Blind Listening Test 2: Neumann KH 80 vs JBL 305p MkII vs Edifier R1280T vs RCF Arya Pro5

[Postlan]

I'm a mixing engineer, but sorry, I don't understand what you mean... If you mean the individual instrument balance can't be changed at the mastering stage, it is partially correct, but not entirely.

 

[dasdoing]

I'm a mixing engineer, but sorry, I don't understand what you mean... If you mean the individual instrument balance can't be changed at the mastering stage, it is partially correct, but not entirely.

what I mean is that this is the job of the mixer. it's not the job of mastering to make individiual instruments louder or quieter

 

[fineMen]

I have a great deal of ...

However, recent mastering has been adapting to more flat speakers compared to 20 years ago, and someday flat speakers may become the standard as time passes. Also, there are differences depending on the source. There are bright sources like Norah Jones that are often used as a reference for Audiophile's listening tests. If such sources were used in this test, the results would have naturally been different. However, this is not the circle of confusion or anything like that. It's simply a fact that there are brighter sources, and there's no need to worry that your environment is too bright, you just need to recognize that the source is bright. We should not treat the source as the word of God.

... wow! So much appreciated! And no 'likes'?! It's childish, yes, but still ... Thank you so much for all the information!

 

[thecheapseats]

The good side though is that is more relevant for the bass region as above humans tend to adapt and hear "though" the room and mainly the direct sound.
Bass always needs EQ to the room but from the other side as Toole recommends it is easy to have a variable bass shelving filter at playback for differently well mixed recordings.

I understand - however I view the discussion as how it applies to recording and mixing recordings - not listening after the fact...

 

[thecheapseats]

I have a great deal of respect for Dr. Toole, but he is not entirely accurate about the circle of confusion.

Commercial music is mastered with the assumption that it will be played in a home listening environment, and is EQ'd during mastering stage for this purpose. While all mastering environments are not the same (for example, at Sterling Sound, the acoustics and speakers are significantly different between Ted Jensen's room and Greg Calbi's room.), mastering studios generally have rooms with acoustics similar to a typical home environment. I would say the mastering environment is similar to that of a listening room in a high end audio shop. The same goes for speakers.

The master tape sent to the mastering studio is completed in the recording studios. Recording studios are designed to have as flat room acoustics and monitors as possible. It is very different from domestic environment. (In recent music productions, mixing engineers have often taken on the final adjustments traditionally done in mastering studios. For example, Serban Ghenea does mastering during the mixing stage. However, his studio is not a typical professional recording studio but rather closer to a mastering studio.)

Neumann are products designed for recording studios and are not intended for mastering (even if they claim so). The speakers used for mastering are usually the audiophile speakers, and for example, Ted Jensen was using B&W800D. B&W800D is a consumer product (even if they claim it's for professionals), and its response is clearly not flat. B&W800D has a BBC dip, and its bass and treble are slightly smiling. This is the typical frequency response of an audiophile speaker, as can be seen in Stereophile.

In other words, commercial music is assumed to be played back in non-flat home environments, and there is no circle of confusion. This is simply a result of historical circumstances, and no matter how much technology advances to make household speakers flat, it cannot fundamentally change the issue of compatibility with recordings that have been sold for over 50 years in the past. No one would be foolish enough to listen to vinyls without RMAA correction just because they hate the correction, but trying to listen to a source that was not originally intended for a flat environment in a flat environment may be wrong. In fact, there are likely many people who intuitively feel that flat speakers are a bit unbalanced.

It's not a matter of science, but rather a matter of history.

However, recent mastering has been adapting to more flat speakers compared to 20 years ago, and someday flat speakers may become the standard as time passes. Also, there are differences depending on the source. There are bright sources like Norah Jones that are often used as a reference for Audiophile's listening tests. If such sources were used in this test, the results would have naturally been different. However, this is not the circle of confusion or anything like that. It's simply a fact that there are brighter sources, and there's no need to worry that your environment is too bright, you just need to recognize that the source is bright. We should not treat the source as the word of God.

all very accurate... in fact it was a mastering engineer who recommended the purchase of my first pair of k+h/Neumann mid-field speakers over forty years ago (one of his pairs in his mastering room were k+h) - which is where I heard the brand for the first time... on my 4th pair today and use them as well as a pair of genelecs (on my third pair of those)...

over the years since that first purchase, friends would ask about speakers I use - and if they might be good for their homes... my standard response was always - "you won't like them - they're very flat"...

