Why do we purchase loudspeakers based off of research done at NRC/Harman without conducting unsighted listening tests for ourselves?

[FrantzM]

Soooooo.... much ado about nothing....

 

[garbulky]

I'm not a self proclaimed objectivist and I don't know which loudspeaker I would prefer without listening. Also with loudspeakers, if you are not going for an out of sight out of mind setup, you have to look at it everyday. I want to know what the whole experience is like so I would perform it sighted.

 

[JJB70]

I don't buy speakers based on looking at how they conform to the Harman /NRC curve.

I guess I might be described as one of the sceptics. It's not that I question the research or their results in terms of statistical correlation between response and preference but rather that a preference is just that, a preference. The essential basis of the research seems to be based on analysis of preference, I can see that such analysis makes perfect sense for manufacturers but one person's preference is irrelevant to me if it differs from my own. If talking about preference then there is no "right" or "wrong". I am told that the Etymotic IEMs are wrong because they don't conform to the headphone curve which has also been developed yet personally I am a huge fan of the ER4SR flat response.

 

[amirm]

I can see that such analysis makes perfect sense for manufacturers but one person's preference is irrelevant to me if it differs from my own.

The research shows that the odds of that are very low. Very few people don't like ice cream.....

I am told that the Etymotic IEMs are wrong because they don't conform to the headphone curve which has also been developed yet personally I am a huge fan of the ER4SR flat response.

Headphone research is not nearly as good as speakers. With speakers, they are measured in anechoic chamber leaving no doubt as to validity of that information. Such does not exist for headphones.

 

[JJB70]

The research shows that the odds of that are very low. Very few people don't like ice cream.....


Headphone research is not nearly as good as speakers. With speakers, they are measured in anechoic chamber leaving no doubt as to validity of that information. Such does not exist for headphones.

But some do dislike ice cream, and if you are one of those people then the ratio of ice cream lovers to haters is irrelevant to your own taste.
A better food analogy is salt. I suspect that if you did a preference study you would find most people like salt in food (there is a reason processed food is heavily salted), yet if you prefer food without salt then you might find adding salt overwhelms the other flavours.
My wife finds European cakes and confectionery to be far too sweet as a typical SE Asian preference for sweetness seems to be different from a typical European. Neither preference is right or wrong.
Both these culinary preferences are largely based on conditioning too and can be changed. A few years ago my doctor advised me to cut salt down, initially I found the resulting food to be bland, now sometimes in restaurants I find food can be like eating salt.

 

[Blumlein 88]

I don't buy speakers based on looking at how they conform to the Harman /NRC curve.

I guess I might be described as one of the sceptics. It's not that I question the research or their results in terms of statistical correlation between response and preference but rather that a preference is just that, a preference. The essential basis of the research seems to be based on analysis of preference, I can see that such analysis makes perfect sense for manufacturers but one person's preference is irrelevant to me if it differs from my own. If talking about preference then there is no "right" or "wrong". I am told that the Etymotic IEMs are wrong because they don't conform to the headphone curve which has also been developed yet personally I am a huge fan of the ER4SR flat response.

I can understand this. I still like my large ESL panels.

A few things make me think Harman may be onto the right track in terms of a reference type loudspeaker. One is doing room correction for a long time now (really speaker correction). You can make two rather dissimilar speakers sound very similar by correcting them to the same target curve. Or with some investigation you can bend any decent speaker to your preference. You learn how some kinds of overall character are related to gentle dips or bumps in response at various frequencies. One of my friends prefers a more distant, but detailed sound. I know by now tweaking his curves what will make him happy, and just how to do that.

Another thing is when I retired and starting doing recreational recording, I needed some monitor speakers. The word if you wanted to keep cost below $1000 was LSR 305 or 308s or spend more among pro studio people. Boy are those some over-performing speakers. On recordings I made they were pretty terrific to use. Not perfect, and not fully capable, but what they could do was get of the way with very little character of their own until pushed rather hard. Great for monitoring. Even if not to someone's preference a very good baseline to start with and tweak to preference.

