Does what we hear correspond to what we measure?

[MAB]

I honestly don't really know where you want to go or even what you would like me to admit as an unforgivable fault, but I just think that you are making this little test into a story that is far too important. Certainly a difference in vision linked to the ocean that separates our two continents.

You got it backwards. Your 'little test' has zero chance of determining a preference in BT devices, nearly 100% chance of delivering a wrong conclusion. Give me a break, "It's just a little test." you say. Studies of preference are not 'little tests', I keep saying this, yet you come back over and over like a carnival barker with this 'little test' phrase. And it certainly is not due to different visions or different continents (your experiment is just as problematic in France).

 

[Vintage02]

The total number of voters is still 16, so the sample size may be too small.
What kind of answer would we get if the sample size reached, say, 100?
Or how would the effort and time of 100 members be useful to the Science of How We Hear subforum?

Do we really need to look for an answer?...
I’m just going to repeat myself, this test was only meant to introduce a set of measurements on a Bluetooth receiver later on. I wanted to be less conventional, but it failed.

 

[Vintage02]

You got it backwards. Your 'little test' has zero chance of determining a preference in BT devices, nearly 100% chance of delivering a wrong conclusion. Give me a break, "It's just a little test." you say. Studies of preference are not 'little tests', I keep saying this, yet you come back over and over like a carnival barker with this 'little test' phrase. And it certainly is not due to different visions or different continents (your experiment is just as problematic in France).

Thank you for your clarifications, which have allowed everyone, including myself, to understand that things need to be done seriously.

 

[Timcognito]

It was a trap from the start... so the OP could then come up with a gotcha conclusion.

Put click bate on ASR and let the AI Bot handle it to see how it does.

 

[RexrothPigeon]

It was primarily a starting point for a discussion on a Bluetooth receiver.

That's weird, the title of the OP (and thus the thread) suggests an objective that is uh a bit more categorical than this

 

[nagster]

Do we really need to look for an answer?...

??
That's what you said.

I think that above all, you should take the listening test and you will get your answer.

I’m just going to repeat myself, this test was only meant to introduce a set of measurements on a Bluetooth receiver later on. I wanted to be less conventional, but it failed.

??
... Okay. You will edit the title and post #1. Then you will ask a moderator to close the thread.

 

[Vintage02]

That's weird, the title of the OP (and thus the thread) suggests an objective that is uh a bit more categorical than this

a matter of point of view

 

[Vintage02]

??
That's what you said.

??
... Okay. You will edit the title and post #1. Then you will ask a moderator to close the thread.

The sentence you quote is related to the recordings, not the number of people who took the test. I really didn't think that asking for a simple listening feeling, as anyone can do on any subject, and don't tell me you never have an opinion, would cause so much ink to flow.

 

[antcollinet]

I agree that there are methodological problems, so the results should be viewed skeptically, but I don't think they're worthless. If anything, I think the discussion about sample 1 being louder would have created confirmation bias in its favor.

Why - preference for loudness is not a given for all participants, plus there are other aspects that can negate that (such as the dynamic compression). Your 75% figure is a complete guess with absolutely nothing to back it up.

You can't put guesses onto bad data and expect to turn the bad data into something meaningful.

 

[MaxwellsEq]

Since the first sample was louder, you would expect it to be chosen more often (all else being equal).

There is a generalized opinion that, all other things being equal, where levels are different, the majority of people prefer louder. I'm not sure that opinion is backed by repeatable tests, but even if it is, it's unlikely to be universally true for all listeners.

 

[antcollinet]

There is a generalized opinion that, all other things being equal, where levels are different, the majority of people prefer louder. I'm not sure that opinion is backed by repeatable tests, but even if it is, it's unlikely to be universally true for all listeners.

And then we are back to small sample size

 

[krabapple]

And then we are back to small sample size

we are also back to a test cohort that, I predict, is unusually knowledgeable about how audio can be made to sound.
The average consumer neither knows nor cares.

If so,
- the ASR cohort knows there's more than one way to make audio subjectively louder: e.g. by turning it up, or by compressing its dynamic range
- many are OK with the former but object to the latter, as a matter of principle

Think about how that might affect 'preference test' predictions and results


Anyway, does anyone else hear this thread crying in a small, piteous voice, "please kill me'?

 

[Timcognito]

Anyway, does anyone else hear this thread crying in a small, piteous voice, "please kill me'?

Not the bot conducting it and those who think its real audio intercourse.

 

[mps]

There is a generalized opinion that, all other things being equal, where levels are different, the majority of people prefer louder. I'm not sure that opinion is backed by repeatable tests, but even if it is, it's unlikely to be universally true for all listeners.

This claims that 1 db difference leads to 90% preference (although I don't know whether to trust that), and I assumed only 75% preference for the louder as my null hypothesis and got extremely strongly statistically significant results so the sample size was more than sufficient for rejecting the null hypothesis that there were no preference-affecting differences outside of loudness. Of course, there are big caveats around test methodology, but it's nice to know that sample size isn't one of them (for the question I was interested in).

 

[MAB]

Of course, there are big caveats around test methodology, but it's nice to know that sample size isn't one of them (for the question I was interested in).

Huh?
Sample size is one of the many problems.

 

[Cbdb2]

The sound that a violin bow makes as it is dragged over the strings goes directly from that point to your ears in a live setting.

Well firstly all instruments are directional so a lot of that sound bounces around a huge room before it gets to you. Look up critical distance. So unless your in the first ten rows or the venue is excellent (rare) that little bit of direct sound is usually buried in the reverb and the result is mush. The only live classical concert that Ive heard that sounded better than my hifi was one in a church in Prague where the unamplified string octet and opera singer where I 20' away and I was sitting in yhe middle.

 

[Cbdb2]

There is a generalized opinion that, all other things being equal, where levels are different, the majority of people prefer louder. I'm not sure that opinion is backed by repeatable tests, but even if it is, it's unlikely to be universally true for all listeners.

Well louder some times is too loud. Is it just me or is there a sweet spot for level? Dosnt matter what source I use (other than Spotify etc with volume correction, altough I find that isnt perfect either) I need a volume remote. I set a volume thats perfect and the next song is 2db louder and its too loud.
I cant tell you why. Maybe my ears get overloaded (there not young) maybe something in the room resonates. Im using speakers and amps that can handle the power so I dont think its the gear.

Measureing loudness is not trivial. Freq. content skews any measurement aka Fletcher munson so electronic measurements dont always agree with what people hear.

Comments?

 

[RickS]

Upon review and discission with OP, thread is closed.