Worst measuring loudspeaker?

[Putter]

1) The results are relevant to normal listening levels. Duration is the key. A moderate listening level of 80 dB on average can easily produce instantaneous peaks of not only 105 dB, but even, albeit occasionally, 115 dB at certain frequencies, from certain instruments, and especially full orchestras, cymbal crashes, loud bass drum (e.g., Telarc's -- or their, or Mercury's cannons), timpani, piano, rimshots, organ, brass, etc. I imagine that is why THX established their peak levels at 105dB above 80Hz, and 115dB in the bass, in truly large rooms.

2) Neither 105dB, nor 115dB can be expected to "damage your hearing" if
brief enough. In the chart below, notice that OSHA thinks that a SPL of 115 dBA is safe for 15 entire minutes. Musical peaks at that level tend to last for 1/4 second or less, then drop back to around 100dBA (OK for much longer). If you have watched your SPL meter during the most powerful Mahler or Beethoven symphony, or perhaps Copland's Fanfare for the Common Man [Richard Morris & The Atlanta Brass Ensemble, adjusted so your pants legs flap in the breeze (my desk seems to go out of square) you will probably see that at 13 feet (my listening distance) it only gets to about 100 dBA at the loudest peaks]. Unlike OSHA and NIOSH I refuse to use dBA, because it ignores the deepest bass, reading peaks lower than they really are. Only dBC or dBZ will be sensitive enough in the bass for music, where there can be great power in the bass. Using C or Z produces a more accurate estimate of the SPL of music.View attachment 476301

View attachment 476301View attachment 476296

View attachment 476296

3) True, the speakers I posted information on are aged, except for the Klipschorn. The Stereophile article is newer.

While all this is interesting and true, the real question is are they relevant? Are those doppler distortions audible in a blind test such that they create a preference

Also to be a little nitpicky, as fan of wide dynamic range recordings, it is very rare to find recordings with 20 db of dynamic range even in classical music although it's certainly possible in certain movie soundtracks. Incidentally I checked on the Dynamic Range Database the dynamic range of the Richard Morris recording which is listed at 15 db. It is certainly possible this is inaccurate, but that may be true of your referenced measurements of Doppler Distortion.

To circle back to your original point which is that Doppler distortion more significant than harmonic distortion. I made a quick check of the AR 98LS review in January 1985 Audio magazine show a chart of a test of IM distortion which shows a maximum of about 5.5% at 100 watts which at a sensitivity of 87db/1m/watt gives a 107 db at 100 watts. Note that Mr. Heyser does NOT note or comment on the audibility of IM distortion which leaves to wonder where and who the reference quote came from, yes Stereophile, but who and from what measurements or tests?

What any of this has to do with the worst measuring speaker is beyond me which is why this is my last response to you on this subject..

 

[Alexander Gödde]

This is appaling - a 15k USD standmount with a 9 db (!) drop between 2 and 3 kHz. Doesn't fill in to the sides either, so will carry on to the in-room response.

Borresen C1 - review by Hifi-News

Of course it gets a rating which falls into Hifi-News' usual 80%-90% range (whatever that means).

 

[Mart68]

This is appaling - a 15k USD standmount with a 9 db (!) drop between 2 and 3 kHz. Doesn't fill in to the sides either, so will carry on to the in-room response.

Borresen C1 - review by Hifi-News

Of course it gets a rating which falls into Hifi-News' usual 80%-90% range (whatever that means).

It's worse than you think, they are fifteen thousand pounds, not fifteen thousand dollars.

 

[kharan]

This is appaling - a 15k USD standmount with a 9 db (!) drop between 2 and 3 kHz. Doesn't fill in to the sides either, so will carry on to the in-room response.

Borresen C1 - review by Hifi-News

Of course it gets a rating which falls into Hifi-News' usual 80%-90% range (whatever that means).

This is quite frankly very interesting.

Here we have the measurements for the Borressen C1 and the Borressen X3. Those are speakers from different ranges and different form factors.

These speakers have different, drivers, they also have different waveguides, and one is a two way bookshelf, and the other is a three way floorstander (so different spacing between drivers and different crossovers).

And yet, we find on both speakers the same dip at 300 Hz and between 2 and 4 kHz. We also have fairly pronounced bass peaks at 80 and 100 Hz respectively.

This tells me this is the house frequency response they are aiming for and do think that those peaks and dips are necessary to achieve a music production they and their customers enjoy.

Spoiler

from Hifinews and EAC

 

[Salt]

This is quite frankly very interesting.

Here we have the measurements for the Borressen C1 and the Borressen X3. Those are speakers from different ranges and different form factors.

These speakers have different, drivers, they also have different waveguides, and one is a two way bookshelf, and the other is a three way floorstander (so different spacing between drivers and different crossovers).

