Should you use Fletcher-Munson loudness compensation?

[ZolaIII]

That's what works for me, yes. But it's subjective. What I would advice is to simply adjust it to your liking. If you're "critically listening" (not a fan of that term), i.e. concentrating on the music and want an accurate representation, you're likely to listen at a louder level anyway.

You can try to make a filter like described by solderdude on page 2:

Should you use Fletcher-Munson loudness compensation?

I know many are familiar with the Fletcher-Munson or equal loudness curves. In a nutshell: “The Fletcher Munson Curve contains a set of graphs that show that when you listen to music at a lower volume, mid-range frequencies will be sound more prominent, whereas high and low frequencies will be...

audiosciencereview.com

E.g. with the values in the pic attached and try to listen to it at lower volume.

It sounds really really bad. It's muddy, dark, and lacks treble. There is no clarity. Vocals are mumbled.

Compare that with the filter in the second screenshot. Sound is clear and airy. Vocals are crystal clear as they would be if listening a much louder level.

My own personal conclusion is that equal loudness contours only tell how a person perceives the level of one frequency vs another in isolation (which is how the tests were conducted). It's not really valid to how we perceive complex sounds such as music. Our brain is a powerful EQ on its own and has expectations how combinations of sounds should sound like.

Edited: As a sentence accidentally slipped out.

It's second screenshot regarding Q (Butterworth) but it's meant to be a lot higher at 105 Hz. What can make or brake things is how attended are transition frequency above the 105 Hz and I advise crossover at 120 Hz for sub's (which then nead to be to the mains and in 2.2 one each chenel configuration). You don't need to do it subjectively, you can use SPL meter and form there adjust to the curve output on speakers by JRiver (ISO 226 2003) screenshot. Rest is more to the subjective taste do you like slight boost (about 2 dB) in sub bass and can do it technically or not. The highs filter is less important and pronounced and comes into play only under 60 dB (normal speech loudness) programme SPL.

 

[Old_School_Brad]

Yes, you should use it

The equal loudness curve *is * psychoacoustics. Why would you think we somehow 'compensate' for it? It is a measurement of how we actually hear. You not only hear disproportionately less bass , you also perceive it so.

It is why loudness buttons were invented.

+1

And Yamahas approach is very nice. If someone (Dirac?) comes up with an active solution like a fixed microphone and a loudness function on- the fly related to SPL I'd buy it.

 

[jeffaegrim]

I don't know what Alexander Walsch did in his EQualizer-APO loudness correction module. It seams to compensate wrong:

here is his -20dB curve vs the one from AutoEQ, https://github.com/jaakkopasanen/AutoEq/blob/master/research/equal_loudness/70vs90_loudness.csv, which also matches this one, https://github.com/dpapavas/alsaloudness/blob/master/filterplots.png

View attachment 247755

now before you tell me that the EQ-APO one has 75dB as a reference, vs 90dB reference for the others, the deltas for the references in this range are very small: https://github.com/andrewjhunt/equal-loudness/blob/main/images/spl-deltas.png

also, and this is even worse: even though it seams to be possible to calibrate to 75dB and listen louder, it isn't. the flat point after calibration is always at 0dBFS....while it should be shiftet down to -20dBFS when you ajust your full loudness to 90dB for example

Thanks for the information. This is a shame, because dynamic loudness controlled over the windows volume slider would be just the thing I’m looking for.

Have you found an alternative solution that doesn’t involve purchasing an ADI2? Or, perhaps, any information on whether the implementation in eqAPO has been fixed?

 

[mmi]

+1

And Yamahas approach is very nice. If someone (Dirac?) comes up with an active solution like a fixed microphone and a loudness function on- the fly related to SPL I'd buy it.

Sonos? It is volume dependent, and does a great job l, I keep it permanently on for all my Sonos speakers.

 

[dasdoing]

Thanks for the information. This is a shame, because dynamic loudness controlled over the windows volume slider would be just the thing I’m looking for.

Have you found an alternative solution that doesn’t involve purchasing an ADI2? Or, perhaps, any information on whether the implementation in eqAPO has been fixed?

you have seen this one apparently: https://www.audiosciencereview.com/forum/index.php?threads/my-custom-loudness-control-tools.50021/
I haven't tried it out since he never replied to my questions

Personally, I use EQ curves that I have to manually select in EQ-Apo. not very convenient, but at least I have total control

 

[jeffaegrim]

you have seen this one apparently: https://www.audiosciencereview.com/forum/index.php?threads/my-custom-loudness-control-tools.50021/
I haven't tried it out since he never replied to my questions

Personally, I use EQ curves that I have to manually select in EQ-Apo. not very convenient, but at least I have total control

Thanks for the response. Yeah, I’ve seen it - I might bump that thread in the hope of summoning back OP.

