Who would buy a speaker without listening to it?

[YSC]

That isn't the same thing as I suggested. I said to have people who are confident in their abilities to rank speakers based on measurements, without knowing which speakers are which, then they listen (blind) and see whether the ranking on listening follows the ranking on measurements alone. The correlation (or lack of) is the interesting part, not the rankings in and of themselves.

I personally this kind of test have missed the point, if say, you compare something measuring obviously poor, say with some wildly varying on axis FR, and then comparing it to something good to great, say the Neumann or Genelec base models, it's highly likely that measurement focused ppl can discern the good vs bad, as with the preference score generating studies, which include layman and those pros/measurement geeks in the mix.

But then when you are comparing something within same performance bracket, any room interaction could mask the minute difference, or say side reflections can vary the ended up correct identification of say KH120 vs 8030. point is that say I enjoy both KH120 and 8030, with difference preference in different setup, measurements can make you know you would get something you like, not something being your holy grail, love and ranked 1st all the time. for well measured speakers more often than not (at least in my very limited exp) comes to which digs deeper or looks better for one's preference, although it might not be the one I definitely love no.1, the purchase would not disappoint, while auditioning in stores or friend's home, or even with different mental status, usually results in something you buying thinking it's great, yet sounds meh at home, then you go upgrade this and that

 

[davlynam]

I purchase equipment when: it comes from a reliable manufacturer/source; it has good reviwes from trustworthy reviwers; the seller has a liberal return policy.

 

[Digby]

point is that say I enjoy both KH120 and 8030, with difference preference in different setup, measurements can make you know you would get something you like, not something being your holy grail, love and ranked 1st all the time.

There is a decent (EQ matched) video on Youtube, which shows how much difference there is in the sound (at least, as captured at stereo microphones) between a KH80 and a 8020D. The difference is not insubstantial. It is likely enough difference to garner a preference. My preference would tilt strongly towards Neumann. Neumann speakers likely have one 'type' of sound (the designs are relatively similar, so easier to say than for Genelec) and the Genelec 80*0 series, another.

Genelec have so many different designs of speakers, that I doubt a particular sound 'presentation' translates across the entire range. While I personally don't favour the 80*0 series, I may well enjoy other types of speaker they manufacture.

although it might not be the one I definitely love no.1, the purchase would not disappoint, while auditioning in stores or friend's home, or even with different mental status, usually results in something you buying thinking it's great, yet sounds meh at home, then you go upgrade this and that

Sure, I take your point, but if a person is so 'easy come, easy go' as to accept any well measuring speaker, even though many have markedly different audio 'presentations', then I can't understand why they would spend so much money/time/energy on the hobby in the first place.

I think we have two different approaches/dispositions at play.

1) Measurements are a tool you bend to your purposes, which is essentially understanding what you prefer best in the world of speakers, all speakers being a compromise of one sort or another. There is no perfect speaker, any imperfections will be more or less pronounced depending on the individuals hearing said speaker. This is why we have such variety in speaker designs at the top, although they are largely all seeking the same goal. The preferences of those buying are not the same, so there is not, and cannot, be one singular, holy grail loudspeaker.

2) Any well measuring speaker, of which there can be vastly different types (2 way vs 3 way, horn tweeters vs AMT, narrow directivity vs wide, small satellites & sub or large floorstanders/large floorstanders + sub) and sound 'presentations', are all essentially as good as each other, more or less, and it is one's preferences that should be bent to the will of all these different types of speaker.

Even if we are to take the position that the stereo recording, with all its loss of sound from the actual musical event, and all the manipulations that have occurred from recording through to engineering, mastering and production, is the grail (accuracy to the stereo recording), then why is precious little said about which speaker is the absolute best in this regard?

Is this because position 2 is incomplete. The picture can only be completed with subjectivity, that of individual preference. It is impossible to have a 'total science' of audio reproduction, because the last piece of the puzzle are the ears on the head of the one listening to said stereo reproduction.

