KEF R3 OR BOWER & WILKINS S706 s2 ???

[TrevC]

Getting back to the topic of the thread. Is there anyone on here that has heard both speakers that are being compared?

 

[HarmonicTHD]

Getting back to the topic of the thread. Is there anyone on here that has heard both speakers that are being compared?

I did. But it doesn’t matter as the test was done sighted and my impression is worthless to anyone else as it only applies to me in that room with all my biases affecting the results.

And that’s why I don’t make statements on sound because there is no validity in it without proper control and statistical relevance for it to be also true for a significant other number of people for them to base a purchase decision on it.

Many reviewer and influencers and dealers do just that to sell you the most expensive product. That’s why people get challenged here if they make such claims to provide facts / evidence and not anecdotes / opinions.

 

[TrevC]

I did. But it doesn’t matter as the test was done sighted and my impression is worthless to anyone else as it only applies to me in that room with all my biases affecting the results.

And that’s why I don’t make statements on sound because there is no validity in it without proper control and statistical relevance for it to be also true for a significant other number of people for them to base a purchase decision on it.

Many reviewer and influencers and dealers do just that to sell you the most expensive product. That’s why people get challenged here if they make such claims to provide facts / evidence and not anecdotes / opinions.

You must be hell to go shopping with.

 

[HarmonicTHD]

T

You must be hell to go shopping with.

That what my wife says ;-) I take it as a compliment

Look here this it what we want to avoid that consumers get taken for a ride without having the chance of making an informed decision based on replicable data which are at least to some extent are valid for a significant number of people (read to the part where he tries to return his 4500USD headphone cables).

New Audiophiler who purchased Abyss/XIAudio stack looking for opinions

Hello all, I'm new to the audiophile world, I just started getting into audiophile gear in November. I kinda jumped into this hobby head first, I started off with the HD6XX and a Topping DX3+ Pro. After hearing those I knew I wanted more. I started doing a ton of review watching + learning on...

www.audiosciencereview.com

 

[test1223]

For Harman and my personal experience pink noise is the most discriminating to hear if some frequency regions are exaggerated, mind you the difference to "Female Pop Rock" is very small and probably pink noise needs more training to get used to the correct overall tilt compared to known music:

Source of above image: http://seanolive.blogspot.com/2010/03/method-for-training-listeners-and.html

White noise is perceived with an uprising tonality.

Here more about it from Toole https://www.audiosciencereview.com/...d-can-it-be-measured.25313/page-7#post-866792

White noise / pink noise needs a bit of training for sure. I have heard it many times for speaker measurements. With white noise it is more easy to hear the higher frequencies (obviously). Since the room and speaker position alters everything below 300Hz I wouldn't use pink noise since you didn't get much inside in this lower frequencies with it anyways since you hear mainly the resonances of the room and placement.

I did. But it doesn’t matter as the test was done sighted and my impression is worthless to anyone else as it only applies to me in that room with all my biases affecting the results.

IMHO you overcompensate the fact that it is very hard to impossible to reliably test "small" differences without some kind of blind experiment. Bigger differences like differences of loudspeakers can be much better observed even without a blind test. Yes there might be a bias but you can only be tricked to some extent. Therefore condemning all sighted test is the same nonsense like taken ever sighted test as accurate data.

And Olive later said the music selection is one of the biggest influences and subsequently tried to eliminate the effect on the studies. He then showed that (and also to probably to simplify things) that certain tracks and pink noise are sufficient without compromising the applicability of the studies, which resulted in the “Harman preference curve”, to other music genres (as seen in the paper you linked).

At least that’s how I understood it.

You can see it from two sides. The music selection which is more similar to noise is better at revealing linear frequency differences. Linear frequency differences are the most important parts of the quality differences of typical loudspeakers. On the other had if you use music like the SQAM material you need longer to get the same impression of the linear frequency differences but you also get more inside of other distortion which isn't included in the preference score model and which is especially interesting if the speakers are more similar from the spinorama data.

 

[thewas]

White noise / pink noise needs a bit of training for sure. I have heard it many times for speaker measurements. With white noise it is more easy to hear the higher frequencies (obviously). Since the room and speaker position alters everything below 300Hz I wouldn't use pink noise since you didn't get much inside in this lower frequencies with it anyways since you hear mainly the resonances of the room and placement.

I would consider in that case rather band limited pink noise as with white noise the higher frequencies are perceived stronger.

 

[test1223]

I would consider in that case rather band limited pink noise as with white noise the higher frequencies are perceived stronger.

Yes you can do that. Pink noise might be better to judge one error over another error than using white noise. But you have to have some experience how different good loudspeaker room combinations sound with that exact noise signal.

