Master Thread: Are measurements Everything or Nothing?

[MattHooper]

Make sense as to why you didn't kind of giving the game away as it's obvious with the responses in two different rooms how influential the room actually is with this type of design.

Rob :}

Yes, they are a finicky speaker to set up!
I’ve heard of people having different and even rather unpleasant results.

This is one reason why I didn’t end up buying them because more than any other loudspeaker I didn’t feel like I could predict how they would sound in my room. Which of course, are the benefits you get with a more
“ best practises” design like the Revel.

But it seems that most people who are able to accommodate them in their room and dial them in properly, report similar sonic qualities. I’ve had many owners of the Devore O/96 and O/93 tell me on other forums that my impressions were bang to what they hear at home.

 

[pablolie]

And yet to me the post important part of the review was not posted. The measured in-room response from the review.

Fig.7 DeVore Fidelity Orangutan O/96, spatially averaged, 1/6-octave response in: AD's listening room (red), JA's listening room (blue).

Now a speaker designed using Harman/Toole guidelines

Fig.7 JBL 1400 Array, spatially averaged, 1/6-octave response in LG's listening room.

And another

Fig.7 JBL 4367 Studio Monitor, spatially averaged, 1/10-octave response in AH's listening room.

And finally Salon 2

So what do these in-room measurements show?

Rob

I have a very good friend that owns these and I have listened to them often.

High sensitivity so they are paired with tube amps. Great for baroque and strings paired with such an amp. Something like Chopin etudes sounds great. Classic guitar also amazing.

Far less so when pushed by several registers at once. It's a specialist. Great in it's domain but not versatile enough for me.

I am sure it can seduce sublimely in blind testing with the right tracks. Just like it will probably underwhelm with more complex, full range tracks. Just my opinion.

 

[MattHooper]

Something like Chopin etudes sounds great. Classic guitar also amazing.

Yeah, some of the classical guitar tracks I played on those speakers were among the most satisfying I’ve ever heard them.

Far less so when pushed by several registers at once.

That has been noted by some other listeners as well, some who gave that very reason for not buying them. They noted a level of congestion for instance during dense orchestral music.

It's a specialist.

I can definitely see where that opinion comes from. To me this kept to the subjectivity of what makes for a “ specialist” loudspeaker.
Ultimately, I feel it comes down to the individual and what he/she finds satisfying on the speaker.

Harbeth speakers, for instance, seem to have something of a dual reputation. On one hand, you have those people who listen to them and conclude they are a “ pipe and slippers” British loudspeaker that does certain things such as vocals and some instrumental music, but can’t do justice to other genres like rock, dance, etc.

Whereas other listeners feel the opposite: that they are a very well-balanced speaker which makes them satisfying across musical genres, and that is why they bought them.

I owned Harbeth speakers and my view was somewhere in between: I found them to have a very satisfying, balanced sound quality that made any genre I played sound very good.
Ultimately, they couldn’t “ slam” like my larger speakers. But I understand somebody being fully satisfied with them.

 

[Axo1989]

And yet to me the post important part of the review was not posted. The measured in-room response from the review.

Fig.7 DeVore Fidelity Orangutan O/96, spatially averaged, 1/6-octave response in: AD's listening room (red), JA's listening room (blue).

Now a speaker designed using Harman/Toole guidelines

Fig.7 JBL 1400 Array, spatially averaged, 1/6-octave response in LG's listening room.

And another

Fig.7 JBL 4367 Studio Monitor, spatially averaged, 1/10-octave response in AH's listening room.

And finally Salon 2

So what do these in-room measurements show?

Rob

Often, when I look at those measured room FR graphs I think "FFS I'd totally run Sonarworks* before listening to that". I know not everyone is happy with (their) room correction, but I am.

*or your EQ of choice (and people will note the smooth DR loudspeakers will respond to EQ well, etc)

 

[pablolie]

... Yeah, some of the classical guitar tracks I played on those speakers (Devore) were among the most satisfying I’ve ever heard them.

Harbeth speakers, for instance, seem to have something of a dual reputation. ..

