How loud do you listen to your system

IB brought up the Fletcher-Munson curves for discussion, highlighting a unique phenomenon in auditory perception.

At very low volumes, our sensitivity to both low and high frequencies significantly decreases, causing these frequencies to appear muted compared to the mid-range. Interestingly, this dip in sensitivity for bass and high frequencies becomes less pronounced at higher listening volumes. In fact, at very loud volumes, our ears become nearly equally sensitive to all frequencies, which can be overwhelming since we’re not accustomed to such intense bass and high frequencies.

While the Fletcher-Munson curves have been superseded by more accurate models like the ISO 226:2003 contours, the basic principle they illustrate – that our ears have different sensitivity to frequencies at different loudness levels – still holds true.

3 Likes

Hadn’t measured the background level before. room is 35-40db using Sound Meter on Android. I listen, depending on mood, at between 75 -85db (I know I am slightly hard of hearing). This is with NDS and 552 at about 7-8am on the dial

1 Like

hmm our ‘sensitivity’ to volume; yes I certainly use this to set my speakers in a surround setup ‘by ear’ when missing an SPL meter and/or the inbuilt microphone for calibration/setup.

By setting the volume at a lower level my ear is a lot better at picking up the respective differences between speakers, and assuming no huge timbre differences, generally gets me .5dB accurate across all the speakers.
Some caveats with this is that not all amps scale well when the volume goes up, but in more than 80% of cases, this ‘set the speaker volumes with the volume set low’ rule seems to work well and scale correctly… (the speakers at listening level remain half a decibel give or take to each other)

Regarding my daily listening volume; completely depends on genre and ‘neighbours’ and whether the weather requires windows and doors open/closed etc.

What I find interesting is that when I go to my friends house who uses ~90dB sensitive Bower and Wilkins monitors, at the listening position in a room that has had effort given to sound control, we are typically at 65dB peaks when we are ‘jamming out’.
Not sure if being drunk/stoned/relaxed aids the ‘sensitivity’ of OUR ears (probably), but that seems like such a low level (in terms of capping the dynamic range) that I feel we are hardly listening ‘to the system’ (always ‘the music’)…
Always a good listen, and as I have matured in life, happy to sit their sober as a monk and still audition cable and equipment upgrades on said system.

My own den setup; owning two high quality SPL metres (one analogue and one digital) that agree with each other…
at 9o’clock on the dial of the XS3, depending on the recording - ie ‘made for radio’ compressed can easily be 10dB higher at the same volume level (so lets IGNORE THAT STUFF); I generally find that 65dB is a very happy level for enjoying music as ‘background’ eg as I write this… (Orchestral Pink Floyd by the London Philharmonic)

which, for orchestral stuff, if listening CRITICALLY, I would generally want the volume to peak at 85dB (so as to maximise the dynamics)…
when measured last week, I found that between 9 and 10 o’clock on the ‘lowly’ Nait XS3, I was swinging from high 50s’ decibels to greater than 85 decibels…

for in room listening this is ‘large scale’ for my ears and any higher wouldn’t be comfortable.
Don’t get me wrong… growing up (before I ‘grew up’) I would use super high powered amplifiers and incredibly capable speakers and really march up the volume… (Seldom to be listened to ‘in room’).

The XS3 seems to have the headroom to render most of my music catalogue very well.
Ben Harpers’ Faded (opening track - Will to Live) is one of the few examples of a track that seems to be missing headroom capability (even at low volume) to recreate ‘the best’ rendition I have heard…
but for everything else, the Nait integrated seem to have a lot of raw grunt and recreate better music than many many “more powerful” (by the numbers) amps I have owned.
The speed of the amp actually reveals a lot of lyrics and fast instruments (or notes between notes when played ‘at speed’) that many other systems just miss.

My setup uses 90dB sensitive speakers (Krix Apex towers), and at the listening position around 3.4metres away, the need to push past 9o’clock on the volume dial (XS3) is basically nil.
Except when playing music into the front yard (which needs to get through ‘a few rooms’ and out ‘small openings’).
Sheryl Crow at 10o’clock would be ‘too loud’ to comfortably listen to ‘in house’, without wanting to tap toes/‘dance’.

TL:DR I’m a 65dB person,… but like peaks around 75dB if trying to pull apart a track and appreciate the capabilities of the system (factoring a little bit of dynamic response as well)

2 Likes

Thanks everyone for the replies. Remarkably similar. Seems most are at upper 60 to upper 70 dB region when actually listening rather than background, occasionally into the 80’s.
This matches my levels!