I’ll do it your way if everything else fails. Maybe that loudness correction setting in eqAPO doesn’t actually sound that bad - it does look ‚relatively‘ compliant on your overlay. It’s just too bad that Alexander is a basshead. Alternatively, that plug-in could be something to look at. There is also a loudness VST plugin by WOK: http://wokwave.com/old-plugins/html/wok_loudness_free_vst_plugin.html

(though it doesn’t seem to offer dynamic loudness)

It’s a pity, anyhow. If anyone has any elegant solutions we haven’t found yet, I‘d love to know them

 

[sethr]

I use the equal loudness function in SlickEQ. I wish it was a little more configurable, however.... I have it installed in Equalizer/PeaceAPO so I can use it system-wide in Windows 11.

 

[HaveMeterWillTravel]

One of the issues with most implementations is it doesn't scale with the actual volume, but only a gain setting or is simply static.
If a passage is 20-40 dB quieter than the rest, it really needs a different "eq" than the louder passages.

 

[dlaloum]

One of the issues with most implementations is it doesn't scale with the actual volume, but only a gain setting or is simply static.
If a passage is 20-40 dB quieter than the rest, it really needs a different "eq" than the louder passages.

Yep, you have to have a known reference level for it to work... which is great with Movies that are recorded to standard reference levels - but for music, where there is no reference, you have to adjust by ear - and that means having the relevant adjustability in your gear/software.

The better 1980's Pre's had an adjustable loudness dial to handle that.

 

[ZolaIII]

Yep, you have to have a known reference level for it to work... which is great with Movies that are recorded to standard reference levels - but for music, where there is no reference, you have to adjust by ear - and that means having the relevant adjustability in your gear/software.

The better 1980's Pre's had an adjustable loudness dial to handle that.

It scales to calibration levels, automatic one. You use normalisation at surce and with EBU R128 it scales good with anything you can apply R128 too and still better with music. Problem is when you can't and broadcast isn't adopted to it or it's adapted to not reference one (basically all streaming services and not EU legally obliged sources). But you compensate it back (R-128 output levels) so it's not a big deal anyway. Manual one with a dial always whose to SPL A reading from listening spot.

 

[sethr]

Another option - JRiver has a loudness function that works with R128 volume leveling. You have to do volume calibrating with some noise first.

Can't use it system-wide though...

 

[ZolaIII]

Another option - JRiver has a loudness function that works with R128 volume leveling. You have to do volume calibrating with some noise first.

Can't use it system-wide though...

You use JRiver WDM driver and it goes system wide, calibration is to 88 dB stereo white noise before enabling JRiver Volume (ISO 226 2003).
EQ-APO also has it and it works similar and him acting as WDM driver as well. WDM is Windows only. You have EBU R128 in form of plugins for many players however you don't get the ISO 226 that way (there is such in iZotope but iZotope is more of a DAW itself and expensive). There is a ISO 226 build in in ALSA Linux sound driver as part of it.

 

[livinon2wheels]

Yes, you should use it

The equal loudness curve *is * psychoacoustics. Why would you think we somehow 'compensate' for it? It is a measurement of how we actually hear. You not only hear disproportionately less bass , you also perceive it so.

It is why loudness buttons were invented.

Agreed that we should use it, but many HT receivers sold today don't have a loudness function built in. Or if they do, it is not user switchable. I am a fan of the Yamaha variable loudness control concept and think that is a brilliant innovation that should be included in all home audio and AV equipment. It is sadly something that is not offered universally anymore.

 

[livinon2wheels]

I think this discussion is getting lost in the weeds. There is not going to be any "perfectly correct" compensation as there are way too many variables. I do think that most people, if listening at very low levels, will prefer some bass (and possibly treble) boost. That is why tone controls are the best way to do this, add as much boost as subjectively sounds better to you and be done with it. Low level listening under any circumstances is not going sound as good as playback at louder levels.

I think you're right. Loudness compensation is, because of variable recording levels and variations in recording engineers flavoring of the raw recording a very inexact bandaid applied to varied source levels. This doesn't even begin to address the differences in speaker efficiencies or their own deviations away from what an ideal frequency response curve is, or even further, personal preference, listening room, and playback level and electronics contribution to the overall target curve. So many variables involved. I would love to see ordinary tone controls as well as a version of Yamaha's approach to the problem brought back as a required feature to all consumer audio gear. It's a bandaid that is necessary due to varying listening environments.

 

[livinon2wheels]

Then indeed, you don't need loudness compensation. But you really listen at 80dB A weighted? So, around -5dB on your receiver, assuming it iscalibrated? I am not criticizing, just curious because I really don't enjoy listening that loud at home for longer periods.

In my case, even when listening attentively, I listen at a lower volume. Typically, 70dBA with peaks at 85dBA. At this level, the tonal composition does sound slightly off.
Here, loudness compensation does magic - it sounds fuller and matches much more what I perceive when (exceptionally) cranking up the stereo. Also in movies, the nice rumbling you hear when there are explosions is not the same and I would miss it otherwise.