Encouraged by the little JBL's, when I decided to put together a 5 channel rig for music and video I looked into some Revel speakers. The F12s were very well reviewed everywhere I looked and made with the Harman approach. I purchased a set for cheap from the last month of production. They may in most ways for music playback be more surprising than the LSR305s I have. They are good enough I can EQ them using a room curve measured with my Soundlabs that they'll remind you of the Soundlabs in general balance. They are not quite the same as the panels with all the extra sound in the room from the dipole panels. But using the same FR curve it is recognizable as that balance I get. They have to be played a few db louder as they don't have the inner detail, but it is surprising for non-critical use.

I don't know if I quite think they are there yet, but I think Toole touches on just the right thing talking about breaking the circle of confusion. If we get speakers figured out a bit more they'll be more like electronics. They'll let us engineer for given parameters(I think more needs doing matching them to room dimensions) and then season to taste.

 

[garbulky]

If Harmon was so good then why aren't all the great speakers Harmon hmmm?

 

[MZKM]

Q1) What data are you given about the speaker?

• Spinorama tells you nothing in regards to how hard it is to drive, distortion, compression, CSD, group delay, etc.

Q2) How does your setup effect the speaker you choose?

• e.g., A small room likely would mean to look for a speaker with narrower dispersion, thus less reflections; vice-versa for a large room.

 

[PierreV]

If 75% of the listeners prefer "Harmon" to "Harman", why haven't they renamed the company yet?

 

[pwjazz]

If 75% of the listeners prefer "Harmon" to "Harman", why haven't they renamed the company yet?

Mark Harmon was named sexiest man alive in 1986, so while 75% might be a bit of a stretch, one could say that perhaps about 50% of people preferred Harmon, at least back then.

 

[digicidal]

Once again an over-emphasized sense of importance related to the topic. Sure, a final decision is more likely to come down to listening in my room (don't care as much whether blinded or not). The objective measurements and specifications are simply a way of avoiding having to buy 6 pairs and return 5 - I don't have enough time for that anyway. I don't test drive every car that I might like driving... I pick two or three based on published subjective and objective data... and then test drive only those.

But at the end of the day... since I haven't ever heard (nor even seen data for) a perfect speaker, nor am I aware of any deaths attributed to poor speaker choices... I'm willing to accept that I might make a mistake again - as I've done before - regardless of how I choose. Still haven't found the perfect car I want to drive for the rest of my life either.

Even under fantastical circumstances in which I could ABX 50 possibilities under perfect conditions... I'll be so tired of listening by the fifth or sixth pair that all the ones afterward are getting an unfair comparison anyway. All of his arguments make more sense if we live in a world where you're only allowed to purchase one pair of speakers for life... but thankfully, you can always just take some depreciation or lose shipping costs... and move on if your selection is truly horrible - which is actually pretty rare in modern speakers IMO.

In Harman's case... where you're concerned with which speaker design you're planning on selling and marketing (at significant expense) - now having a much more defined selection process makes perfect sense. Purchasing as an end user just isn't at that level of importance - unless you really can't afford the speakers you're considering at all... in which case you should be investing first and buying much later on.

 

[FrantzM]

Venturing an answer to the OP question.

"Why do we purchase loudspeakers based off of research done at NRC/Harman without conducting unsighted listening tests for ourselves?"

A few reasons:

Unsighted test of loudspeakers is difficult, very difficult. The logistics are beyond what the amateur want to consider... or can even pull off ...

Aside from Harman research there are few research with clear metrics for loudspeakers adequacy. Not too many companies have put forth their findings in such a way. Not too many entities, commercial or otherwise have found what constitute a set of elements of choices/preferences when choosing a loudspeaker. While it might well be a thousand data points, they make clearer, simpler with a few things that are relatively easy to measure: 1) Flat on-axis Frequency Response. 2) Gradual fall off with directivity 3) Let's not forget their contribution on bass reproduction which seems to account for the perception of proper sound reproduction in a big way. They have found (along with other people such as Earl Geddes) that the best bass can only be obtained with several subwoofer .. That should have been a welcome thing for other manufacturers, you would think that would help them sell more wares.

