IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Louder sounds better, Looking for a citation
Notat
post Mar 8 2010, 00:43
Post #1





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



I think we're all aware that given two audio sources, all things being equal, listeners will overwhelming identify the louder as better. Level matching when doing comparisons is therefore of great importance. Can someone please point me to research where this was discovered or experimentally verified?
Go to the top of the page
+Quote Post
aclo
post Mar 8 2010, 01:09
Post #2





Group: Members
Posts: 132
Joined: 17-December 09
Member No.: 76103



QUOTE (Notat @ Mar 8 2010, 00:43) *
I think we're all aware that given two audio sources, all things being equal, listeners will overwhelming identify the louder as better. Level matching when doing comparisons is therefore of great importance. Can someone please point me to research where this was discovered or experimentally verified?


Do you need a reference demonstrating that changing the loudness changes the perception (apart from the trivial change of sounding louder!), or specifically that it is perceived as "better"?
Go to the top of the page
+Quote Post
Notat
post Mar 8 2010, 03:50
Post #3





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



Specifically that it sounds better or would be the preferred choice. The context here is loudness wars. The reason we have a loudness war is because, all else being equal, louder sounds better.
Go to the top of the page
+Quote Post
Robertina
post Mar 8 2010, 09:21
Post #4





Group: Members
Posts: 1306
Joined: 4-January 09
Member No.: 65169



I think a search for Fletcher–Munson curves could be a suitable starting point.

I came from here, a (probably less serious) website but it explained the topic for me in a more understandable way.

Please respect that I am no audio engineer; if my hint should be wrong I beg a moderator to delete this post.
Go to the top of the page
+Quote Post
Notat
post Mar 8 2010, 14:55
Post #5





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



I hadn't thought of asking why louder sounds better. Your second link speculates that quieter has less bass due to Fletcher-Munson. We all know that more bass is better smile.gif I don't buy it. My anecdotal understanding is that the louder-is-better effect has been demonstrated for level differences on the order of 1 dB - doesn't seem like a big enough difference to bring F-M into play.

Thanks for taking a stab at it. Still looking for a citation.

This post has been edited by Notat: Mar 8 2010, 15:34
Go to the top of the page
+Quote Post
Slipstreem
post Mar 8 2010, 15:22
Post #6





Group: Members
Posts: 966
Joined: 7-July 06
Member No.: 32660



QUOTE (Notat @ Mar 8 2010, 13:55) *
Your second link speculates that quieter has more bass due to Fletcher-Munson.

You're interpreting the Fletcher-Munson curve upside-down. wink.gif
Go to the top of the page
+Quote Post
Notat
post Mar 8 2010, 15:35
Post #7





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



QUOTE (Slipstreem @ Mar 8 2010, 07:22) *
You're interpreting the Fletcher-Munson curve upside-down. wink.gif

Thanks. Fixed that.
Go to the top of the page
+Quote Post
aclo
post Mar 8 2010, 17:52
Post #8





Group: Members
Posts: 132
Joined: 17-December 09
Member No.: 76103



QUOTE (Notat @ Mar 8 2010, 14:55) *
I hadn't thought of asking why louder sounds better. Your second link speculates that quieter has less bass due to Fletcher-Munson. We all know that more bass is better smile.gif I don't buy it. My anecdotal understanding is that the louder-is-better effect has been demonstrated for level differences on the order of 1 dB - doesn't seem like a big enough difference to bring F-M into play.

Thanks for taking a stab at it. Still looking for a citation.


One citation I found in my archive is

Gabrielsson et al., "Perceived sound quality of reproductions with different frequency responses and sound levels."; The Journal of the Acoustical Society of America (1990) vol. 88 (3) pp. 1359-66

and other work by the same author(s). However,
a) they used a tape
b) there is a 6kHz low-pass filter over everything because of their equipment
c) they used 12-but DA and AD converters
d) they looked at the effects of two different sound levels, different by 10dB
e) the subjects filled in a questionnaire on perceived clarity, nearness and this kind of thing

so, it's not what you're after (ie the effects of 1dB differences or thereabouts), they used equipment and methods that opens you up for attack (bad equipment=attacks from audiophiles, no double blind tests=attacks from people who like arguing independently of any information content in the argument).