You two guys have just illustrated the problem faced by manufacturers of the gear we use...no two people listen at the same levels EVER...and those levels vary all over the map. Again, thinking like a manufacturer, do you accept that there is no such thing as a "one size fits all" compensation because source levels vary all over the map and listening levels do too, so now again as a manufacturer, you just leave out loudness compensation all together because its one less piece they have to include due to any given implementation will be a mismatched feature for some or even most of their cutomers? Why include a feature that many will see as poorly implemented because it doesn't fit everyone's needs or any particular person's needs? People's perception of quality would be colored by the inclusion of something intended to be a feature but because it doesn't quite fit their preferences, it degrades their overall perception of the preamp or receiver that includes it. You see what I mean? As a result, maybe these customers buy a different brand next time because the loudness function was poorly implemented and didn't fit their needs. Maybe not the sole reason for switching brands, but it factors into the decision. Better to leave it out than include something (at additional cost) that is perceived as poorly implemented.

 

[ZolaIII]

@livinon2wheels it whose and is implemented as two shelving filters low and high self Q 0.71 (Butterwort) first and more important is bass one at 105 Hz and that is a knew frequency. High self one is much more mild and isn't needed until you get to relatively low listening levels. What most folks don't get is that you can't make woofer stop and get back to it's base line fast enough when the boost under the knee is big. This produces nus effect of boost up to 1 KHz and leak of the bass in mids. You only can fix this addressing it physically making crossover point above the knee (120 Hz) so that it stays on sub's but then you need 2.2 setup as from 80~90 Hz it becomes directional. Variation of even 2 dB in low bass makes a big difference in how we perceive it. As long as you have SPL meter and single free PEQ (supporting low self filter) you can play with it and even make cuple pressets you can quickly change to the mood and usual for you listening SPL. Bass slope goes inverted above calibration point (88 dB SPL stereo white noise).

 

[OldHvyMec]

It is why loudness buttons were invented.

Although I'm not familiar with Yamaha's loudness implementation, I sure like the way Mcintosh has done it for the last 50+ years. There was variable loudness (or trim) in the older analog preamps and they also had a sub/bass contour. The old Bose 901s you could toss their EQ and have a LOT better control from the preamp alone. When they dropped the bass contour (I think on the C11/22) it cleaned up some of the added tone control circuitry but it took a nice feature that worked great on NO Bass Bose.

The newer OpAmp models are still exceptional with onboard "Trim" at the remote. I'm sure there is a way to preset a few different loudness settings but, to be able to
vary the loudness is a nice feature for people like me, who enjoy the color and flexibility of good tone controls and at lower volumes

I still use L-Pads when I build a set of speakers. It takes a tad more maintenance but, it sure makes for easier room-to-room implementation if you're into non-DSP
systems like I am. I don't mind DSP in the sub/bass 250hz < but the upper frequencies, it does something my ears don't like. It reminds me of throwing a blanket
over a great-sounding set of speakers.

The older I get the more HF roll-off my ears have. That variable bump (trim) is better than eatin' bugs with a little cayenne pepper. C2500 is one of the wife's all
time favorite preamps. She just loves tinkering with the options Mac uses. I'm the same way for the most part. I just happen to love the look of a Cary SLP-05 a
little more. I can get by pretty well with a good bass management system and that preamp. It's one of the few LS preamps I learned to tolerate and add features
without messing with the house sound and variables valve rolling offer. It's one of the few preamps that you can change the flavor pretty easy, with a different
pair of valves here or there. It would be nice if it had a Variable Loudness but with my luck, they would have to add another case to the mix. I'm not sure if I like
a 2 part preamp yet.

I wish someone would make the plugs for different source options on the top with a carousel for us old guys who hate fishing behind a rack/? to add, swap, or try
something else from time to time. Keep the knobs and switches in the bottom case. I suppose it could be swan song project for an older fart like me. Oh well
thinking out loud again.

Regards

 

[ZolaIII]

@OldHvyMec maybe you can find a mini active (digital) mixer with such orientation of inputs and outputs and controls to your needs and use it as preamp of course. Sometimes it's good to think out loud.

 

[TimoJ]

Monoprice HTP-1 preamp has adjustable loudness with two selectable correction curves (official and vintage), bass and treble controls with adjustable corner frequency and 16 channel PEQ. And Dirac in addition to those. All of those can be active and used at the same time.

 

[Jaxx1138]

The Fletcher Munson curves are a set of experimentally determined graphs that show how loud—in dB SPL—a sound at one frequency must be in order to be perceived as equally loud as a sound at another frequency. It is HUMAN sensitivity to different frequency bins. It is not an EQ curve that you change your system response too. That would be adding linear distortion to your sound system.