 

[thecheapseats]

I'm a mixing engineer, but sorry, I don't understand what you mean... If you mean the individual instrument balance can't be changed at the mastering stage, it is partially correct, but not entirely.

what I mean is that this is the job of the mixer. it's not the job of mastering to make individiual instruments louder or quieter

you're both on the same page - mostly...

 

[MatthewS]

I would love to see the in-room meassurements unsmoothed,

The REW files with the raw data is in the original post.

 

[Floyd Toole]

EDITED:

B&W speakers are extensively used in recording studios, often classical music biased, but not always. The topic has been talked to death on this site in the thread:

"Which speakers are the Classical Music Pros using?"​

Injecting some hard data into the conversation here is a spinorama on the 800D. The so-called BBC dip (which was not in the design specs for their own named monitors) is evidence here of questionable design: a rather large midrange becoming directional before crossing over to an unbaffled - very widely dispersing - tweeter at an uncommonly high frequency. The result could be predicted from visual inspection. The lower figure is Figure 12.3 in the 3rd edition of my book, showing a newer model 802D, with similar behavior. I don't see a "smiley face", just a sagging midrange in the 800D and very uneven on-axis frequency response and directivity vs. frequency in both.

response and directivity vs. frequency in bo

th

.

 

[fineMen]

Nobody said the B&W would follow a 'flat' paradigm. To the contrary, the D123 was said to resemble typical high-end speakers. If that suffices to produce one hit after the other, what exactly is wrong with it?

I'm still under the impression that the music program here wasn't chosen with a lucky hand. Please, once in a while listen to records with natural reverberation (reportedly recorded in airport hangars, The Supremes). Nowadays every (!) of those 'female voices' is doubled, tripled up in the recording, see the selection in post #1. The Delfonics, Nina Simone ... ? Nobody seems to (honestly) care why it is that the most synthetic most popular product is the most revealing. Which in my understanding means, that it brings more people to agree on the same 'better' speaker.

Provide some DIY recordings, mono without alterations, except for adjusting to predicted head related transfer function. If the target is 'flat' that would be the way to go.

Try to not only make a low ranking speaker better w/ equalization. Try the other way round, the higher ranking is trimmed to appear as bad as the lower, objectively. Ask if the latter is still recognized by what remaining virtues.

I know the winner of this test all too well. It reveals it's deficiencies not that easily. But on the long run they can't be missed when listening to music that means something. To be understood, I'm not at all a so called subjectivist.

 

[Blumlein 88]

science says otherwise https://en.wikipedia.org/wiki/Precedence_effect

you can easily test this with two speakers. send them the same signal, but delay one of those a few ms. you will see that you ear wont filter out the later one, but treat both as a single signal, but shifted in space

From wiki:

The "law of the first wavefront" was described and named in 1948 by Lothar Cremer.[3]

The "precedence effect" was described and named in 1949 by Wallach et al.[4] They showed that when two identical sounds are presented in close succession they will be heard as a single fused sound. In their experiments, fusion occurred when the lag between the two sounds was in the range 1 to 5 ms for clicks, and up to 40 ms for more complex sounds such as speech or piano music. When the lag was longer, the second sound was heard as an echo.

So a few milliseconds for clicks and longer for other more complex sound before an echo is heard. Delay in one channel below the point where echo is heard causes a shift in image. Something like a 15 msec difference is enough to shift where we hear stereo all the way over to one side. Yet we aren't hearing echoes (which is what they are). What you are describing in your test is within the zone where it is not heard as an echo. Similarly in a room the direct sound from a speaker will be heard and this effect filters out echoes until they get rather long. So a zone of time where we hear direct mostly, a transition zone where it merges location perceived (though not an echo or perceived reflection) and finally a zone where echoes are perceived.