I couldn't find anything much closer to what you wanted by looking up later work of these authors or work that cites them (but I may have missed lots of things; this is hardly my own area)

So good luck smile.gif

PS: let me know if you can't get hold of the paper
Go to the top of the page
+Quote Post
2Bdecided
post Mar 8 2010, 18:25
Post #9


ReplayGain developer


Group: Developer
Posts: 5142
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



You could ask on the auditory mailing list:

http://www.auditory.org/

It's one of those things that's "just known" (just like no one questions that if you increase the amplitude of higher frequencies, the audio sounds "brighter"!) - but I bet there was some kind of experiment showing the just noticeable level difference vs bandwidth of that difference at some time. It's alluded to (but not quoted) here:

http://www.bostonaudiosociety.org/bas_spea...abx_testing.htm
QUOTE
Meyer handed out a sheet photocopied from the ABX manual which showed typical level-matching required for reliable detection of differences between sources with 1/3 octave frequency-response aberrations. When the aberrations span a wider spectrum, level-matching becomes increasingly critical, dropping to less than 1/3 of a dB especially in the ear-sensitive 2-5kHz region. Acuity (ability to hear difference) also depends sometimes on how close to the threshold of hearing the level of the frequency is. At threshold, a small increase in level will make the sound audible and enable the listener reliably to distinguish A and B when different.


Hope this helps.

Cheers,
David.
Go to the top of the page
+Quote Post
Notat
post Mar 8 2010, 19:32
Post #10





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



Thanks for the suggestion David. There's a separate question as to what's the minimum level difference that is perceptible in an ABX test. I don't think I'd have trouble finding citations for that. My question concerns preference so it would be worked out some sort of blind test but not ABX.
Go to the top of the page
+Quote Post
greynol
post Mar 8 2010, 19:53
Post #11





Group: Super Moderator
Posts: 10001
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Notat @ Mar 8 2010, 05:55) *
My anecdotal understanding is that the louder-is-better effect has been demonstrated for level differences on the order of 1 dB - doesn't seem like a big enough difference to bring F-M into play.

Fletcher-Munson does provide the answer for the general case.

While I familiar with the common claim that most people can simply distinguish a volume change on the order of 1 dB, I don't recall that I've ever seen it coupled with any claim that the small increase will bring improvement.

Aside from your anecdotal understanding, where do you see people assert that louder on the order of 1 dB sounds better?


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
Lynch
post Mar 8 2010, 20:03
Post #12





Group: Members
Posts: 11
Joined: 17-June 06
Member No.: 31933



louder-is-better may as well mean: better because it sells more. what i'm trying to say is, that maybe some research in a more marketing related context might provide some answers as well. i can't point you in a more specific direction, though. i just always assumed that research which could have instigated the loudness wars, came from the marketing buffs.
Go to the top of the page
+Quote Post
greynol
post Mar 8 2010, 20:10
Post #13





Group: Super Moderator
Posts: 10001
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Maybe Notat can clarify, but it doesn't seem that this has to do with the loudness war since mastering something to be louder requires more than simply increasing the volume. To me the use of dynamic range compression doesn't appear to satisfy the criteria in the original post of all other things being equal.


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
Notat
post Mar 8 2010, 21:48
Post #14





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



Nobody I've spoken to or heard speak claims that artists, record executives, radio station program managers or listeners prefer the sound of stereo bus dynamic range compression. What they claim is that they like the sound of louder. "My CD sounds louder than the other CDs in the changer therefore my CD sounds better." If they could, they'd back off on DRC and dictate that listeners turn it up louder than whatever it was that they last listened to.

I admit I made up the 1 dB figure. The level gains that mastering engineers are able to achieve with DRC on what is more frequently an already pre-squashed mix that they receive is on the order of a few dB. Artists and record executives listen to the mastered version in comparison to their final mix and pat these chaps on the back. They've made it louder. They've done other improvements too. The situation is too messy to know what is actually going on. So I was hoping to find someone who has actually studied it.

