Music: how loud is loud? (video)

[ernest16A]

What do you think about applications for iPhone like this called Decibel meter ( here: httрs://decibelpro.app/ ), which ensures that it can detect the exact sound levels of any sound all around you? What would you choose to re-check it's measurements? Maybe you know some other great applications, that may help with sound measurements?

 

[aarons915]

This has been a very informative thread and I admit I'm one of those people who think Amir listens too loud lol, not so much in a way that he's going to go deaf in a short amount of time but I personally think expecting small speakers to play loud full range is asking too much when the vast majority of people will be using subs with those kind of speakers. At the very least I think it's a good idea to to crank them full range and then apply a high pass to see if they are fine with the high pass in place.

I do agree with one poster who questioned the logic of trying to duplicate a live concert in our living rooms. I think almost all live music is too loud and always have some musicians earplugs on my keychain in case I need to use them, I certainly don't want SPLs that are anywhere close to concert or club levels in my living room.

I've always used an SPL meter off amazon (about $20) to measure my levels and assumed I was very low so haven't worried about this much but after checking with the logger in REW with a UMIK I realized my meter was only A weighted and was reading lower than reality. I listened to a song about 3 db louder than I ever normally listen just to get an idea of my max SPL and was a bit surprised at the peaks but the weird thing is I don't perceptively notice any large peaks in this song so it makes me wonder if we should really be worrying about this as far as distortion in our speakers. As many know, distortion in the bass already has to be at very high levels to detect so common sense would tell us that very fast peaks would be even harder to discern. My motivation is purely to high pass my speakers (LS50) high enough to play loud cleanly, I think the whole hearing loss issue should be decided by listening to your body.(Ie if you're ears are ringing or in pain then lower the friggin volume lol) So anyway this is my SPL logger in REW playing a song a bit louder than I ever listen, this is of 3 LS50 (L,C,R) playing with the subs turned off at a 9 foot listening position. This is with them crossed over at 110Hz and they sounded clean with no distortion to my ears at this level. Does this seem loud to others who have measured their max SPL? Just curious what others think.

 

[Aerith Gainsborough]

90 - 95dB(Z) peak is a comfortable volume for most things. 90 is my "go-to" target pre EQ.
Keep in mind that the peaks usually come from deep bass, hence they aren't that loud perceptively.

95dB in the 2-4KHz region would make you turn down the volume pretty much instantly. Quite unpleasant.

 

[hege]

playing with the subs turned off at a 9 foot listening position. This is with them crossed over at 110Hz

So basically you are really showing A-weighted figures, since there is no bass. 75-80dB(A) is good engaged listening. 85-90dB(A) is starting to get quite loud and uncomfy depending on material.

Try enabling bass and should see atleast 5-10dB more..

edit: copypaste my usual figures here too:

Engaged listening: LAeq 75 LCeq 90 LZpeak 105
Relaxed/background listening: -10 all
Rocking out: +10 all
(note: this is with average full range music, not pink noise.. yeah I have slightly tilted up bass)

 

[aarons915]

So basically you are really showing A-weighted figures, since there is no bass. 75-80dB(A) is good engaged listening. 85-90dB(A) is starting to get quite loud and uncomfy depending on material.

Try enabling bass and should see atleast 5-10dB more..

edit: copypaste my usual figures here too:

Engaged listening: LAeq 75 LCeq 90 LZpeak 105
Relaxed/background listening: -10 all
Rocking out: +10 all
(note: this is with average full range music, not pink noise.. yeah I have slightly tilted up bass)

Actually before I used REW I would use an A weighted meter to test my max SPL so I thought I was much quieter than reality, I used that during this same test and it was reading at the LCSMin level for the most part, A-weighted mostly covers 1k-8k, I think you mean C-weighted which does roll off under 100hz. I ran the test again using Z weighted figures just to check but it matches my C weighted values closely. Like I said in the post, my motivation is mostly about the capability of my mains, not hearing damage, that's why I turned the subs off. The level I showed was about the loudest I listen and the speakers didn't sound stressed at all and no audible distortion was really heard. This is important because we see in many of Amir's reviews, most speakers do ok in the 86db distortion test but many smaller ones fall apart at 96db. Now it's possible that the 95db peaks that REW is measuring are inaudible either due to their fast nature or the fact that humans just aren't very sensitive to distortion in the bass but just based on my listening it really seems the LZSmax might be the more accurate figure than LZPeak. I only say that because I can see most modern music varying by 10-13db but I'm just not hearing 25db, that's over a 4x difference in loudness perception. Until some kind of research is established on all of this it seems like our own subjective thoughts are all we have regarding distortion being audible or how high of a frequency to high pass the mains to minimize this distortion.