We have preferences in musical styles, some strongly detest what others love, it follows that we would also have preferences in types of speaker 'presentation'. We don't wilfully listen to music we dislike, why would we do similarly with speakers that, however good the measurements, do not satisfy us the most upon listening?

 

[Digby]

Tests and measurements will tell you which messenger is the best.

So which speaker, out of all known speakers, is the best/most accurate? This a question that has yet to be answered. My position is that it can't be answered conclusively, and the position of those who say measurements are 'all that matters' seems to be....it can't be answered.

At least we agree on something, but I am left scratching my head as why a most accurate speaker cannot be pinpointed from all the measurements and data available. If one removes preference from the equation, then you are left only with measurements. What is the difficulty?

 

[Kal Rubinson]

At least we agree on something, but I am left scratching my head as why a most accurate speaker cannot be pinpointed from all the measurements and data available. If one removes preference from the equation, then you are left only with measurements. What is the difficulty?

The concept of measured accuracy is based on the measurement of multiple parameters, many of which are dependent on external factors, e.g., the space in which a speaker is measured, and not on any single measurement. To combine all these one must apply weighting to the multiple parameters and there is no universally accepted scheme for that.

How would you weight the value of "flatter frequency response" against "lower distortion?"

 

[Digby]

The concept of measured accuracy is based on the measurement of multiple parameters, many of which are dependent on external factors, e.g., the space in which a speaker is measured, and not on any single measurement. To combine all these one must apply weighting to the multiple parameters and there is no universally accepted scheme for that.

It seems very complicated even without personal preference in the mix.

How would you weight the value of "flatter frequency response" against "lower distortion?"

With some difficulty, I imagine. I suppose it all depends on what kind of distortion and it's audibility vs how much frequency variation and where in the spectrum. Certain individuals may be more or less annoyed by one fault or the other, which is why I think subjectivity should not be removed from the process of speaker selection, unless one is willing to have less optimised, by way of less personalised, results.

 

[YSC]

So which speaker, out of all known speakers, is the best/most accurate? This a question that has yet to be answered. My position is that it can't be answered conclusively, and the position of those who say measurements are 'all that matters' seems to be....it can't be answered.

At least we agree on something, but I am left scratching my head as why a most accurate speaker cannot be pinpointed from all the measurements and data available. If one removes preference from the equation, then you are left only with measurements. What is the difficulty?

because, even if an imagininary speaker goes with full range 20-20khz on axis dead flat, without directivity mismatch, the width of directivity, distortion and optimum listening distance will be different, and after attaining some sort of good enough quality, it can't be said for one is definitely the best, as in different room, even the same person can prefer wide/narrow directivity, and summation of sound between drivers can have a big part on preference. furthermore, even say the stand position is fixed, different speaker design have different in cabin size and shape, so the driver to front/side wall, ceiling and floor will be different, which in turn affect the SBIR nulls and peaks.

the later is what I feel is most useless for the EQ level matched comarison, you have an uncalibrated mic, and in a room with acoustics of that particular room, same speaker, same setting in your room will sound very different, which makes that audition more or less useless.

that also applies to why I say measurement gives you the idea of "good enough" for your desire, so one could opt for the option that fits best and thing like REW room simulation could give you an idea of expected SBIR, sure if there are free return police and you can try the narrowed down choices in your space for the final bit extra confidence of choosing the best speaker for you it would be perfect, but more often than not that is not the case, and also, in a case where you would want to put in extra money for option B over option A or not. For my own case of KH120 vs 8030C, I decided that they will both make me happy, and the desk footprint for me is small, so the 8030 with class D amp which don't run as hot, and have the isopod allowing me to tilt them towards my ears, plus the slightly smaller footprint allows me to put them with a few extra cm of freedom gets the vote.

audition IMO only matters when the audition is in your own place, own setup, coz that way you have the fixed room effects in play, measurements done 95%+ of decision making considering extension, neutralness, distortion, directivity and SPL capability, audition only helps you to decide between say, top 3 of your own choice and get the best one out. more importantly it makes the educated guess of what you want, so you can then focus on things like connectivity, extra functions, price, WAF etc.

