Neumann KH420 Review (Studio Monitor)

[FeddyLost]

In this video the active SCM50 sound much better

I mentioned TRI-amp. So, 3 different and expensive amps through existing passive crossover was used to subjectively overplay standard amppack. Not single P1/P2.

If everything is done right in active integrated speaker, there is no easy way to outperform it even with chip-amps.
But people try sometimes if they have resourses.

 

[hardisj]

Ah - why doesn't Erin use the technique deployed for single drivers for speaker-boxes also? Yes, it would, see above, need to seek out for design-specific problems. Hence it wouldn't be the same for all, and is thus considered "unfair"?

This has been asked of me a few times already. Here is a quote reply from one of those instances:

I don’t have a way to perform those anechoically (for a loudspeaker). That means that any influence in the room would override the true results of the speaker. For drive unit testing it’s fine because you can get in the near field and negate the room. But with a loudspeaker you need to be far enough back to capture the summed response of the drivers, and unfortunately when doing so you also capture the the room.

For HD testing I can use a room correction curve built by using Klippel’s ISC module so it’s not a problem.

 

[Tangband]

The 2 versions don't have the same the crossover transfer functions and thus the FR and directivities are different, so this result cannot be used for generalisations like "active is better". Active can be better but due to other reasons, mainly as more complex crossover functions can be implemented without getting insanely complex or expensive.

There is more - active almost always has a big advantage mainly in the bass region, in the EMF area where most passive solutions is wastly inferior . In the bass region , the directly coupled low resistance to the amplifier will control the bass in a better way. Its audible .

Youre right about that an active system can be really bad if not done right. But rightly done, its a much better solution than passive .

 

[DSJR]

In this one brand, I don't believe it's possible to 'do' the passive crossover as well as the active version does, so there will be compromises in the passive model which at the time, I found unacceptable. The active amp pack then wasn't too expensive to retro-fit and at the time, factory visits to drop off and/or collect the speakers by appointment was a huge advantage, as the owners could get better immersion into their chosen speakers if they wanted to (why I'm still so fond of the brand). No idea if Neumann and Genelec are thus equipped for visits and perhaps today, Youtube factory tours can remove this load from them?

I say the above as someone who loved the chance to poke around a favourite manufacturer and get to speak to and learn from the powers that be. The one manufacturer I didn't get the chance to officially visit was Quad in Huntingdon, as I was between jobs and politically, I couldn't go (I know, I'm sure I could have visited some other time, but this was the era of the then brand new ESL63 and I suspect factory visits were done in groups).

 

[FrantzM]

Exactly.
And there is an opinion that just good measurements are not equal to ultimate sound quality. Gladly, I can't verify this, because my qualification is worse than measurement mic and I follow "better measurements" way.
But if I'd have time, resources and competence, it would be nice to play with "already good enough" design .. maybe.
Some people mentioned that they was able to tri-amp passive big ATC in the way that they outperformed actives. But at ridiculous price of equipment attached, for sure.

With such price.
Of course, I'm sane.

Not obvious, I think. Otherwise it wouldn't be accepted by customers.
But I really think that every industrial equipment can be slightly improved ... if the cost is not an option.

Bold emphasis is mine and ... I'll drop the discussion...

Some people mentioned

Again? Seriously ???

 

[Aperiodic]

Remarkable. I put this under the category of 'See, it can be done.' Is the $5.2k price for one or two of these gems?

I finally found this info after some scrolling, but how about standardizing on either 'each'or 'per pair' pricing and including that info in the opening post? I suggest 'per pair' as many speaks are not sold as single units (with the exception of centers) and are probably not bought that way even if they are.

 

[thewas]

There is more - active almost always has a big advantage mainly in the bass region, in the EMF area where most passive solutions is wastly inferior . In the bass region , the directly coupled low resistance to the amplifier will control the bass in a better way. Its audible .

I agree that its theoretically active is better there, though it must be also said that the usually passive crossovers don't have much resistance in bass branch passband as no attenuators are usually used the corresponding circuits. Also the control can be increased by other techniques, like current drive or motional feedback. Now about the audible differences, I think they are overestimated and I haven't read of any scientific proof (like blind test comparison) between them.

Youre right about that an active system can be really bad if not done right. But rightly done, its a much better solution than passive .

Also there my answer is quite different, active can be much (also audibly?) better for difficult cases, but not for most of them, but on the other side it also does not harm.