And yet, we find on both speakers the same dip at 300 Hz and between 2 and 4 kHz. We also have fairly pronounced bass peaks at 80 and 100 Hz respectively.

This tells me this is the house frequency response they are aiming for and do think that those peaks and dips are necessary to achieve a music production they and their customers enjoy.

Spoiler

from Hifinews and EAC

I would not call it house curve, but unawareness of the simplest basics of constructing a crossover, with emphasis on the polarity of the mids driver, and that measurements are not performed at all or not correctly interpreted.

 

[MattHooper]

This is quite frankly very interesting.

Here we have the measurements for the Borressen C1 and the Borressen X3. Those are speakers from different ranges and different form factors.

These speakers have different, drivers, they also have different waveguides, and one is a two way bookshelf, and the other is a three way floorstander (so different spacing between drivers and different crossovers).

And yet, we find on both speakers the same dip at 300 Hz and between 2 and 4 kHz. We also have fairly pronounced bass peaks at 80 and 100 Hz respectively.

This tells me this is the house frequency response they are aiming for and do think that those peaks and dips are necessary to achieve a music production they and their customers enjoy.

Spoiler

from Hifinews and EAC

On Audiogon a retailer made a thread pimping Borresen speakers. I pointed out that measurements of Borresen speakers does not suggest the company knows what it’s doing in terms of speaker design.

The retailer claims that they also sell more linear, measuring speakers like Magico, KEF, YG Acoustics etc and claims that their customers almost always prefer the Borresen.

A lot of self interest packed into that claim I’m sure. On the other hand, if they are continuing to represent the brand, I guess they have customers purchasing them in numbers significant enough to keep selling.

 

[Mart68]

We've seen other companies designing for showroom appeal - PMC domestic and B&W for example although they all do it a little differently.

The B&W chief engineer is on record as saying 'We do it because it sounds better' and Borresen make a similar statement.

They may sell on audition but do those speakers continue to satisfy long term? Do they remain consistent with a wide variety of programme or do owners start to discover 'Just how many bad recordings there are'?

 

[kharan]

I would not call it house curve, but unawareness of the simplest basics of constructing a crossover, with emphasis on the polarity of the mids driver, and that measurements are not performed at all or not correctly interpreted.

The are fairly different speakers, that diminishes the likelyhood that the same errors are produced when designing them, hence my speculation for intentionality.

On Audiogon a retailer made a thread pimping Borresen speakers. I pointed out that measurements of Borresen speakers does not suggest the company knows what it’s doing in terms of speaker design.

The retailer claims that they also sell more linear, measuring speakers like Magico, KEF, YG Acoustics etc and claims that their customers almost always prefer the Borresen.

A lot of self interest packed into that claim I’m sure. On the other hand, if they are continuing to represent the brand, I guess they have customers purchasing them in numbers significant enough to keep selling.

Some people like their speakers to interpret music rather than reproduce what's on the record. Nothing wrong with that. I know the hobby is called HiFi, but not everyone is about fidelity.

 

[MattHooper]

We've seen other companies designing for showroom appeal - PMC domestic and B&W for example although they all do it a little differently.

The B&W chief engineer is on record as saying 'We do it because it sounds better' and Borresen make a similar statement.

They may sell on audition but do those speakers continue to satisfy long term? Do they remain consistent with a wide variety of programme or do owners start to discover 'Just how many bad recordings there are'?

Yes, good questions. When it comes to B&W I have two audiophile friends who continue over time to be thrilled with their B&W speakers (even though I couldn’t live with them myself).

As to Borresen, as somebody who watches the used market, I would be hesitant about buying them in terms of resale value. Borresen keeps coming up with new models, with plenty of Borresen on the used market, and they don’t seem to hold their value that well.

 

[Salt]

The are fairly different speakers, that diminishes the likelyhood that the same errors are produced when designing them, hence my speculation for intentionality.

Some people like their speakers to interpret music rather than reproduce what's on the record. Nothing wrong with that. I know the hobby is called HiFi, but not everyone is about fidelity.

If they are not aware of grand flaws at one speaker, how can you be assured they don't do the same at others?

 

[Mart68]

Yes, good questions. When it comes to B&W I have two audiophile friends who continue over time to be thrilled with their B&W speakers (even though I couldn’t live with them myself).

A friend of mine has the original 603 from about 15 years ago or so. Have to say I have not noticed any major issues with them either for music or HT and I've spent a fair bit of time listening to them. The top end lift is noticeable against a more neutral speaker but it isn't unpleasant.

I'm pretty sure it is carefully calculated, possibly/ probably with listening tests they have not published. B&W measure their designs, no question about that, so they know what the FR looks like and they say they're fine with it.

