can act as mild stressors that might reduce the enjoyment of music, even if it's subconscious.

    If you've settled down and relaxed the body such that your ears have opened up enough for low level listening, you are probably in a better mental and physical state overall.

    edit: think of it like this... your body's overall sensory organs have a baseline "energy" or "stimulation" level. Incoming stimuli has to meet or exceed this level for you to really notice, and even then it's fighting all the residual energy lingering in the system. Bring it all down, and there's less garbled bleh in your body (which in general is a good thing) and your sensitivity to input improves.
     
  9. Tchoupitoulas

    Tchoupitoulas Friend

    Pyrate Contributor
    Joined:
    Aug 17, 2019
    Likes Received:
    3,711
    Trophy Points:
    93
    Location:
    PA
    Please excuse the following for not offering anything substantial but two thoughts spring to mind from reading the above:

    I've been meaning to post this in the SW51+ thread but the following is germane here: over the past few months I've been loving my Bifrost 2 (Unison) --> SW51+ --> HD 600 setup. For the first time, I've found myself routinely turning down the volume; if anything, I've found the experience of listening at lower volumes more enjoyable. I'd assumed this had something to do with the even, perhaps slightly forward mid-range response. But I gather now from @atomicbob that this may be attributed to the benefits of SET amps. Thank you! (The SW51+ is my first SET amp).

    I wonder if this may be part of the reason why many of us enjoy listening to music late at night.
     
  10. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    This is an excellent point. I find playing music that has content conducive to relaxation at the beginning of a listening session to produce a very pleasant period following with what ever styles of music I wish to experience for the evening with the stage set for lower listening levels.
     
  11. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    First all the distortion graphs are when you push things at their peak. The lower the power the lower distortion artifacts are, especially for SET amps. In fact when you look at THD vs power graphs you can see that THD lowers in a Class AB amp with more power whereas distortion rises with a Class A amp:

    [image 1 Class AB, edited later]

    [image 2 Class A]

    Very sensitive speakers mean even lower amplifier demands as well as powering drivers below when their break-up modes.
     
  12. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,696
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    Another thought: a system becomes resolving enough when the volume control is really more of an "engagement" control? I noticed this effect first when I had EC Af + HD800SDR. Now that my speaker rig is capable of similar resolve, it's a similar effect, but only about 827% more convincing since it's not headphones. Sort of like the volume is no longer associated with sheer loudness, or turning things up to hear more detail, or more bass, or whatever. But instead, changing the volume affects the "liveness", or the level at which you're choosing to be emotionally engaged with the music.

    either way, maybe this is a shitty humblebrag, but IME, this stupid Elekit amp and third-hand AN UK speaker setup of mine is convincing me of the ways of the flea watt. I feel like i've crossed a threshold or something. whether that's a good thing or a bad thing, I dunno.
     

Share This Page