 

[Kal Rubinson]

Certain individuals may be more or less annoyed by one fault or the other, which is why I think subjectivity should not be removed from the process of speaker selection, unless one is willing to have less optimised, by way of less personalised, results.

I see no value in a personalized "subjective accuracy" apart from garden-variety personal preference.

 

[YSC]

It seems very complicated even without personal preference in the mix.


With some difficulty, I imagine. I suppose it all depends on what kind of distortion and it's audibility vs how much frequency variation and where in the spectrum. Certain individuals may be more or less annoyed by one fault or the other, which is why I think subjectivity should not be removed from the process of speaker selection, unless one is willing to have less optimised, by way of less personalised, results.

I agree with this, subjective preference is in the mix, but then after a few demo in a well treated room of a few speakers one could easily read the graphs and know what you prefer or not, but since this is personal for everybody, it's meaningless for the classification or grading a product, same goes for cars, cloths, food, or even marriage, don't dare jump into prefernce of a person, but say for cars, if all important parameters are listed, one could easily choose which car they likely won't regret buying, say I love heavy and direct feedback steering, heavy clutch, or I prefer a comfort, sofa like luxury car with is as effortless as possible.

 

[symphara]

because, even if an imagininary speaker goes with full range 20-20khz on axis dead flat, without directivity mismatch, the width of directivity, distortion and optimum listening distance will be different, and after attaining some sort of good enough quality, it can't be said for one is definitely the best, as in different room, even the same person can prefer wide/narrow directivity, and summation of sound between drivers can have a big part on preference. furthermore, even say the stand position is fixed, different speaker design have different in cabin size and shape, so the driver to front/side wall, ceiling and floor will be different, which in turn affect the SBIR nulls and peaks.

the later is what I feel is most useless for the EQ level matched comarison, you have an uncalibrated mic, and in a room with acoustics of that particular room, same speaker, same setting in your room will sound very different, which makes that audition more or less useless.

that also applies to why I say measurement gives you the idea of "good enough" for your desire, so one could opt for the option that fits best and thing like REW room simulation could give you an idea of expected SBIR, sure if there are free return police and you can try the narrowed down choices in your space for the final bit extra confidence of choosing the best speaker for you it would be perfect, but more often than not that is not the case, and also, in a case where you would want to put in extra money for option B over option A or not. For my own case of KH120 vs 8030C, I decided that they will both make me happy, and the desk footprint for me is small, so the 8030 with class D amp which don't run as hot, and have the isopod allowing me to tilt them towards my ears, plus the slightly smaller footprint allows me to put them with a few extra cm of freedom gets the vote.

audition IMO only matters when the audition is in your own place, own setup, coz that way you have the fixed room effects in play, measurements done 95%+ of decision making considering extension, neutralness, distortion, directivity and SPL capability, audition only helps you to decide between say, top 3 of your own choice and get the best one out. more importantly it makes the educated guess of what you want, so you can then focus on things like connectivity, extra functions, price, WAF etc.

It‘s always nice to have measurements, of course.

But if we were to restrict our speaker choice to stuff that’s measured, that would hugely restrict our choice, which is already restricted by space, budget, type of speaker, looks etc. Plus, it’s not like there’s some gold standard of measurement or central repository.

 

[Robh3606]

The concept of measured accuracy is based on the measurement of multiple parameters, many of which are dependent on external factors, e.g., the space in which a speaker is measured, and not on any single measurement. To combine all these one must apply weighting to the multiple parameters and there is no universally accepted scheme for that.

How would you weight the value of "flatter frequency response" against "lower distortion?"

Hello Kal

From reading Toole and the Geddes distortion metric I would say flatter frequency response.

Rob

 

[symphara]

Hello Kal

From reading Toole and the Geddes distortion metric I would say flatter frequency response.