Side note, you listen much louder than me, about 10-15 db. Not sure if you or I listen at more average levels but it might explain why the LS50 sound fine to me but not to someone that listens as loud as you.

 

[JJB70]

I do worry a little about the possible consequences of the current obsession with crazy high power output of headphone amplifiers. Yes there are outliers but most headphones are pretty easy to drive to high volume and don't need anything like the power outputs becoming common on headphone amplifiers. Quite aside from the consequence of long term exposure to high volume I see potential for people to end up causing serious damage accidentally with the volume knob.

 

[Robbo99999]

I do worry a little about the possible consequences of the current obsession with crazy high power output of headphone amplifiers. Yes there are outliers but most headphones are pretty easy to drive to high volume and don't need anything like the power outputs becoming common on headphone amplifiers. Quite aside from the consequence of long term exposure to high volume I see potential for people to end up causing serious damage accidentally with the volume knob.

Yes, I hope there is not a tendency for people to turn it up louder than they otherwise would just because they think they should if they've gone to the bother of a high output headphone amplifier, or if instead they get influenced that it's somehow safer because otherwise why would they offer all this power. I worked out for my different headphones that I listen at a max possible RMS peak of something like 75-86dB depending on which of my headphones I listen to, which I worked out mathematically from my headphones listed sensitivity combined with what I know my DAC/amp can output combined with manipulating a digital preamp in EqualiserAPO to my normal listening levels whilst doing my music listening tests to work out the levels I listen at. I think it's important to look after your ears.

 

[Spkrdctr]

90 - 95dB(Z) peak is a comfortable volume for most things. 90 is my "go-to" target pre EQ.
Keep in mind that the peaks usually come from deep bass, hence they aren't that loud perceptively.

95dB in the 2-4KHz region would make you turn down the volume pretty much instantly. Quite unpleasant.

Yes and YES!

 

[RayDunzl]

Does this seem loud to others who have measured their max SPL? Just curious what others think.

No.

When I listen "loudly", like when Audio Buddy comes over, 85Lzeq 105dBpeak is common. My EQ is flat, no bass boost over what is on the recording, justified by noting that at that level the Fletcher Munson curve is just about flat.

More is available, briefly (for a few seconds) tried a calibrated drum solo and registered 116.9dBpeak, sounded like a drum kit in the room (been there done that), but it also sounded like someone was taking a baseball bat to the speakers.

 

[aarons915]

No.

When I listen "loudly", like when Audio Buddy comes over, 85Lzeq 105dBpeak is common. My EQ is flat, no bass boost over what is on the recording, justified by noting that at that level the Fletcher Munson curve is just about flat.

View attachment 146748

More is available, briefly (for a few seconds) tried a calibrated drum solo and registered 116.9dBpeak, sounded like a drum kit in the room (been there done that), but it also sounded like someone was taking a baseball bat to the speakers.

Kind of my thoughts as well, the music sounds loud to me but I can tell I'm not rocking out as loud as someone like Amir because I've never even seen the woofers on my LS50 visibly move at all. I also calibrate my system to have flat bass at my loudest volumes, some have commented that it should have a "house curve" but that sounds too tubby to my ears.

 

[hege]

Side note, you listen much louder than me, about 10-15 db. Not sure if you or I listen at more average levels but it might explain why the LS50 sound fine to me but not to someone that listens as loud as you.

Yes there could be many reasons:

- I'm no stranger to clubs and bass since 90's (some tinnitus from early days, but I've used moulded plugs most of my life and hearing is still excellent)
- My room is extremely treated/dry
- I have lots and lots of headroom with very little distortion

Hmm, I should take my UMIK-1 to my untreated living room, which only has my LG plasma tv's internal speakers (which are actually decend).. would be interesting to see how loud I can crank before it sounds too distorted/loud.

 

[hege]

My TV speakers could achieve this at around 3 meters:
LAeq 75 LCeq 80 LZpeak 97

It cuts off hard at 100hz. Clearly straining a bit at those levels, reduce -5dB and distortion is already decend.

So it achieved about my normal listening level, not especially loud.

It's quite clear that SPL figures alone convey nothing about the actual playback quality and physical sensations.

