Why don’t ASR members laud Neumann speakers?

[Bounce44.1]

The Adams are quite popular in the mixing world. They made it to my final 3. I went with the BMR Philharmonitor from Philharmonic audio for midfields and CCB-8 from HSU Research for nearfields. These retired my Lsr 305's and 310 sub.

The other one on my final 3 was the neumann.

 

[KSTR]

Some older ADAMs (SX and AX series) certainly qualify as bright and crisp, even ear-bleeding, notably the very popular A7X. Those are great for EDM but fail with eg electric guitars or brass sections and this impression is shared by many people.
There are reasons for this and of course it can and has been measured. It's a combination of significant tweeter distortion, midwoofer cone breakup, directivity mismatch (no waveguide) and the overall "bathtub" EQ.

 

[jhaider]

The high level of port output - almost the same level as driver output, but out-of-phase, manifests as destructive interference. Credit for the graph goes to Sound and Recording magazine of Germany, which professionally measures speakers in a research institute's anechoic chamber. It is anyone's guess as to why Harman didn't engineer it out.

Because they could not eliminate it without either changing the acoustic design (higher tuning, bigger cabinet) or raising build cost substantially (passive radiators).

Any port that tunes such a small cabinet so low will have some out of band resonance. As discussed here at least it's very narrow in bandwidth. The S-shaped port with wide flare is likely the best they could do for a low-tuned small ported enclosure.

 

[jhaider]

It's clear this has become personal for you. No worries. Have a great week ahead.

Cheers,
Mitch

It does not reflect well on you to flood the zone with condescending statements that, taken charitably, could be accurate as to the brand's past products but badly misinterpret current products based on the data presented, and then rear up into High Grievance Stance as if somebody took your Sharpie away when those inaccurate statements are questioned. But so it is. Too bad. It seems we actually have very similar sonic tastes, based on our shared appreciation of NAD’s headphones.

While you refused to answer the question presented, here's the "teacher's edition" answer anyway: the speakers corresponding to the anechoic on axis frequency responses and horizontonal/vertical polar maps I posted were correctly identified by others as Neumann KH 80 DSP (A) and JBL LSR705P (B). Presumably LSR705P was designed using Harman's defined best practices, but KH 80 DSP was not designed using Harman's defined best practices. Yet they seem to have arrived at strikingly similar results both in on-axis and directivity design targets. I am still unaware of any property that would cause KH 80 DSP to maintain its anechoic on axis treble response at a normal home listening distance, while LSR705P would show a downscoping response at a normal home listening distance. Care to enlighten?

Lastly, you haven’t explained why you put such primacy on in-your-room measurements, made with unclear or undisclosed spatial averages so they cannot even be replicated by someone with access to the same room, while disregarding universal data. Do you care to explain? You may have good reasons, but are acting contrary to the current research-based consensus. As @Floyd Toole wrote earlier (emphasis added)
“Second, from comprehensive anechoic data on a loudspeaker (the spinorama for example) one can predict with good accuracy the steady state room curve in a typically reflective room. However the reverse is not true. The spinorama can identify good or bad loudspeakers - this capability is seriously compromised if in-room measurements are all that is possible.
“Third, a loudspeaker that is not flat on axis but which has well-behaved directivity as a function of frequency can benefit from in-room equalization, but in order to know that one needs comprehensive anechoic data. If one had such data, the optimum equalization (above the transition frequency) would be based on the anechoic data, not a room curve.“
(https://www.audiosciencereview.com/...r-favorite-house-curve.2382/page-5#post-85321)

 

[jhaider]

Back to KH 80 DSP: my first thought on my first listen to KH 80 DSP was "wouldn't that be a great 'head' for a modern day KEF 105/107 interpretation?”

Unfortunately I have not had much seat time with them, for two reasons.
First, strike one for the "pro stuff is more reliable" hypothesis, unfortunately. One had an issue with the on/standby circuit. It wouldn’t stay in standby. Any slight vibration would trigger it back on. So it was sent back to Sennheiser for warranty service last week. I chose repair to exchange because Neumann sent me a shipping label for warranty. There's no need to burden Sweetwater with a defect return when the manufacturer will handle it reasonably.
Second, the app is missing features that are important to me. The available highpass filter is fixed at 2nd order. I would like experiment with odd-order. Andrew Jones is adamant that subwoofer crossovers should be odd-order so the power response subs flat. He uses 5th order in the ELAC products that integrate subwoofers and correct the response. The app's frequency window is also fixed, which means there is little screen real estate for fine-grained correction in the modal and bass regions. Lastly, you can’t import your spatial average from FuzzMeasure/REW/ARTA and apply filters directly to the measured response, as miniDSP and Hypex allow. Neumann wants to sell their own forthcoming microphone interface, so this will not be fixed. So I had to change up the signal chain. Instead of using a presumptively clean pro interface (Apogee Duet or Focusrite Clarett) I plugged everything into a spare miniDSP DDRC-24D to perform for bass management and bass/modal correction. That required making RCA-to-XLR cables.