Rob

Speaker coloration can be somewhat EQed (or you might just like it) but distortion cannot.
I would also rate a good bass response, lack of resonances or port chuffing above flat FR.

 

[DonR]

I bought my Energy RC-50's after listening to them in the showroom for about 30 minutes. They sounded completely different when I set them up at home and it took several hours of repositioning before I was satisfied with them.

 

[Kal Rubinson]

Hello Kal

From reading Toole and the Geddes distortion metric I would say flatter frequency response.

Rob

Me, too, but if you are balancing really awful distortion against moderately flawed (and correctable) FR, how would you weight them? It's not absolute.

 

[Kal Rubinson]

Speaker coloration can be somewhat EQed (or you might just like it) but distortion cannot.
I would also rate a good bass response, lack of resonances or port chuffing above flat FR.

There you go. Multivariate.

 

[killdozzer]

That isn't the same thing as I suggested. I said to have people who are confident in their abilities to rank speakers based on measurements, without knowing which speakers are which, then they listen (blind) and see whether the ranking on listening follows the ranking on measurements alone. The correlation (or lack of) is the interesting part, not the rankings in and of themselves.

And again, it points out how little you know.

You're trying to criticize knowledge from the position of non-knowledge. It makes no sense. But I get it. You could probably buy speakers based on measurements alone if Amir published measurements for microdynamics, right?

 

[killdozzer]

You're looking at the problem from the wrong angle. Standards are not codified to determine which is "best". Codified standards exist to provide a reference for the examination of a DUT. In other words, to what degree does this device conform to the standards? To what degree does it not conform to the standards? To which section of the standards does it comply, and to which does it not comply? Those characteristics are shown in the tests and measurements, and as a group, they show us the nature of the DUT.

They aren't meant to show us which device is "best", although they may incidentally do that. They simply give us a clear picture.

Jim

Oh, tell me about it. Not one wrong angle. He is even mentioning horn and dome and ribbon (????). Talking about apples to oranges. If there's one minimum you should abide by, it would be comparing speakers with the same array. Best bookshelf can't compete with best tower. Ah, yes, bookshelves are toys for him, I forgot. And the only scientists he'll have are the ones with big, fat, huge speakers.

 

[Pearljam5000]

In general it's a mistake
No one told me KH120 would sound dark (to me) and they did.
And i bought them without hearing them.
Them when i listened to Genelec i discovered i like "their sound" more
You can read a million reviews and measurements, no one can tell 100% how will they sound to you.
On the other hand if you know the
"house sound" of a certain brand then its pretty easy and predictable.

 

[ahofer]

So yes, it’s multivariate. But I see @Digby ’s proposition as a kind of predictive “threshold of audibility” question for dispersion and frequency response measurements.

If we included a bunch of speakers that were, say, above 5 on the preference rating, but presented with different flaws - sound power humps in different places, wider/narrower dispersion, etc. How accurately could we armchair critics translate the measurements into identification of the speakers? Surely it is a skill to turn the visual measurements into an idea of sound, and I suspect I’d have a tough time. But I bet others who work with those measurements more regularly (@amirm ?, Erin?) could do pretty well.

 

[Digby]

In general it's a mistake
No one told me KH120 would sound dark (to me) and they did.
And i bought them without hearing them.
Them when i listened to Genelec i discovered i like "their sound" more
You can read a million reviews and measurements, no one can tell 100% how will they sound to you.
On the other hand if you know the
"house sound" of a certain brand then its pretty easy and predictable.

Precisely. Is your assessment wrong? No, it is your assessment. A pair of speakers for your use, should have character that is pleasing to you. Could you have determined they were 'dark' from the measurements alone? I doubt it, but a few minutes listening to them first would likely have identified this trait reasonably quickly, certainly if listened to alongside comparable Genelec speakers.

These are two manufacturers of speakers that are considered more or less on par with each other, so according to some here you are either imagining the difference or the difference is essentially inconsequential, and you should just adapt your preferences to the KH120. How far this is actually possible in the face of something about the character of a speaker that one doesn't appreciate, who can say?