 

[Foulchet]

I have always been pretty sensitive to sound and according to my iPhone or Apple Wtahc mic, I generally set the sound at no more than 65 dB and generally 55-60 is good.
I need more when the sound is of lower quality/not neutral (especially recessed mids). I notices for instance that I set MP3/AAC or non-neutral headphones to a bit higher levels.

Tenet movie in theater was horrible, they put the sound at unbearable levels.

 

[MC_RME]

Out of curiosity last night, I checked my SPL levels playing my upright piano while singing. The slow time average SPL was around 90dB, with peaks up around 112dB. This is all acoustic and a very reasonable level for a musician. I do this for about an hour, 3-4 days a week, and am not deaf. I routinely check my hearing and can still hear reliably up to 16kHz.

Start playing drums instead and this will change...

 

[Sancus]

So anyway this is my SPL logger in REW playing a song a bit louder than I ever listen, this is of 3 LS50 (L,C,R) playing with the subs turned off at a 9 foot listening position. This is with them crossed over at 110Hz and they sounded clean with no distortion to my ears at this level. Does this seem loud to others who have measured their max SPL? Just curious what others think.

Seems pretty reasonable to me. A lot depends on the spectral content of the music and the dynamic range. Whatever you were listening to there has 79ish LCeq and 95LZpeak. That's only 16dB above the average. The music I posted earlier in the thread has a 30dB difference between average and peak(switching weightings confuses things a bit), but the average level is even lower.

I don't have any problems tolerating any number of >100dB temporary peaks but >85dB average loudness starts to bother me after maybe only 10-20 minutes. Not all my music is very high dynamic range, but I obviously want my speakers to be able to handle anything I throw at them.

Now it's possible that the 95db peaks that REW is measuring are inaudible either due to their fast nature or the fact that humans just aren't very sensitive to distortion in the bass but just based on my listening it really seems the LZSmax might be the more accurate figure than LZPeak.

Given how short auditory memory is it's really hard to say. Stuff that you think sounds fine on an LS50 might all of a sudden sound pretty weak if you had a JBL M2 next to it and the ability to swap back and forth in 500ms. And there's not just distortion to be concerned about but compression as well. I agree we don't really know exactly what to measure when it comes to SPL requirements, so I prefer to use LZpeak and if that results in too much speaker well I'm fine with that. Considering I still don't have/feel a need for giant floorstanders or anything I think that approach worked fine.

 

[Genkishi569]

I'm definitely going to put some time aside to watch this video. I've been wondering for the longest time about my listening levels.

I hope I can come away with something actionable especially since my left ear is weaker than my right ear. In an off-putting constantly noticeable kind of way. Which is to say I've got some pretty bad natural channel imabalance.

 

[amirm]

I sit about 2 to 3 meters from speakers I measure. When running 86 dBSPL sweep, I have to wear hearing protection for mid to treble frequencies. Bass however, is barely audible and I could listen to it all day long without hearing protection. So please folks, play attention to what I explained in the video. If you are hearing substantial bass, your volume is far, far more than what some of you are assuming.

 

[Robbo99999]

I'm definitely going to put some time aside to watch this video. I've been wondering for the longest time about my listening levels.

I hope I can come away with something actionable especially since my left ear is weaker than my right ear. In an off-putting constantly noticeable kind of way. Which is to say I've got some pretty bad natural channel imabalance.

The good thing about a natural channel imbalance between your ears - you don't need to correct for that in headphones or speakers because your brain is used living it's life to compensate for that in everyday life, so I wouldn't correct for that imbalance in headphones or speakers. The one caveat to this is if you use a hearing aid in your bad ear, in which case I would wear headphones without the hearing aid and instead add the associated hearing aid boost to the channel of your bad ear.

 

[Deleted member 27948]

I sit about 2 to 3 meters from speakers I measure. When running 86 dBSPL sweep, I have to wear hearing protection for mid to treble frequencies. Bass however, is barely audible and I could listen to it all day long without hearing protection. So please folks, play attention to what I explained in the video. If you are hearing substantial bass, your volume is far, far more than what some of you are assuming.

Amir, have you ever tried to adjust LF so that the loudspeakers (+sub?) sound flat to your ear, on your preferred listening level? Or this is an inherently stupid idea?

 

[amirm]

Amir, have you ever tried to adjust LF so that the loudspeakers (+sub?) sound flat to your ear, on your preferred listening level? Or this is an inherently stupid idea?

That would sound bright. You want the bass at higher level than treble.