Borresen I am not sure about, I would not be surprised if they were totally designed by ear by a one - or just a handful - of people.

 

[MattHooper]

A friend of mine has the original 603 from about 15 years ago or so. Have to say I have not noticed any major issues with them either for music or HT and I've spent a fair bit of time listening to them. The top end lift is noticeable against a more neutral speaker but it isn't unpleasant.

I'm pretty sure it is carefully calculated, possibly/ probably with listening tests they have not published. B&W measure their designs, no question about that, so they know what the FR looks like and they say they're fine with it.

Borresen I am not sure about, I would not be surprised if they were totally designed by ear by a one - or just a handful - of people.

That’s cool. I mentioned before that I actually enjoy listening to B&W speakers, just as a place to visit and I get their appeal. They just aren’t a speaker that I would choose for my long-term listening.

When it comes to loudspeakers, I’ve always looked for a sound that seemed subjectively well balanced. I don’t want to feel aware of some chunk missing from the frequency spectrum, or some boost that is a consistent artificial sound. So in the main, I tend to be attracted to loudspeakers that don’t stray too far from neutral. However, I can be OK with the loudspeaker that has some level of colouration or frequency deviation so long as subjectively I find the results pleasant, and to me that usually means that they are colorations that have been carefully balanced so that they don’t stick out like a sore thumb.
For instance, my current speakers measure fairly neutral but they do have a little rise on axis anechoically starting around five or 6K, though which is smoother in room. This gives a little bit of “ sparkle and air” that I enjoy (I really don’t care for dark sounding speakers), but it doesn’t stick out in terms of producing excessive vocal sibilance or unnaturally emphasized or peaky upper frequency transients. It’s a very canny balance to my ears so I’m happy with it.

But again my pal with the current B&W flagship stand mount speakers (and flagship B&W sub) is just over the moon happy with the sound. He’s been reviewing speakers for decades and I’ve never heard him so happy and enthusiastic “ this I could live with forever.” He’s totally aware of the frequency profile, both in terms of measurements and in terms of hearing some classic B&W sculpting of the sound, but he loves it because he finds it sounds incredibly open, boxless, ultra clear and detailed and dynamic, but also subjectively vary smooth (lack of grit/grain) so that he can actually listen comfortably and enjoy that extra vividness.

I haven’t had a chance to hear them yet.

 

[Mart68]

I can be OK with the loudspeaker that has some level of colouration or frequency deviation so long as subjectively I find the results pleasant, and to me that usually means that they are colorations that have been carefully balanced so that they don’t stick out like a sore thumb.
For instance, my current speakers measure fairly neutral but they do have a little rise on axis anechoically starting around five or 6K, though which is smoother in room. This gives a little bit of “ sparkle and air” that I enjoy (I really don’t care for dark sounding speakers), but it doesn’t stick out in terms of producing excessive vocal sibilance or unnaturally emphasized or peaky upper frequency transients.

Likewise. Thing is, it may have changed in recent years but 'acceptable accuracy' for a speaker was always anechoically flat to within plus or minus 3dB.

Which isn't unreasonable even though we now have fairly inexpensive passive speakers that do better than that.

So there is some room for 'subtle voicing' without creating a speaker that's a total dog's dinner.

 

[thewas]

A friend of mine has the original 603 from about 15 years ago or so. Have to say I have not noticed any major issues with them either for music or HT and I've spent a fair bit of time listening to them.

The B&W up to approximately the new century where tuned quite neutrally on-axis and had also smooth directivity:

B&W DM603 S3 loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements The handsome B&W DM603 S3 has higher-than-average voltage sensitivity, at an estimated 89dB(B)/2.83V/m. Though this is slightly lower than the specified 90dB, the difference is not that significant, given the usual experimental error. The speaker's impedance plot (fig.1)...

www.stereophile.com

Newer ones aren't like this, exemplary:

Bowers & Wilkins 702 S2 loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure 7the Bowers & Wilkins 702 S2's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield responses. My estimate of the B&W's sensitivity was a high 90.2dB(B)/2.83V/m...

www.stereophile.com

 

[kharan]

If they are not aware of grand flaws at one speaker, how can you be assured they don't do the same at others?

We have two speakers from different lines, with different drivers, different cabinets and different crossover designs. If they truly didn't know what they were doing at all, the grand flaws would appear at different places. Instead, were have very similar frequency responses, with the same characteristics.

More anecdotally, the reports from people people listening to different Boressen speakers across their whole line seem to indicate that there is some consistency in their house sound.

That suggests intentionality.

The grand flaws you mention strongly belong to the realm of audibility, they might even have learned to enjoy those flaws and decided to replicate them across their entire speaker range.

