Why don’t ASR members laud Neumann speakers?

[watchnerd]

Everything I could find on these speakers especially the Neumann model KH 420 should make them one of the most accurate sound reproduction devices available. The KH 120 and the KH 310 are no slouches in their own right.

Granted they are low on the WAF scale however the top shelf specs are reproducible and that’s saying a lot. This can’t be said for all of the newer generation of active monitors.

I’ve offered as have many other to support Amir’s speaker testing venture. I hope we’re collectively going to move forward with the project. The interesting fact is the outcome we’re aspiring to find is already available, at least one company is putting their money into audio science.

I’m going to pick up a pair and will find out what if anything i’ve been missing.

Will share my findings.

I've used Neumann KH420 in mixing / recording engineering environments.

They're very good speakers, but at that size and price, there are also a lot of other good active monitor speakers.

But for home use, I'd probably go with either (for less than the KH420) the Dynaudio Core 59 (full disclosure, I use Contour 20 in my living room and LYD5 in my music / mixing studio), or the (for more money than the KH420) Barefoot Micro Main 27, just because I prefer the sound of those monitors to the KH sound a bit when it comes to long term listening.

I also like having direct digital inputs. Last I saw, that was an add-on option to the KH.

 

[mitchco]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I spent a decade in studio control rooms with monitors like these and made sure to keep the volume down (or heaven forbid eq the treble down). Ear bleeders is right, especially if they are placed on the mixing bridge right in your face. If the band wanted to rock out, they gathered around the mixing console and I showed them where the volume is and I went out for a coffee or ...

It is still prevalent today and if I were purchasing one of these astoundingly engineered monitors (just look at the tech specs on the Neumann KH-310!), I would be sure to turn down the treble (alas not enough range built into the speaker system) with some sort of tilting tone control to at least sound pleasant and not rip your ears off. That's the fundamental issue with Genelec's GLM software - eq steady state response to flat. I am not singling out Genelec either as most pro monitors have the same issue. Contrast that with the D&D 8c that has the downward tilt "out of the box"...
It's no wonder folks have a preference for the 8c's, even if they don't know why.

 

[Totoro]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I spent a decade in studio control rooms with monitors like these and made sure to keep the volume down (or heaven forbid eq the treble down). Ear bleeders is right, especially if they are placed on the mixing bridge right in your face. If the band wanted to rock out, they gathered around the mixing console and I showed them where the volume is and I went out for a coffee or ...

It is still prevalent today and if I were purchasing one of these astoundingly engineered monitors (just look at the tech specs on the Neumann KH-310!), I would be sure to turn down the treble (alas not enough range built into the speaker system) with some sort of tilting tone control to at least sound pleasant and not rip your ears off. That's the fundamental issue with Genelec's GLM software - eq steady state response to flat. I am not singling out Genelec either as most pro monitors have the same issue. Contrast that with the D&D 8c that has the downward tilt "out of the box"...
It's no wonder folks have a preference for the 8c's, even if they don't know why.

Thanks, that was informative. I didn’t realize you couldn’t set the target in the GLM software,

 

[Ilkless]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I spent a decade in studio control rooms with monitors like these and made sure to keep the volume down (or heaven forbid eq the treble down). Ear bleeders is right, especially if they are placed on the mixing bridge right in your face. If the band wanted to rock out, they gathered around the mixing console and I showed them where the volume is and I went out for a coffee or ...

It is still prevalent today and if I were purchasing one of these astoundingly engineered monitors (just look at the tech specs on the Neumann KH-310!), I would be sure to turn down the treble (alas not enough range built into the speaker system) with some sort of tilting tone control to at least sound pleasant and not rip your ears off. That's the fundamental issue with Genelec's GLM software - eq steady state response to flat. I am not singling out Genelec either as most pro monitors have the same issue. Contrast that with the D&D 8c that has the downward tilt "out of the box"...
It's no wonder folks have a preference for the 8c's, even if they don't know why.