I commented long ago in this thread saying that's why IMO measurements are great, but not "everything". Some aficionados know how to optimally set up - and outright flatter - their known focused favorites with their tailored system.

I like to see versatile, well engineered measured basics, but my favorite tracks are my ultimate guideline. But I have very different kinds of music I love, so the Devore or Harbeths are too single minded to me. The Devore clearly are designed to work with low wattage tube amps and focused acoustic presentation (that's how my friend uses them extremely effectively, and utterly aware of their limitations), the Harbeths I have heard cater to the LS3 legacy... which is not a bad thing at all, my Dad has Spendors that turned me into an early "audiophile". But one can have more extension than the classic LS3 these days. LS3 derivatives seem to share quite similar measurement patterns.

 

[MattHooper]

Often, when I look at those measured room FR graphs I think "FFS I'd totally run Sonarworks* before listening to that". I know not everyone is happy with (their) room correction, but I am.

*or your EQ of choice (and people will note the smooth DR loudspeakers will respond to EQ well, etc)

Understandable.

I was able to listen to those speakers in rooms they were pretty capacious, so both the speakers and my listening position were well away from room boundaries, with my listening as close as I could get (8ft) so I suspect I heard something somewhat closer to what the anechoic data suggests. Especially in terms of high frequencies.

It would be interesting for sure if I had in-room measurement data of the set ups that I heard. But I would still be left asking “ OK, but everything taken together, what exactly does that sound like? I could see bumps and blips here and there, but will those stick out to me and detract from enjoyment? Or will those be less obvious across my music selection then they might look on the graph?
Hence, I would want to listen anyway to see how I perceive them.

Interestingly, I found that the measurements for “ better measuring” loudspeakers tended to track more obviously/predictably with what I heard.

In other words when I see the measurements for something like the Kii Audio or Revel performa measurements, I think “ yeah that’s pretty much exactly what I heard.”

But when things get more complex and wacky as with a Devore speaker it can be harder to know in any particular set up the precise subjective results. (for me anyway.)

 

[MattHooper]

I commented long ago in this thread saying that's why IMO measurements are great, but not "everything". Some aficionados know how to optimally set up - and outright flatter - their known focused favorites with their tailored system.

I agree.

I think it’s worth remembering that some of the reasons certain designs get demerit points is that their measurements indicate they will be tricky to set up for good sound in an actual listening room (e.g. tricky off axis behaviour). That’s one of the benefits of the measurement criteria extolled here: it indicates loudspeaker that is more likely to sound good in more rooms.

That doesn’t mean that in particular rooms with particular care, some less perfect measuring loudspeakers can’t also end up sounding pretty good (and good with the type of music somebody likes).

 

[pablolie]

..

That doesn’t mean that in particular rooms with particular care, some less perfect measuring loudspeakers can’t also end up sounding pretty good (and good with the type of music somebody likes).

And I should note I am a zero believer in "PEQ can fix everything". I am actually convinced it pushes gear into a direction it was not designed for, and will cause it to fail in its original design goal and sound merely run of the mill at best.

You cannot correct imperfect gear to sound better with EQ, you'll just push it into failing elsewhere. Pick gear that measures well and you enjoy. EQ should be used extremely sparingly, if at all... that's my premise that does not align with the usual ASR storyline.

 

[Newman]

That's largely how ASR works. A few manufacturers of suitable goods provide them free and use it as a testing service and promotional channel. The remainder are purchased by members and sent in for testing. It's sustainable as a business model because work is voluntary, costs are low and donations come in from members, who perceive benefits.

Business model? It's a labour of love, not a business at all, and I'm sure Amir is substantially out of pocket. Even if the donations did match the costs, it's by the bye and would not determine whether ASR works or not.

ASR is one of those special projects where an individual just wants to do something good, reaches into his own pocket to make it happen with no anticipation of recompense, and some of the people freely using this good thing just want to show their gratitude with a gift. There is no 'business model'. ASR would be exactly the same if there were no donations.

 

[Newman]

[to Axo1989] Well, that's what you should have said, then. Can you show in support of that, the data sets concerned are "somewhat limited"? Can you point to a larger dataset that shows something different? What are your conclusions based on the available data?

Lay out your own argument in detail. I'm here to learn.

