Skip to main content

Topic: Looking for papers on how small level changes are misheard as quality  (Read 2493 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • mzil
  • [*][*][*][*][*]
Looking for papers on how small level changes are misheard as quality
I've often heard it said, by many audio writers I trust, that very small level differences [on the order a 1 dB or even less] are often (if not usually) perceived as quality differences by humans, and not simply level differences. I've always accepted this without question, however in arguing with others in other forums, I need documentation to back this. What I'm asking for people's help with is in finding the original work(s) which demonstrated this, in the form of a scholarly paper(s). Anyone?
  • Last Edit: 01 February, 2013, 04:42:22 PM by mzil

  • julf
  • [*][*][*][*]
Looking for papers on how small level changes are misheard as quality
Reply #1
I've often heard it said, by many audio writers I trust, that very small level differences [on the order a 1 dB or even less] are often (if not usually) perceived as quality differences by humans, and not simply level differences. I've always accepted this without question, however in arguing with others in other forums, I need documentation to back this. What I'm asking for people's help with is in finding the original work(s) which demonstrated this, in the form of a scholarly paper(s). Anyone?


Unfortunately I don't have any pointers to proper papers - but I can confirm that last time I did a blind listening test between different hi-res resolutions, I put in one "+1 dB" control track, and it was the one that (not surprisingly) came out on top.

  • mzil
  • [*][*][*][*][*]
Looking for papers on how small level changes are misheard as quality
Reply #2
Thanks. I have no doubt the phenomenon is real, but I need a scholarly source to cite.

The closest I have found is in an article by veteran audio writer Tom Nousiane. He refers to a test he conducted, but I don't know if it was an informal test or one of his published works:

"For example, we tend to interpret small changes in volume as sound quality. I conducted an experiment several years ago where thirty-one subjects were asked to listen to ten sets of musical passages, with each set containing two 30-second samples.
In half of the sets, both samples were played at precisely identical volumes. In the other half, there was a 1-dB difference in level between them. Although people had a strong tendency to 'prefer' the louder alternative (especially when it came as the second of two), not one of the subjects reported volume or level as a discriminating factor. All comments on how the sound changed were couched in quality terms such as 'cleaner' or 'more harsh' even though volume was the only thing that had changed."

Source:
http://www.nousaine.com/pdfs/Can%20You%20T...Your%20Ears.pdf

He also mentions the same test in this article:
http://www.nousaine.com/pdfs/Smoke%20and%20Mirrors.pdf

I don't have access to reading his published AES papers because I am not a member, nor do I have any academic credentials or connections which would give me free access. At best, all I can get is the abstracts, but I know he has written at least some general papers about audio testing, and is mentioned in some  such as this. If anyone here can access this paper and see if it is the one he speaks of in this quote I've provided, I would appreciate it. Thanks.

edit : AES paper link corrected
  • Last Edit: 02 February, 2013, 12:18:58 PM by mzil

  • Nessuno
  • [*][*][*][*]
Looking for papers on how small level changes are misheard as quality
Reply #3
I casually found a reference to the topic at hand in the discussion to this blog article about (supposed) headphone sound quality modifications after break-in.
At a certain point a poster makes a generic reference to a test made by german speaker manufacturer Nubert (better use your browser search function on that page with "nubert" as a key, as the discussion is quite long). He says they proved that different cables are not ABXable (nothing new here) but an 1dB increase made the same cable appearing a better sounding one.

Anyway a fast Google search returned nothing. Maybe some German speaking fellow here could know (or search for) better.
  • Last Edit: 02 February, 2013, 12:59:11 PM by Nessuno
... I live by long distance.

Looking for papers on how small level changes are misheard as quality
Reply #4
This is just a thesis but still interesting (and free):
http://etd.lib.fsu.edu/theses/available/et...esis-Kimura.pdf

  • Wombat
  • [*][*][*][*][*]
Looking for papers on how small level changes are misheard as quality
Reply #5
I casually found a reference to the topic at hand in the discussion to this blog article about (supposed) headphone sound quality modifications after break-in.
At a certain point a poster makes a generic reference to a test made by german speaker manufacturer Nubert (better use your browser search function on that page with "nubert" as a key, as the discussion is quite long). He says they proved that different cables are not ABXable (nothing new here) but an 1dB increase made the same cable appearing a better sounding one.

Anyway a fast Google search returned nothing. Maybe some German speaking fellow here could know (or search for) better.

I quickly did read some Nubert forum posts and technical papers. One paper called "ts_klangentscheidende-faktoren.pdf" describes the same thing that you already explained in 2 sentences, not more.
It only tells that a 1dB difference to the louder resulted in 'better' or 'more dynamic' felt sound in direct comparison by listeners. There is no info or further linking to that speaker cable test itself i have found.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Looking for papers on how small level changes are misheard as quality
Reply #6
Thanks. I have no doubt the phenomenon is real, but I need a scholarly source to cite.

The closest I have found is in an article by veteran audio writer Tom Nousiane. He refers to a test he conducted, but I don't know if it was an informal test or one of his published works:

"For example, we tend to interpret small changes in volume as sound quality. I conducted an experiment several years ago where thirty-one subjects were asked to listen to ten sets of musical passages, with each set containing two 30-second samples.
In half of the sets, both samples were played at precisely identical volumes. In the other half, there was a 1-dB difference in level between them. Although people had a strong tendency to 'prefer' the louder alternative (especially when it came as the second of two), not one of the subjects reported volume or level as a discriminating factor. All comments on how the sound changed were couched in quality terms such as 'cleaner' or 'more harsh' even though volume was the only thing that had changed."

Source:
http://www.nousaine.com/pdfs/Can%20You%20T...Your%20Ears.pdf

He also mentions the same test in this article:
http://www.nousaine.com/pdfs/Smoke%20and%20Mirrors.pdf

I don't have access to reading his published AES papers because I am not a member, nor do I have any academic credentials or connections which would give me free access. At best, all I can get is the abstracts, but I know he has written at least some general papers about audio testing, and is mentioned in some  such as this. If anyone here can access this paper and see if it is the one he speaks of in this quote I've provided, I would appreciate it. Thanks.

edit : AES paper link corrected


I'd try emailing Tom.