 

[HarmonicTHD]

Yes you can do that. Pink noise might be better to judge one error over another error than using white noise. But you have to have some experience how different good loudspeaker room combinations sound with that exact noise signal.

Yes. I presume that’s why Olive used trained listeners in his later studies and I think to remember that Amir took one of these trainings (I might be wrong though).

 

[thewas]

There was a reason why I asked you to read up more and not just pick bits and pieces without having read the larger context.

Like for example from the same paper:

The programs were selected on the basis of their ability to reveal spectral and preferential differences between different loudspeakers in over 100 different listening tests and various listener-training exercises.

 

[HarmonicTHD]

Like for example from the same paper:

The programs were selected on the basis of their ability to reveal spectral and preferential differences between different loudspeakers in over 100 different listening tests and various listenertraining exercises.

Yes and some other related paper from Olive said: 70 speakers and some 260 listening tests lead to the „Harman Preference Curve“ plus he also keeps referring back to the initial studies financed by the Canadien Consumer Union before both moved to Harman later and derived the anechoic flat preference.

But I would have to look it up as I am quoting from memory and might be wrong.

@BDWoody There is a whole bunch of OT stuff about Harman studies. If the OP is fine and you have a better place, I don’t mind moving it.

 

[preload]

Here's the answer, btw:

Artist	Track	Music Genre
James Taylor	That’s Why I’m Here	Folk rock
Little Feat	Hangin’ On to the Good Times	Blues rock
Tracy Champan	Fast Car	Folk/Blues rock
Jennifer Warnes	Bird on a Wire	Country rock

You conveniently picked the 5 tracks which include the word rock to support your incomplete picture and misconceptions - again don’t pick the bits and pieces out of context.

Hold on a sec, @HarmonicTHD
Are you saying there were more than 4 music tracks utilized in Olive's research paper that correlated loudspeaker preferences with measurements?
And that I deliberately picked only the tracks that were considered "rock music" in order to mislead people here?

There was a reason why I asked you to read up more and not just pick bits and pieces without having read the larger context.

@HarmonicTHD On multiple occasions, you accused me of deliberately picking only the rock tracks out of the music library used in Harman's study that correlated loudspeaker preference with measurements, implying that I didn't understand the research (and that you did). And you opined that I had an "incomplete picture" and that I needed to "read up."

Below is Table 2 from Olive S, "A Multiple Regression Model for Predicting Loudspeaker Preference Using Objective Measurements: Part 1 - Listening Test Results." JAES 2004.

As you can plainly see, there were, in fact, only 4 tracks used. And these 4 tracks match 1:1 with the 4 tracks I listed previously.
They are all considered rock music.

I think you owe me an apology.

 

[Benedium]

Ok ok so the worst is the best and cheap is good cos human hearing is completely subjective and undiscerning.

 

[HarmonicTHD]

@HarmonicTHD On multiple occasions, you accused me of deliberately picking only the rock tracks out of the music library used in Harman's study that correlated loudspeaker preference with measurements, implying that I didn't understand the research (and that you did). And you opined that I had an "incomplete picture" and that I needed to "read up."

Below is Table 2 from Olive S, "A Multiple Regression Model for Predicting Loudspeaker Preference Using Objective Measurements: Part 1 - Listening Test Results." JAES 2004.

View attachment 259537
As you can plainly see, there were, in fact, only 4 tracks used. And these 4 tracks match 1:1 with the 4 tracks I listed previously.
They are all considered rock music.

I think you owe me an apology.

You keep picking bits and piece which only support your ill conceived belief that Harman (Olive) studies are based on “rock” songs and ergo the studies only apply to rock music (did you even read, let alone understand the paper and graph @thewas linked for you).

I asked you multiple times to look at the entire work of Toole, Olive and the conversation with @thewas today gives numerous hints that not only other genres were used but also (pink) noise (even the audioscience link you got from thewas and linked to me included pink noise) and your claim is not supported by the publications.

I can’t apologize for your continued deliberate selective perception and I am wasting my time here doing your homework.

I am out. Please return the favor and put me on ignore too.

 

[preload]

You keep picking bits and piece which only support your ill conceived belief that Harman (Olive) studies are based on “rock” songs and ergo the studies only apply to rock music (did you even read, let alone understand the paper and graph @thewas linked for you).

I asked you multiple times to look at the entire work of Toole, Olive and the conversation with @thewas today gives numerous hints that not only other genres were used but also (pink) noise (even the audioscience link you got from thewas and linked to me included pink noise) and your claim is not supported by the publications.

I can’t apologize for your continued deliberate selective perception and I am wasting my time here doing your homework.

I am out. Please return the favor and put me on ignore too.

Apology accepted.