:+1:

With extra for ‘When the Levee Breaks’ in my case….

2 Likes

All true but what does it have to do with how loud people are listening? Genuine question BTW, not trying to be clever.

Just interested. I know how loud I listen and that I need a certain level for the sound to become solid. I also know that in demo rooms it tends to be even louder than I had it. Plus, I’ve read somewhere speaker manufacturers use about 80dB at the listener position. Also, there are often comments about a system being better at lower volumes than other systems.
So, interesting to get the context and see if my listening levels, how I use the system is typical. It seems to be

I listen to classical music. Everything sounds unnaturally loud above 9-9.30.

It would be useful to run some test tones.
I would imagine a great deal of volume adjustment to maintain an 80db sound level at listening position through the frequencies.

It would depend on how even the frequency response is of the system including room, and unless great care has been taken with positioning of speakers and listener there may very well be huge variation - and in some rooms regardless, unless well treated acoustically. But raather than individual test tones it is easy to do a frequency sweep with software like REW, using a measuring microphone at the listening position, to then show graphically. But do several with the microphone moved a few inches between to represent all positions your ears might be in, as the precise positioning can make significant differences (making nonsense of millimetre adjustments of speakers if head isn’t clamped in one precise spot!).

Yes. I can play some material that is quieter than the ambient noise level in the rooms listening point. Yet the bass sounds are easily discernible in the room upstairs.

It’s debatable whether one should aspire to having a system which has a ruler-flat response from 20Hz-20kHz (and beyond), given ears don’t behave that way.

Even the microphone you would use for such measurements wouldn’t have a ruler-flat response either. The classic Shure SM58 (with which the vocals on most of your favourite pop-rock recordings from the last several decades were probably recorded) has a response which is deliberately engineered not to be flat!

[just spotted that the link I was about to post to a helpful review with graph and explanation is from a site with adverts, so can’t post it under forum AUP. Google ‘Shure SM58 frequency response’ for yourself…]

The equivalent is true for other microphones, hence why engineers and artists will each have their own favourite microphones.

Mark

2 Likes

I’m sat at a table in my day room, listening to the radio playing in the kitchen. Its source is a Unitiqute. SPL dBA is around 44, with a background level 28. I can hear the birds singing outside, but traffic noise is muted by secondary glazing.

I mention this because, although the kitchen radio was on when I was preparing breakfast I thought I would need to switch to my study source when I was eating it. But there was no need. In fact the SPL dBA sat at the same table but listening to the source from the study is around 45. Perhaps I do not need the Unitiqute in the study?

My main system’s volume knob has to be altered depending on source, TT, radio, NDS or TV: it provides with an excuse to cross the room. Whatever the source, the SPL is around 60-65 dBA, with the knob moving from 2000 hrs to 2200 hrs.

1 Like

It depends of course on whether you are happy with the sound as on the recording you’re playing! That said, relatively small deviations of just a few dB from an even response aren’t generally very noticeable. As for levelness, my understanding is that a normal sounding in-room response would be a gradual increase in sound level across the spectrum with bass end several dB louder than top end. I’m not sure where/when I learnt this, but to me that certainly sounds more natural (and pleasant) than a completely level response.

1 Like

Many have put forward the notion that some music material needs the volume set at a specific point to bring everything into focus. I think this is a generalisation best avoided.

Regardless of what I think, for a system to perform very well giving excellent replay at very low levels will need to take into account many variables - perhaps different to a system that performs best at very high levels - or indeed a system that can do both.

1 Like

Measurement mics, on the other hand, do have a flat frequency response…

1 Like

My friend Mike. Getting a flat top response, front end party and back end party makeover.

5 Likes

When we inquired about playback loudness, the aspects of sound quality, such as perceived dynamics and noise floor, emerged as two of the key contributing factors. While the actual dynamic range of a system or recording remains constant, the perceived dynamic range can vary with loudness levels. This variation is especially notable in environments with a high noise floor, due to the masking effect.

1 Like

But as I said, whilst noise floor will almost certainly have an influence on low chosen sound levels, the significance reduces with higher level listening, once people play at 70 to 80 dB average, or more, which are generally chosen for personal presentation preference and/or realism in presentation. (Here I am assuming a domestic environment not industrial, with ambient noise levels typically 30-40, maybe in more extreme cases up to mid to high 40s)

1 Like

Measured with phone 60-75 dB.
About 9:30 on my SN3