An anechoically flat on-axis speaker is likely to have a downward tilting steady-state room curve at the listening position though. Look at the Revel and JBL spins. Flat on-axis/within the listening window. Downward tilting early reflections and power response curves. All these speakers, in every instance I've seen them measured in room tend towards a downward-sloping steady-state curve. The Neumann speakers are not profoundly/fundamentally different from other well-engineered speakers like Harman's good stuff. I'm not sure why you'd argue otherwise, because this sounds like a slightly more nuanced (but nonetheless fraught with confirmation bias) version of the flat studio monitor = bright argument.

What is uncontroversially problematic is flat power response, which formed the Consumer Reports target that Dr Toole found to negatively-correlate with listener preference. Flat power response yields a bright direct sound because it means tilting up the response on-axis and within the listening window. Neumanns and Genelecs don't do this.

 

[watchnerd]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I spent a decade in studio control rooms with monitors like these and made sure to keep the volume down (or heaven forbid eq the treble down). Ear bleeders is right, especially if they are placed on the mixing bridge right in your face. If the band wanted to rock out, they gathered around the mixing console and I showed them where the volume is and I went out for a coffee or ...

All of this is one reason why I like the fact that the Core 59 has a tilt control EQ, implemented in DSP

"Users can also alter the overall presentation. Core 59 uses the same concept as the LYD series Sound Balance filter – a different design to typical shelving-EQ tweeter and woofer adjustments found on most active monitors. Instead of simply fine-tuning the tweeter level up or down, which will affect phase response, Core 59 uses a full-spectrum band-pass filter that tilts depending on the desired tonal response. In the ‘Dark’ setting, the entire frequency response is tilted so that 20kHz is down -1.5dB, while 20Hz is up +1.5dB. In ‘Bright’, the tilt is reversed (+1.5dB at 20kHz and -1.5dB at 20Hz). This gentle filter maintains the proper phase response between the drivers while providing the tone that meet your tastes. The changes are subtle, but often this little tweak is enough to make the sound balance just right."

 

[jhaider]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I don't understand your objection as applied to these speakers. The Neumanns I've seen measurements for have similar anechoic treble trendlines as Revel, M2, or 7-Series. KH 80 DSP specifically measured with a very slight anechoic tilt. From Sound und Recording's measurements they are about 2dB down at 20kHz re: 100Hz.

https://www.soundandrecording.de/equipment/studiomonitore-neumann-kh-80-dsp-die-messdaten/

For that matter, how about Adam studio monitors.

I mentioned on another thread that Adam had a touring demo when they introduced their current line. I only heard the 7" 2-way and topped MTM nearfield. The AMT seems more sensitive to head motion up close than dome tweeters are. I did not want to listen long.

At normal home listening distances, Adam's 3-way stand-mount sounded very good to me. They were $6000 or $7500 a pair. They're also in the Sound und Recording Studiomonitor Special PDF, at 74. At least in that treated studio room they sounded better to me than they measure. On axis is very flat. The only issue to note is a resonance just below 2kHz. The horizontal polars do betray the crossover points, though compared to most speakers they measure very well. Unfortunately, I could not get them into our home, because when I was having the living room wired I did not have the foresight to have XLR run to the speaker locations. I thought 18 channel of speaker cable (7 bed including triamp capability for mains, 4 heights, 4 subs) would be more than enough!

 

[Totoro]

As shown by the great research by @Floyd Toole and Sean Olive, a good sounding speaker has a downward tilting steady-state room curve. While there are several posts on this at ASR, including links to AES papers and other references, I am going to link to one of Floyd's responses that explains why the downward tilt. It also explains why Neumann, Genelec and many others don't get as much love with his key statement, "A fundamental problem has been the incorrect assumption, made long, long ago (in the age of RTAs), that the audio rule "flat is beautiful" should apply to steady-state room curves, not the direct sound." Can you dig it?

I spent a decade in studio control rooms with monitors like these and made sure to keep the volume down (or heaven forbid eq the treble down). Ear bleeders is right, especially if they are placed on the mixing bridge right in your face. If the band wanted to rock out, they gathered around the mixing console and I showed them where the volume is and I went out for a coffee or ...