It bears to keep in mind that this is one way to make speakers ... Some manufacturer have gone a different route with excellent results. Constant directivity is one way. The works on Don Keel with CBT are another... Siegfried Linkwitz is not the Harman way... Completely different take. A bit more difficult IMO to follow but the results seems to be there... I have rarely heard people not being impressed by his designs from the Audio Artistry to his latest: the LX Mini and the 512.4 ... Must be doing something right.. The Science behind the speakers is not disputed either...

I should not forget about two of my favorite designer Earl Geddes and Tom Danley ... Different philosophies, superb results nonetheless both subjective and objective.

 

[Blumlein 88]

Q1) What data are you given about the speaker?

• Spinorama tells you nothing in regards to how hard it is to drive, distortion, compression, CSD, group delay, etc.

Q2) How does your setup effect the speaker you choose?

• e.g., A small room likely would mean to look for a speaker with narrower dispersion, thus less reflections; vice-versa for a large room.

Resonances do show up in the spin-o-rama and are to be avoided. Resonances are something CSD would show you. It is true distortion is not shown, but also has not been found to be a factor in listener preference. Harman speakers generally are easier than average to drive.

They don't seem to take into account room size, and I'd think maybe that is worth doing. Then again maybe it is of relatively little importance.

 

[nhunt]

The research shows that the odds of that are very low. Very few people don't like ice cream.....


Headphone research is not nearly as good as speakers. With speakers, they are measured in anechoic chamber leaving no doubt as to validity of that information. Such does not exist for headphones.

I'm interested in this upcoming presentation that might further classify ice cream flavors:

https://twitter.com/i/web/status/1191427099573178368

I'm not sure if this is new research or a recap of their findings on demographics from earlier studies.

Edit: Looks like it recaps this study from March: http://www.aes.org/e-lib/browse.cfm?elib=20289

 

[Hipper]

Is there a practical set of procedures that an ordinary home listener, who enjoys the principles of this site, can use to select and decide on speakers?

 

[PBB]

Is anyone here familiar with Scott Wheeler? This is his question.

I am not sure he believes the research at NRC/Harman was/is of scientific nature -- Yet he claims to not object to or disagree with the research -- although he's previously professed his issues with the research. His circular reasoning begets circular arguments. To no end, can you have a reasonable conversation with this person. He enjoys calling out both objectivists and subjectivists while at the same time, he is a die hard objectivist and die hard objectivist. A part of me used to believe he was a troll and I once referred to him as a nihilist based on our prior conversations but I can see now that he is dead serious.

So the question for us "self proclaimed objectivists" -- his words -- How can we know which loudspeakers we prefer without personally taking part in blind comparisons to substantiate our preference? Is this practical? Is this realistic?

Lastly, this is his proposal to properly testing each loudspeaker to determine a preference: "The difference for me would be setting both speakers up as they are supposed to be used with room treatments that suit each individual design."

Thoughts?

That's why we buy our speakers from professionals so we will not have to build our own anechoic chamber like that of the NRC to test them!

 

[FrantzM]

Is there a practical set of procedures that an ordinary home listener, who enjoys the principles of this site, can use to select and decide on speakers?

Excellent question . Need answer too

 

[BillG]

Is there a practical set of procedures that an ordinary home listener, who enjoys the principles of this site, can use to select and decide on speakers?

This site has been devoted to objective evaluation for quite some time now, and is a trusted source of mine:

https://www.audioholics.com/loudspeaker-design/understanding-loudspeaker-measurements

 

[BillG]

Is there a practical set of procedures that an ordinary home listener, who enjoys the principles of this site, can use to select and decide on speakers?

Another article from them:

"In a nutshell, if a speaker exhibits flat and linear on-axis frequency response with consistent off-axis performance to preserve critical early reflections, then the speaker will score very highly in blind listening tests and also provide more consistent performance from room to room. The speaker must also be free of pesky resonances and the transition between drivers must be as seamless as possible, hence why waveguides are an essential part in most Harman loudspeaker products."

https://www.audioholics.com/loudspeaker-design/measure-loudspeaker-performance

 

[BillG]

Is there a practical set of procedures that an ordinary home listener, who enjoys the principles of this site, can use to select and decide on speakers?

Or you could just do what I do: Check out any speaker from a Harmon International owned company, and take it from there. That's a bit of a short cut though, as they've already done the hard part with their extensive R&D...