Unfortunately, we have the answer: "I'm too busy".

I am reminded of SIY's description of how to deal with this: "My first question is always a simple, "Evidence?" When invariably such is not forthcoming but there's a wall of handwaving text, my terse dismissal is, "OK, so no evidence.""

Let's be frank: if there were, as you requested, "larger datasets that show something different", are we to assume that Toole would be unaware of it? That would be a heck of an assumption. Or that he would be aware of it but ignore it in subsequent editions of his book? We are talking about a doyen of audio research. Let's be sensible.

 

[Blumlein 88]

I agree.

I think it’s worth remembering that some of the reasons certain designs get demerit points is that their measurements indicate they will be tricky to set up for good sound in an actual listening (e.g. tricky off axis behaviour). That’s one of the benefits of the measurement criteria extolled here: it indicates loudspeaker that is more likely to sound good in more rooms.

That doesn’t mean that in particular rooms with particular care, some less perfect measuring loudspeakers can’t also end up sounding pretty good (and good with the type of music somebody likes).

The worst pickiest speakers I've setup were some Magnepan 3.3Rs. The sweet spot was pretty small, distance from the rear wall was crucial, it was picky, picky about toe in, and equally so about tilt of the panel. I spent more time and effort on it than anything I've messed with. It could indeed be made to sound excellent. It also could have drumming bass issues, weird imaging, and badly balanced uneven midrange response. Off axis measurement show why most of this is a problem. If I hadn't owned them, I don't know if they ever would have ended up sounding right as I wouldn't have experimented that long with someone else's room. I did set them up for someone else and it also was hard. I believe the person I purchased them from never had them right which is why he sold them. Ease of setup should not be devalued.

K-horns in a different way are similar. In the right room, used as intended they can be better than most people ever hear them. They aren't hard to set up if you have the right room. They go in one place in the corner. The room however needs to be at least 12 feet tall and that wide (preferably wider), and needs to be pretty long. The walls on the speaker end need to be very sturdy, preferably brick or brick and underground. If you don't have the right space, get something else.

 

[MattHooper]

Business model? It's a labour of love, not a business at all, and I'm sure Amir is substantially out of pocket.

“ASR is relies on goodwill of the visitors and members of the site to cover its expenses and enable testing of more products.”

^^^ as far as I know that pretty much constitutes a business model.

I think the charitable reading of Axo1989’s comment simply references that fact.
We all recognize that this is not a “ get rich quick” endeavour by Amir. I’m sure he’s not buying yachts on the donations.
Many businesses, even ones that don’t earn the proprietor much money, are also labours of love. Many YouTuber reviewers will tell you that. And all of us very much appreciate Amir’s time and effort.

 

[Blumlein 88]

“ASR is relies on goodwill of the visitors and members of the site to cover its expenses and enable testing of more products.”

^^^ as far as I know that pretty much constitutes a business model.

I think the charitable reading of Axo1989’s comment simply references that fact.
I think we all recognize that this is not a “ get rich quick” endeavour by Amir. I’m sure he’s not buying yachts on the donations.
But many businesses, even ones that don’t earn the proprietor enough money, are also labours of love. Many YouTuber reviewers will tell you that. And all of us very much appreciate Amir’s time and effort.

Amir has a camper van and vegetable garden. I bet he paid for it all from ASR proceeds/

Now how he paid for the Klippel and AP testers is another matter..........................

 

[MattHooper]

The worst pickiest speakers I've setup were some Magnepan 3.3Rs. The sweet spot was pretty small, distance from the rear wall was crucial, it was picky, picky about toe in, and equally so about tilt of the panel. I spent more time and effort on it than anything I've messed with. It could indeed be made to sound excellent. It also could have drumming bass issues, weird imaging, and badly balanced uneven midrange response. Off axis measurement show why most of this is a problem. If I hadn't owned them, I don't know if they ever would have ended up sounding right as I wouldn't have experimented that long with someone else's room. I did set them up for someone else and it also was hard. I believe the person I purchased them from never had them right which is why he sold them. Ease of setup should not be devalued.