It makes sense because in a smaller space this allows more precise location of a sound vs if this filtering did not occur. That stereo works is something of a happy accident.

 

[Tom C]

The battle never ends…

 

[Thomas_A]

Nobody said the B&W would follow a 'flat' paradigm. To the contrary, the D123 was said to resemble typical high-end speakers. If that suffices to produce one hit after the other, what exactly is wrong with it?

And then there were the B&W Matrix series 2, 15 years in production? It was after that B&W departed from the on-axis flat frequency response.

 

[fineMen]

The battle never ends…

Better be it so, that's what science is about, but without the 'battle'. (didn't count the 'ad hominems', calling for authorities etc)

I appreciate the (spinorama) standard, and also the requirement for 'flat'. What I'm missing a tiny little bit is to cross-check the validity of the 'preference' assessment. Why replicate the test methodology? Let's challenge it! Science.

Add.: fruitless, why not? We as humans are quite competent in detecting the tone of a voice, decoding tiny little cues in the articulation. It gives us the idea to understand the speaker (we might be wrong, but ..). With the 'musical' test program used here there is no such thing to begin with. All what I feel to be of that sort is replaced by nasty sticky studio technology (synth/ rrreverb/, doubling up voices etc). I don't feel a voice, but some brutal substitute. It is a matter of the mind, not the brain. Brown/pink/white noise doesn't have articulation by definition. Some 'music' neither. But those two are the most 'revealing'. I only ask, what does it reveal? The 'flatness'? So the test reveals that 'flat' is prefered over every other parameter? Is it so because every other parameter was ruled out by the test set-up?

The winner here I know very well. When listening I quickly feel a bit stressed. There's something missing that I get from my other speaker. It's defiencies interfere with those cues mentioned above, resonances, distortion, intermodulation, tons of? It doesn't feel real to my mind--not brain, I have none.

Cherio

 

[bodhi]

Better be it so, that's what science is about, but without the 'battle'. (didn't count the 'ad hominems', calling for authorities etc)

I appreciate the (spinorama) standard, and also the requirement for 'flat'. What I'm missing a tiny little bit is to cross-check the validity of the 'preference' assessment. Why replicate the test methodology? Let's challenge it! Science.

Is the word preference itself bothering you?

Sometimes in these discussions people seem to interpret it to mean that if some speaker gets a higher preference score than another then everyone should prefer the former. I have always thought that it's trying to give as good as possible info to compare two speakers. And in your mind you must read it like "in these conditions, with this kind of material, statistically people seem to prefer speaker A to speaker B".

If this is the case wouldn't it be useless to debate the accuracy of current research as it is not claiming anything about how accurate this score is if and when the conditions change? So if you would like to test some other preference calculcation you would have to re-create the research with different conditions and see where you arrive at.

 

[fineMen]

Is the word preference itself bothering you?

Sometimes in these discussions people seem to interpret it to mean that if some speaker gets a higher preference score than another then everyone should prefer the former. I have always thought that it's trying to give as good as possible info to compare two speakers. And in your mind you must read it like "in these conditions, with this kind of material, statistically people seem to prefer speaker A to speaker B".

If this is the case wouldn't it be useless to debate the accuracy of current research as it is not claiming anything about how accurate this score is if and when the conditions change? So if you would like to test some other preference calculcation you would have to re-create the research with different conditions and see where you arrive at.

Sorry, no it is not the word, but the methodoloy. Preference is decidedly unspecific. I gave an example in the edit of my post #317. Sorry again.

 

[test1223]

EDITED:

B&W speakers are extensively used in recording studios, often classical music biased, but not always. The topic has been talked to death on this site in the thread:

"Which speakers are the Classical Music Pros using?"​

Injecting some hard data into the conversation here is a spinorama on the 800D. The so-called BBC dip (which was not in the design specs for their own named monitors) is evidence here of questionable design: a rather large midrange becoming directional before crossing over to an unbaffled - very widely dispersing - tweeter at an uncommonly high frequency. The result could be predicted from visual inspection. The lower figure is Figure 12.3 in the 3rd edition of my book, showing a newer model 802D, with similar behavior. I don't see a "smiley face", just a sagging midrange in the 800D and very uneven on-axis frequency response and directivity vs. frequency in both.

response and directivity vs. frequency in boView attachment 279048thView attachment 279052.