It is still prevalent today and if I were purchasing one of these astoundingly engineered monitors (just look at the tech specs on the Neumann KH-310!), I would be sure to turn down the treble (alas not enough range built into the speaker system) with some sort of tilting tone control to at least sound pleasant and not rip your ears off. That's the fundamental issue with Genelec's GLM software - eq steady state response to flat. I am not singling out Genelec either as most pro monitors have the same issue. Contrast that with the D&D 8c that has the downward tilt "out of the box"...
It's no wonder folks have a preference for the 8c's, even if they don't know why.

I had thought from this part of the blurb about the GLM software that I would be able to set a target curve,

• Use the personalisation feature to customise your preferred response - without sacrificing predictability.

Is what you’re saying that I can’t? That seems like it would kill my enthusiasm.

 

[Blake Klondike]

What I don't know about the physics of audio would fill a book, but from the manual for my Genelec 8020b speakers:

"Tilt Treble Tilt control (switch 1) attenuates the treble response of the loudspeaker at frequencies above 5 kHz by 2 dB, which can be used for smoothening down an excessively bright sounding system."

Wouldn't this address any perceived brightness built into these speakers, if their flat response does indeed cause it?

 

[Blake Klondike]

Speakers seem to be the hardest component in a system to get a handle on to me. So many variables, and the electro-mechanical nature of their design makes design and craftsmanship so important! This is even before the stupid room we have to sit in gets involved in the equation.

 

[watchnerd]

Speakers seem to be the hardest component in a system to get a handle on to me. So many variables, and the electro-mechanical nature of their design makes design and craftsmanship so important! This is even before the stupid room we have to sit in gets involved in the equation.

If you think that's bad, try recording (not measurement) microphones, with different radiation patterns, colorations, etc.

 

[pierre]

Thanks, that was informative. I didn’t realize you couldn’t set the target in the GLM software,

I have a glm 3 and you can set the target you want. You can also have multiple curves optimized for and;switch between them quickly while,having a consistent loudness.

 

[Totoro]

I have a glm 3 and you can set the target you want. You can also have multiple curves optimized for and;switch between them quickly while,having a consistent loudness.

Thanks, that’s what I had thought

 

[mitchco]

Re: Genelec - yes, you can adjust the target - I wonder how many people do that. Maybe just the folks on this forum ;-)

-2 dB @ 20 kHz... Well, don't shoot me, I am talking about -10 dB down at 20 kHz folks. A bit of reading here.

@Ilkless I agree with your 2nd paragraph, but not your first. I have measured in-room Revel 208, Revel Salon2 and JBL M2 and confirm the nicely tilted response to about -10 dB at 20 kHz (ref 20 Hz). Not so with a couple of Neumann's I have measured and can confirm -2 dB at 20 kHz as per article you linked. The Revel 208 measured in-room response -10 dB @ 20 kHz also confirmed by Sean Olive along with his study on room correction products in the link above.

I am not saying one size fits all and it is not always a tilt exactly from 20 Hz to -10 dB at 20 kHz. As stated here, unless one is using the spinorama method as embodied in the ANSI/CEA 2034-A Standard Method of Measurement for In-Home loudspeakers, then I suspect that the speaker designer/manufacture is doing it wrong.

PS the D&D 8c also tilts 20Hz to -10 dB at 20 kHz "out of the box" as measured in-room in my previous post. Just sayin...

 

[CTRLM]

-2 dB @ 20 kHz... Well, don't shoot me, I am talking about -10 dB down at 20 kHz folks. A bit of reading here.

Hi Mitch, I've followed your progression with DRC for quite a while. If the OP doesn't mind an off topic question that I've been wanting to ask you for a while....does the target slope depend on the room and how dead or alive it is? I've tried the straight line to -10 db at 20 kHz using Dirac in my situation and it sounds quite dull on most material, where -6 or 7 db is more balanced. My room is carpeted with lots of soft furnishings and absorption on the wall behind the listening position. There is also a minimum of 2 meters between the speakers and side walls.

 

[jhaider]

Re: Genelec - yes, you can adjust the target - I wonder how many people do that. Maybe just the folks on this forum ;-)

-2 dB @ 20 kHz... Well, don't shoot me, I am talking about -10 dB down at 20 kHz folks. A bit of reading here.