Also, @Krunok: quasi-anechoic measurements are not that difficult, though they do require some space for resolution down to the transition region. You want to clear as much of the room as you can, and probably aim the speaker diagonally to maximize reflection pathengths. Just set up to measure indoors - clear as much space as you can, try to get the reference vertical axis (ask the speaker maker; usually it's tweeter or between midrange and tweeter) centered floor to ceiling. Try to measure from as far away as possible if the speaker has lots of drivers or wide center-to-center spacing. IMO the ideal is measuring distance just a little bit shorter than the first reflection (usually floor/ceiling) arrival. I think JA does ~55”. After you shoot your measurement, window out reflections in the impulse response. Then turn the speaker 10 degrees and repeat, until it's firing 180 from the microphone, or turned all the way back around if the speaker's not symmetric. Happily, with speakers in real rooms all the resolution you lose is in the transition region and below, which you're going to correct with EQ measured where you and your friends/family sit anyway.

PS: I can’t find it here but I recall some discussion earlier about Neumann's studio monitor drivers. It looks like Sennheiser pulled their new soundbar's drivers from the Neumann KH parts bin, or at least based them on Neumann parts.

The woofer cones are perforated to vent the voice coil. The vents are sealed from the front by the dust cap. Tymphany often uses this design instead of a pole piece vent behind the motor.

 

[q3cpma]

It might seem useless to chime in for this, but where did you get the idea that Neumann wasn't lauded here? As one of the manufacturers giving a lot of useful product data, if not the most exhaustive, I'm sure they're appreciated here.
The only thing that could give this idea is the strange audiophile-like passion for passive technology in here and the arguably over the top Harman/JBL fanboyism eclipsing the other brands.

 

[mitchco]

Hey @jhaider Jay, did you stew on that for two weeks since I posted that? Lol! Thanks for the teacher’s edition, but I knew that from Floyd’s 2002 paper. I have no disagreement with any of @Floyd Toole research or publications. So who’s condescending who here man? Again, seems personal.

You are missing my point Jay. Have you downloaded and read: https://www.audiosciencereview.com/...adphones-or-speakers.7910/page-10#post-219508

As an Fyi:

From: History of Harman Target Curve. There are several other AES papers referenced, as well as links I posted earlier. Some great reads there.

Based on much of the subjective/objective listening tests that Sean Olive and friends have documented and repeated over and over again, there is indeed a “preferred” in room target response for what makes for a good sounding loudspeaker in a room (and headphone for that matter, including the wonderful NAD Viso HP50).

The predicted in-room response (PIR) can be predicted, quite accurately, based on the ANSI/CEA-2034-A Standard Method of Measurement for In-Home Loudspeakers I posted above., which has a great section on the Basis for Estimated In-Room Response Calculation.

My point is that some speaker manufacturers are not quite up to date on using the latest techniques, even though they say they are. In other words, I check the in-room response to see if a speaker manufacture is full of shit or not. As I have pointed out in my review of speakers, some hit the Harman target and some don’t. For example, the Kii THREE is about +5 dB hotter from 2 kHz to 20 kHz as compared to the D&D 8c using out of the box settings. The D&D 8c and the KEF LS50 being speakers that hit the Harman target (quite accurately) out of the box, whereas The Kii’s, Dynaudio 600 XD, Devialet Phantom and several others that I have measured have "the same" treble lift characteristic, including several unnamed studio control room monitors. Same issue with DSP products – see attached.

So the only conclusion one can draw is that not all anechoic measurement techniques are the same. Even though the anechoic data looks the same. The loudspeakers produced using CEA-2034 or before it was a standard, sound neutral to me ears and I “prefer” those types of loudspeakers. It is also the same (translated) target response as the HP50 headphones. When I switch from my speakers to these headphones, the tonal balance sounds the same to me.

Personally, I have nothing against the well designed and engineered Neumann’s, even though in the past I have measured some of their monitors to fit in the bright sounding (to my ears) category. I agree with @q3cpma about where did this idea about Neumann not being lauded here? Not only is it much ado about nothing, and again, seems personal to me.

PS. My repeatable in-room measurement process is well documented in my book referenced in my sig and too long to post here. But it does involve calibrating levels and distance, the latter down to a single sample offset between two speakers at the LP. But who cares! Not my point in this thread.