What are your thought on some facts which imply that a flat /smooth frequency /sound power response is not always ideal for a good loudspeaker.

There is the different perception of loudness of diffuse sound compared to directional sound due to the head related transfer functions.

But we only have one stereo mix. Therefore you potentially have to compensate differences which occur due to more or less diffuse sound compared with the amount of diffuse sound at the recording studio. This differences are most of the time greater if you have a shorter or longer listening distance and a more or less omnidirectional speakers. And there also might be an effect that room acoustic cues of the recording provide the wrong tonality due to most likely less diffuse sound in the speaker play back system.

And there are some other difficulties due to only one stereo mix. There is the dependency spl and loudness.

So you potentially also have to compensate the difference in playback spl at home vs. at the mastering studio.

There is also a difference in tonal perception from a spherical, cylindrical and flat wave front due to the HRTFs and some other potential smart to use compensation which introduce some peaks and dips.

So I think it is to strickt to dismiss every loudspeaker as not good which has some smaller like +/-2dB deviations from a flat frequency response or sound power response. It is of cause extra hard to tell if these deviations are good (for your specific case) or occur there by bad engineering.

I would like to hear your thoughts on this.

 

[thewas]

What are your thought on some facts which imply that a flat /smooth frequency /sound power response is not always ideal for a good loudspeaker.

There is the different perception of loudness of diffuse sound compared to directional sound due to the head related transfer functions.

But we only have one stereo mix. Therefore you potentially have to compensate differences which occur due to more or less diffuse sound compared with the amount of diffuse sound at the recording studio. This differences are most of the time greater if you have a shorter or longer listening distance and a more or less omnidirectional speakers. And there also might be an effect that room acoustic cues of the recording provide the wrong tonality due to most likely less diffuse sound in the speaker play back system.

And there are some other difficulties due to only one stereo mix. There is the dependency spl and loudness.

So you potentially also have to compensate the difference in playback spl at home vs. at the mastering studio.

There is also a difference in tonal perception from a spherical, cylindrical and flat wave front due to the HRTFs and some other potential smart to use compensation which introduce some peaks and dips.

So I think it is to strickt to dismiss every loudspeaker as not good which has some smaller like +/-2dB deviations from a flat frequency response or sound power response. It is of cause extra hard to tell if these deviations are good (for your specific case) or occur there by bad engineering.

I would like to hear your thoughts on this.

I have no problem to accept such deviations when their result is supported for example from corresponding literature or controlled listening tests, the problem though is that most loudspeaker companies and fans use such as excuses without any such foundation.

 

[bodhi]

Sorry, no it is not the word, but the methodoloy. Preference is decidedly unspecific. I gave an example in the edit of my post #317. Sorry again.

It is very specific. The methodology is well described and we have to assume that it was followed precisely. Not having read the book I believe there was some serious thought put into the choice of test material so that it would give as accurate results as possible also with different material.

What I'm deciphering from your explanation is that as you don't listen to same kind of music that was used in the research the preference score is not accurate, for you. As you have proven yourself, that is indeed the case. I'm just failing to see how this could be seen as a flaw in the methodology.

 

[fineMen]

What I'm deciphering from your explanation is that as you don't listen to same kind of music that was used in the research the preference score is not accurate, for you. As you have proven yourself, that is indeed the case. I'm just failing to see how this could be seen as a flaw in the methodology.

You're correct, of course I don't listen to such material! Would I crave for a good stereo otherwise? And as I said, it seems to be a fruitless attempt to discuss "the book" and the methodology of 'preference test'. It's on everbody to decide if such a situation defines 'scientific'. From my education I'm empowered to call myself a scientist with all the paperwork. You see, I'm still on that very topic. But how long? I'm afraid I'm starting to bore people

 

[Postlan]

you're both on the same page - mostly..

Interesting, but I don't know him. Contact amir