Perhaps you don't read German, but if you did you would have noted that cited -2dB @ 20kHz cite was an anechoic, not in-room, measurement. Generally, when people compare speaker measurements we use anechoic or quasi-anechoic measurements. You're the only one I know who compares in-room measurements.

By comparison, according to Harman's published Spins, re: 100Hz
-M2 is a couple dB hotter at 20kHz, though it has a larger waveguide exit so power response falls off relatively more
-708 is a little hotter at 20kHz, and because the waveguide exit is so small the listening window almost exactly tracks the on-axis
-705 has a bump at 20kHz but is generally flat or just slightly down in the top octave. A little hotter than KH 80 DSP. Again, because the waveguide exit is so small the listening window almost exactly tracks the on-axis
-Revel F208 (Spin on your page) is hot on axis and less hot but still hot over the listening window

PS the D&D 8c also tilts 20Hz to -10 dB at 20 kHz "out of the box" as measured in-room in my previous post. Just sayin...

Your comparable (sort of - 30cm measurement distance?) measurement of the D&D 8C is this one:

 

[mitchco]

@CTRLM ideally speaking, if the speaker is properly designed, then one only needs to compensate for room modes below 500 Hz like here: https://www.audiosciencereview.com/...the-magnepan-30-7-road-tour.8451/#post-216435 My "general" complaint is that most speakers don't seem to be properly designed and based on the available research on what sounds neutral, most to my ears anyway, sound too bright, as mentioned in my post above. But then again, I mainly listen to (unfortunately) highly compressed music that tends to add brightness.

How much tilt depends on the things you mention, plus listening distance, off axis response, constant directivity eq, type of music being listened to, etc., and of course preference I don't know if Dirac can do partial correction, like to 500 Hz or lower only?

 

[mitchco]

@jhaider sorry,I don't read German. And that's the wrong measure for the D&D 8c - that's at 30cm as labelled on the chart - keep reading
This is the 8c at the listening position:

 

[Blumlein 88]

@CTRLM ideally speaking, if the speaker is properly designed, then one only needs to compensate for room modes below 500 Hz like here: https://www.audiosciencereview.com/...the-magnepan-30-7-road-tour.8451/#post-216435 My "general" complaint is that most speakers don't seem to be properly designed and based on the available research on what sounds neutral, most to my ears anyway, sound too bright, as mentioned in my post above. But then again, I mainly listen to (unfortunately) highly compressed music that tends to add brightness.

How much tilt depends on the things you mention, plus listening distance, off axis response, constant directivity eq, type of music being listened to, etc., and of course preference I don't know if Dirac can do partial correction, like to 500 Hz or lower only?

This brings up a good topic. Once you have your speakers close to transparent is that enough? I know I keep more than one room correction curve. With basic pop music, I basically double up on the in room downward tilt (roughly 3db per decade above 200 hz). This seems to be an improvement on lots of pop music with the brightening effects of compression and the low treble excess of studio condenser microphones. It seems to do little harm and makes a much wider swath of regular recordings highly enjoyable for the music. Do you have any cogent thoughts in this regard?

I of course know this departs from simple fidelity to the source veering into personal preference and there is no "correct" answer once you do that other than you like the result or you don't.

 

[Mtbf]

As for other companies that are known to have extensive R&D budgets for consumer grade speakers, it's already widely known that the Harmon Group does this. As a matter of fact, Dr. Floyd Toole, who is/was the head of Harmon research, is a member of ASR...

Harman Group.

 

[jhaider]

@jhaider sorry,I don't read German. And that's the wrong measure for the D&D 8c - that's at 30cm as labelled on the chart - keep reading
This is the 8c at the listening position:

View attachment 31862

Why do you keep bringing up out-of-the-scope listening position measurements when everyone else is clearly referring to anechoic measurements as commonly shown on data sheets and 3rd parry testing? It’s frustrating trying to keep time with someone who insists on clapping on every fifth upbeat for no discernible reason.

Or have you found some crazy physics-defying magic in current Neumann speakers that causes them to defy physics and perform differently in room (whose room?) than other speakers with similar anechoic on axis frequency response and dispersion measurements?

That said, I recently submitted a review of a studio monitor based speaker that has the kind of power response you’re railing against. You may enjoy reading it when it drops