If it turns out that turning it up sounds better because, due to F-M effects, it fills out the bass, that would be interesting. It would say that, to sound better, music should be mastered with more bass, not necessarily louder. But that's not what's happening. Aggressive DRC we see in the loudness wars invariably skews the energy towards the treble.
Go to the top of the page
+Quote Post
Axon
post Mar 8 2010, 23:58
Post #15





Group: Members (Donating)
Posts: 1984
Joined: 4-January 04
From: Austin, TX
Member No.: 10933



Another angle to consider is the positive response to a stimulus, and how a constant stimulus is sometimes preferred to intermittent stimulus. Basic dynamic range compression helps ensure that the stimulus (the sound) is always audible. Multiband compression helps ensure that all critical bands of hearing are consistently stimulated over time. Limiting/clipping spreads the spectral content out so that more critical bands are stimulated at a given time.

IIRC, extremely loud volumes can also create a dopamine response, which can play a big part in the rock concert experience.

Obviously that's a very pop-sciencey look at the issue. Somebody who has actually studied such effects (*cough* jj) could explain it with far less hot air and handwaving.

This post has been edited by Axon: Mar 8 2010, 23:59
Go to the top of the page
+Quote Post
2Bdecided
post Mar 9 2010, 09:41
Post #16


ReplayGain developer


Group: Developer
Posts: 5142
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



QUOTE (greynol @ Mar 8 2010, 18:53) *
Aside from your anecdotal understanding, where do you see people assert that louder on the order of 1 dB sounds better?
I've experienced that in ABX tests - a small level change, not easily perceptible as a level change in itself (especially with a switching delay), perceived (and ABXed) as "better".

The test was EQ-ing loudspeakers (i.e. frequency response correction) - we could only match the standard and EQ-'d overall level to within 1dB - and it turned out that whichever was the slightly louder (either EQ'd or flat) sounded better.

This is was on rather good speakers that (IMO this test shows) didn't need correction.

I should point out these were single blind tests.

It's hardly the "evidence" you're looking for, which is why I didn't offer it earlier!

Cheers,
David.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 9 2010, 15:20
Post #17





Group: Members
Posts: 3797
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (Notat @ Mar 7 2010, 18:43) *
I think we're all aware that given two audio sources, all things being equal, listeners will overwhelming identify the louder as better. Level matching when doing comparisons is therefore of great importance. Can someone please point me to research where this was discovered or experimentally verified?


I really don't know exactly what *better* means. ;-)

However, there is plenty of evidence that there is an optimal SPL for hearing small details and small changes to sound. It's generally thought to be around 85-90 dB for the loudest parts. Of course there is some ambiguity in that figure - what spectral balance, what sort of dynamics, and all that. Also, I find that people find speech to be more natural when its level is about 10 dB less than music, including singing. Others have noticed this as well.

The basic idea - that sound sources can be too loud or too soft pretty well agrees with "common sense". So the only argument is over what is too loud and what is too soft, right?
Go to the top of the page
+Quote Post
Notat
post Mar 9 2010, 15:39
Post #18





Group: Members
Posts: 581
Joined: 17-August 09
Member No.: 72373



There is an upper limit on useful listening levels. Anywhere below that, if you do back-to-back comparisons, louder program is perceived as "Better".

"Better" means that a listener would identify louder version as a superior version. It is a robust effect. In my experience, you don't have to give very specific rating criteria for the effect to appear. You can ask for "Better", "Accurate", "High-fidelity", "Pleasant", "Engaging". Just don't ask for "Quieter" and you're good.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 9 2010, 15:46
Post #19





Group: Members
Posts: 3797
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (Notat @ Mar 9 2010, 09:39) *
There is an upper limit on useful listening levels. Anywhere below that, if you do back-to-back comparisons, louder program is perceived as "Better".

"Better" means that a listener would identify louder version as a superior version. It is a robust effect. In my experience, you don't have to give very specific rating criteria for the effect to appear. You can ask for "Better", "Accurate", "High-fidelity", "Pleasant", "Engaging". Just don't ask for "Quieter" and you're good.



Of course you realize that this is just an unfounded assertion as presented, right? ;-)
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 24th September 2014 - 02:51