 

[DSJR]

We've seen other companies designing for showroom appeal - PMC domestic and B&W for example although they all do it a little differently.

The B&W chief engineer is on record as saying 'We do it because it sounds better' and Borresen make a similar statement.

They may sell on audition but do those speakers continue to satisfy long term? Do they remain consistent with a wide variety of programme or do owners start to discover 'Just how many bad recordings there are'?

I think 'we' know the answer to that one - the wealthy elderly men who buy these (not sure very many ladies are into HiFi as a hobby, although a lot love music), can't hear much over 10kHz in-room, so the massive response errors up there don't matter, the presence dip gives a nicely 'rounded' tone to go with the bass on the larger models and in PMC-Domestic's case, the things are usually pointed straight out (little to no toe-in) and the mid-bass shenanigans due to cancellations in the line, can sometimes work well in rooms with bass peaks at those frequencies. I just find the latter speakers boringly dull and more than a bit coloured, except for the Fact Fenestria, which fills a large space with a vaulted ceiling with consummate ease (I suspect it's too much in a smaller room however).

 

[DSJR]

A friend of mine has the original 603 from about 15 years ago or so. Have to say I have not noticed any major issues with them either for music or HT and I've spent a fair bit of time listening to them. The top end lift is noticeable against a more neutral speaker but it isn't unpleasant.

I'm pretty sure it is carefully calculated, possibly/ probably with listening tests they have not published. B&W measure their designs, no question about that, so they know what the FR looks like and they say they're fine with it.

Borresen I am not sure about, I would not be surprised if they were totally designed by ear by a one - or just a handful - of people.

The B&Ws around the first 600 and later 500 series models, had a subjectively slightly 'scooped out' upper mid, which wasn't at all unpleasant and if anything, I remember the slightly lively top being pleasant rather than acidic. Sadly, the s2 and s3 models really *did* go for paint stripping/filling removal and I couldn't listen for long as it was so unpleasant. Their last last baby models I heard was twenty years ago now with the 68 series, and these just sounded a little 'hollow' in the mids (slight cupped-hand 'awww' sound), as if extra wadding needed to be put in the enclosures to stop the inside of the box bleeding to the outside (highly technical, ain't I). No idea of later models but they all seem seriously in error to me and how studios can use and adapt to the N800 models of any age for proper tonal monitoring is beyond me (I chuckle remembering Abbey Rd using a N801 once as a door stop and their vinyl room uses pro twin-box PMCs it seems)

 

[Mart68]

The B&W up to approximately the new century where tuned quite neutrally on-axis and had also smooth directivity:

B&W DM603 S3 loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements The handsome B&W DM603 S3 has higher-than-average voltage sensitivity, at an estimated 89dB(B)/2.83V/m. Though this is slightly lower than the specified 90dB, the difference is not that significant, given the usual experimental error. The speaker's impedance plot (fig.1)...

www.stereophile.com

Newer ones aren't like this, exemplary:

Bowers & Wilkins 702 S2 loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure 7the Bowers & Wilkins 702 S2's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield responses. My estimate of the B&W's sensitivity was a high 90.2dB(B)/2.83V/m...

www.stereophile.com

Ah I got that wrong, the speakers I'm referencing are the original 683 not the 603.

Bowers & Wilkins 683 S2 loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the Bowers & Wilkins 683 S2's frequency response in the farfield; and an Earthworks QTC-40, with its small, ¼"-diameter capsule, for the nearfield responses.

www.stereophile.com

 

[Arindal]

would not call it house curve, but unawareness of the simplest basics of constructing a crossover, with emphasis on the polarity of the mids driver, and that measurements are not performed at all or not correctly interpreted.

Would consider this rather unlikely as people both on sales and customer side seem to specifically highlight their preference for the given frequency response and sound characteristics. We might not like it, call it not linear or bemoan its technical flaws, but it seems to be intentional. Maybe we should accept the fact that some people like bloated upper bass and recessed presence.

the speakers I'm referencing are the original 683 not the 603.

I remember the original 683 as one of the brightest (or rather dullest) examples of a speaker offering tonality not even remotely resembling what one would expect from the frequency response graph.

Since Bowers introduced first order crossover topology and broad overlapping frequency bands some 20 years ago, it became completely impossible to judge or predict the outcome looking solely at on-axis graphs, as the result heavily relies on directivity and lobing effects.

 

[dfuller]

Since Bowers introduced first order crossover topology and broad overlapping frequency bands some 20 years ago, it became completely impossible to judge or predict the outcome looking solely at on-axis graphs, as the result heavily relies on directivity and lobing effects.

Said it once and I'll say it again: first order crossovers are stupid!