K-horns in a different way are similar. In the right room, used as intended they can be better than most people ever hear them. They aren't hard to set up if you have the right room. They go in one place in the corner. The room however needs to be at least 12 feet tall and that wide (preferably wider), and needs to be pretty long. The walls on the speaker end need to be very sturdy, preferably brick or brick and underground. If you don't have the right space, get something else.

All you have to do is lower your standards

Out of curiosity, have you ever heard a true infinite baffle speaker set up? If I remember correctly, something like loudspeaker drivers placed in a large wall would achieve this. Or is there other ways of achieving an infinite baffle performance? And do you happen to know if the standard measurement criteria, especially in terms of off axis performance, applies just as much to an infinite baffle, or whether that brings in other considerations? Do they increase the challenges of getting good sound in an actual listening room or somehow ameliorate some of the challenges?

Sorry to lay this on you, but I figure you’d be a good person to ask.

 

[Blumlein 88]

All you have to do is lower your standards

Out of curiosity, have you ever heard a true infinite baffle speaker set up? If I remember correctly, something like loudspeaker drivers placed in a large wall would achieve this. Or is there other ways of achieving an infinite baffle performance? And do you happen to know if the standard measurement criteria, especially in terms of off axis performance, applies just as much to an infinite baffle, or whether that brings in other considerations? Do they increase the challenges of getting good sound in an actual listening room or somehow ameliorate some of the challenges?

Sorry to lay this on you, but I figure you’d be a good person to ask.

I've heard two, but it was so long ago it wouldn't be a good basis for whether that is a good approach by how it sounded. It should be. I would have to give more thought about how the off axis behavior would figure into this. Certainly no reason the part of response used in the listening window would be any different. It has some advantages in the low end.

 

[Axo1989]

Business model? It's a labour of love, not a business at all, and I'm sure Amir is substantially out of pocket. Even if the donations did match the costs, it's by the bye and would not determine whether ASR works or not.

ASR is one of those special projects where an individual just wants to do something good, reaches into his own pocket to make it happen with no anticipation of recompense, and some of the people freely using this good thing just want to show their gratitude with a gift. There is no 'business model'. ASR would be exactly the same if there were no donations.

Check the plain English dictionary meanings of "business" and try again. I'll wait.

 

[MattHooper]

Check the plain English dictionary meanings of "business" and try again. I'll wait.

I hope we don’t ever discover that Amir has secretly been using donations to fund his startup high-end AC cable business

 

[sigbergaudio]

I thought about posting those as well, but I saved them for when I posted the link to the review so that people could see them.

I wanted people to try and infer from the Anechoic type data, so they could include their speculations about how such a loudspeaker might interact in different rooms. Hence why I didn’t post the particular in-room results from Stereophile.

Those in-room measurements certainly track pretty well with what I'd expect to see based on the anechoic measurements, but they do not track very well with your experience. We don't know how the in-room response was in the rooms where you listened to them of course. But it is likely they were not very bass heavy there either.

Some of what you are describing can be attributed to the extra energy in the midrange, but that you generally found the sound to be rich and full bodied is somewhat strange. We could theorize that the resonances (adding energy over time, which we do see in some of the measurements) to some extent compensate for the lack of direct energy. Also the relatively wide baffle will give you more directed energy towards you in the midbass.

But yes, expecting to understand the sonic signature of a speaker purely from the measurements is a tad optimistic. Seeing the speaker design helps a little bit.

 

[Mart68]

That peak in the mids in both rooms pretty much guarantees that the owner will often complain about 'recording quality.'

They never blame the speaker which is why it's possible to claim that so many owners are 'satisfied' with them.

I don't think I could describe the sound of a speaker from measurements but do know to look for the red flags.

I have had a demo of these speakers they were pleasant enough with 'gentle' music, the dealer didn't take a chance with anything else.

 

[Galliardist]

Yeah, some of the classical guitar tracks I played on those speakers were among the most satisfying I’ve ever heard them.

From those measurements, that's a surprise. It might do the "hyperrealistic" thing with solo guitar where it's "in the room", sounds like a really nice classical guitar, but the details in the playing are missing, but I doubt you'd fall for that as others do, even if you do look for that realism thing overall.