Over and out.

 

[GrimSurfer]

It might seem useless to chime in for this, but where did you get the idea that Neumann wasn't lauded here?

Thanks for expressing in words what I was feeling. To take this a bit further, I don't get too excited about brands.

Products? Sure. When they spec, measure, and perform well.

But brands? Not really. Depending on the motivation, each is capable of producing greatness and utter shite. Occasionally at the same time.

 

[q3cpma]

I feel the same about this, but there are some few I really consider trustworthy. I could almost buy blindfolded from Genelec, Neumann and especially Yamaha since I trust them to never make something bad from a price quality ratio point of view. Well, not that I buy blindfolded, since the specs and technological marketing is what made me trust them in the first place.

Isn't there at least one brand you consider honest and good?

 

[GrimSurfer]

I feel the same about this, but there are some few I really consider trustworthy. I could almost buy blindfolded from Genelec, Neumann and especially Yamaha since I trust them to never make something bad from a price quality ratio point of view. Well, not that I buy blindfolded, since the specs and technological marketing is what made me trust them in the first place.

Isn't there at least one brand you consider honest and good?

I hold Benchmark and Bryston in high regard but would never buy sight unseen. Otherwise I'd be like Mc owners, who seem to ignore the "misses" that company makes from time to time.

I'm more agnostic about speakers, though I'm inclined towards brands whose manufacturers have proper anechoic chambers for product development.

 

[Cbdb2]

First I would like to say hello to everyone, this is my first post. Im a recording engineer ( 25 years) with an EE degree. Just had to add another pro manufacturer that helps advance the science and is conspicuously missing from this thread, Meyer Sound.

 

[jhaider]

@mitchco
1) My most sincere apologies for having a life and a job outside of audio forums.

2) Thank you for the free CTA-2034A link. I almost bought it from ANSI for a hundred bucks last month, so you've saved me money. As best I can tell, your position is based on logical flaw of seeing X then Y, and inferring Y then X from it. The 2034 standard describes acquisition and display of standard anechoic or quasi-anechoic measurements. That standard battery of measurements can be used to generally predict in-room response (X->Y). However, 2034 is silent as to the predictive power of measurements taken in an unclear and undisclosed way in some random room (Y->X).
Your "so the only conclusion one can draw is that not all anechoic measurement techniques are the same" statement flows from this reasoning error. "In room measurements are not a consistent way to evaluate speaker performance above the transition reason" is a better conclusion one can draw from your anecdote and other available information, such as the graph of the same Genelec monitor in different rooms in Sound Reproduction, the Olive curve-drawing study Dr. Toole describes, and so on. Here, your invocation of Dr. Toole seems little different to me from Magnepan's video discussed in another thread.

3) I think it's reasonable to write Neumann products aren't "lauded" here. There is a simple reason for that: Neumann does not market to home users unlike Kii or Dutch&Dutch, and they don't have a "crossover hit" analogous to JBL M2 or 3-Series. Most of us don't work in studios and don't have much visibility into equipment used in them, so those speakers are not generally "lauded" here. I bet Sennheiser-branded versions of Neumann speakers with Airplay front ends would likely be highly lauded here and, depending on price point, commercially successful. Assuming their on/standby circuits work reliably!

4) Your Kii Three comments are really confusing. Are you stating that the standard (i.e. anechoic/quasi-anechoic) on-axis response of your sample Kii 3 was 5 dB hot in the treble, or that some in-your-room measurement made with unclear and undisclosed spatial averaging was 5dB hot? The distinction matters.
Thankfully, I did not see any Sharpie scribbles on any graphs in a skim of your review.
However, I also did not see any useful universal anechoic or quasi-anechoic measurements, except for the polar map reproduced from AudioXpress. All I saw was in-your-room measurements taken with unclear and undisclosed spatial averaging, that did even not show the claimed HF rise.
Here's are the findings on the Kii Three by reviewers who measured the speaker with universal, standard techniques and traditional disclosure of measurement practices have found:
Atkinson - slightly declining treble over a 30º horizontal average.

Single-point on axis, anechoic (Wilke) and quasi-anechoic (Colloms) show a slight on-axis treble rise:
https://www.audioxpress.com/files/attachment/2609
https://www.kiiaudio.com/media/GENERAL/docs/reviews/hificritickiithreeen.pdf
The tension is resolved by considering directivity. It seems to me that the 2034 drafters were wise to redefine DI as referenced off of the listening window response.

5) Compared to Neumann KH 80 DSP or JBL LSR705X, KEF LS50 has lower performance and is also less practical for home use (no grille). I've heard the standard ones multiple times (unfamiliar environments) and never ran from the room, but also never heard anything compelling enough to bring a pair in for more detailed evaluation.

 

[Ilkless]

First I would like to say hello to everyone, this is my first post. Im a recording engineer ( 25 years) with an EE degree. Just had to add another pro manufacturer that helps advance the science and is conspicuously missing from this thread, Meyer Sound.

Because they talk a big talk but don't have any data remotely close to the level of detail Genelec and Neumann have. Nor are their design choices remotely evidence-based (eg. mating a large woofer to a direct radiating dome tweeter on a flat baffle, as in the HD-1 and its ilk). I could only find primitive off-axis measurements on a white paper from Meyer I can't find anywhere else, and only hosted on a dubious third-party site. These measurements of the HD-1 are wholly unremarkable and very, very far from the level Meyer conceives itself to be, and is imagined to be.

 

[Krunok]

Because they could not eliminate it without either changing the acoustic design (higher tuning, bigger cabinet) or raising build cost substantially (passive radiators).

Would transmission line design produce better results?

 

[Erik]

I could almost buy blindfolded from Genelec, Neumann and especially Yamaha since I trust them to never make something bad from a price quality ratio point of view.

Yamaha HS5 are one of the worst 5-inch monitors on the market. Their other products are not that bad, but not on Genelec or Neumann level for sure.

HS5 on top right.

 

[q3cpma]

Well, you aren't wrong but not completly right. While the LSR305 has a way better frequency response, it also has a very cheap cabinet full of resonances and a badly/cheaply made class D amp known for its hiss. I'd still take the LSR over the HS5, but it's not a clear cut which one is the best (a cumulative spectral decay comparison would be nice too).

That 1 kHz peak and 3-4 kHz dip look quite suspect, though, and aren't that prominent in the data given by Yamaha. I still agree that the HS5 is curiously bad considering how good (for their price) the HS7 and HS8 are; as if it was an afterthought.

By the way, I didn't manage to find any data provided by JBL, even after downloading the user manual. Even the frequency response range is given without any amplitude ("43 Hz – 24 kHz"). It's a bit disappointing for a brand known for its research.

 

[Vintage57]

The only thing that could give this idea is the strange audiophile-like passion for passive technology in here and the arguably over the top Harman/JBL fanboyism eclipsing the other brands.

Well said, I couldn't of put it better.

One company, Harman/JBL chose to stay out of the active arena and I for one think of this as a disservice to the customer. I've owned some actives, ATC's (good) Meyer's (not as good) and Neumann's (very good). Some passives, B&W, Revel, Dunlavy, JBL, KLipsch, this over the past 30 years and know what I prefer. Not that I believe passive speakers are bad, just incomplete.

Who better to chose the engine for the car than the manufacturer. Even if it is not their own manufactured engine they would have it built to their specifications.

Very few speaker manufactures make their own drivers. Why not make driver selection and wiring choices an option in the future as well lol. A choice of drivers and cone/dome materials. We could add to the cable wars and end up too busy to listen to the music.

I'm not attempting to restart the debate, YMMV.

 

[audiophool]

Question for

Based on much of the subjective/objective listening tests that Sean Olive and friends have documented and repeated over and over again, there is indeed a “preferred” in room target response for what makes for a good sounding loudspeaker in a room (and headphone for that matter, including the wonderful NAD Viso HP50).

There are actually multiple target curves, depending on the preferences of the listener.

 

[Ron Texas]

Question for


There are actually multiple target curves, depending on the preferences of the listener.

Research at Harmann produced statistically significant prefefrences. That's what a manufacturer wants, happy customers in the long run. Of course there will be those with minority tastes and manufacturers who voice bright for a quick comparison in the showroom followed by listening fatigue at home.

 

[Ron Texas]

Well, you aren't wrong but not completly right. While the LSR305 has a way better frequency response, it also has a very cheap cabinet full of resonances and a badly/cheaply made class D amp known for its hiss. I'd still take the LSR over the HS5, but it's not a clear cut which one is the best (a cumulative spectral decay comparison would be nice too).

That 1 kHz peak and 3-4 kHz dip look quite suspect, though, and aren't that prominent in the data given by Yamaha. I still agree that the HS5 is curiously bad considering how good (for their price) the HS7 and HS8 are; as if it was an afterthought.

By the way, I didn't manage to find any data provided by JBL, even after downloading the user manual. Even the frequency response range is given without any amplitude ("43 Hz – 24 kHz"). It's a bit disappointing for a brand known for its research.

305's measure with a downward response in room, especially if the switch is set to reduce HF. Below 200 it's mostly the room changing frequency response.