 

[fineMen]

This has been asked of me a few times already. Here is a quote reply from one of those instances:

First of all, thanks for all the effort You already take.

I personally don't doubt the outcome of my own IM measurements. To the contrary, they led me to rethink my design paradigms altogether. I'm happy with the results of the new set of beliefs (multi-multi-way etc). Reasoning is, the speakers won't be listened to in a sterile environment. Hence a measurement directly in that filthy room acoustics may appear quite right. It is discerning, but admitted, not that presentable.

Example: 12" JBL PA/studio bass in 60 liter ported design versus one 7", 40y old quite humble "hifi" driver in double tuned bandpass experiment. IM over a naturally small bandwith was different by a linear factor of 3 in favour of the bandpass. I tend to believe the outcome, and so to the day the sub-par bandpass suports the already quite noble JBLs.

Again, an in-room measurement of a critical quality parameter for sure is corrodible (I looked that up). People might easily tell, it wasn't valid. Mostly, me thinks, because they cannot replicate and verify such an attempt on their own, and so badly feel the need to just believe--or not.

 

[hardisj]

HD alone with in-room vs anechoic can be vastly different. I actually just wrote a small bit about this elsewhere with 3 examples.

For a DIY'r that's fine. It gives you an idea. And you know your measurement rig/room. For someone like me, I try to provide the most "true" results I possibly can. I try to avoid having to provide measurements that need caveats that the general public won't understand. Thus, my reticence for providing loudspeaker IMD. I've been toying with the idea again, though.

 

[Spocko]

It's all about playing loud to "reference" levels, way beyond what mere mortals would ever need - simply amazing!

 

[Vintage57]

It's all about playing loud to "reference" levels, way beyond what mere mortals would ever need - simply amazing!

I’d compare it to 0-60 and other data for cars and what they are capable of. As you have said “beyond what mere mortals would ever need”.
However by pushing the limits we get trickle down innovation that mere mortals can enjoy.

 

[Spocko]

I’d compare it to 0-60 and other data for cars and what they are capable of. As you have said “beyond what mere mortals would ever need”.
However by pushing the limits we get trickle down innovation that mere mortals can enjoy.

YES, like it's smaller brother the KH310 which captures all the magic without requiring a gym membership to lift on stands.

 

[Pearljam5000]

What basically does it mean?

 

[changer]

That the vertical pattern is designed to be narrow, to reduce early reflections like floor or ceiling bounce.

 

[Sancus]

For a DIY'r that's fine. It gives you an idea. And you know your measurement rig/room. For someone like me, I try to provide the most "true" results I possibly can. I try to avoid having to provide measurements that need caveats that the general public won't understand. Thus, my reticence for providing loudspeaker IMD. I've been toying with the idea again, though.

The main problems with IMD to me are:
1) No standardized way of testing this. Different sources use different methods. S&R's methods make their results incomparable even with other speakers they've reviewed so it's very strange...
2) No research I've seen, or even anecdotal evidence, that IMD would be audible in cases where HD would not, with the exception of a report from Harman about audible IMD+doppler distortion in older Kef 2-ways specifically. But you've already covered the doppler distortion issue(and thanks for doing those measurements, I haven't seen anybody else even attempt it).

Given these factors I haven't a clue how anybody is even supposed to interpret IMD to correlate with anything audible. It's hard enough with HD, which, as typically depicted, is already incredibly difficult to interpret in terms of correlation with audible problems.

 

[Tangband]

The main problems with IMD to me are:
1) No standardized way of testing this. Different sources use different methods. S&R's methods make their results incomparable even with other speakers they've reviewed so it's very strange...
2) No research I've seen, or even anecdotal evidence, that IMD would be audible in cases where HD would not, with the exception of a report from Harman about audible IMD+doppler distortion in older Kef 2-ways specifically. But you've already covered the doppler distortion issue(and thanks for doing those measurements, I haven't seen anybody else even attempt it).

Given these factors I haven't a clue how anybody is even supposed to interpret IMD to correlate with anything audible. It's hard enough with HD, which, as typically depicted, is already incredibly difficult to interpret in terms of correlation with audible problems.

If IMD didnt exist, then a singel coaxial driver should sound exactly the same as a four-way active loudspeaker, even at rather high volumes.
Thats not the case. Theres measurement evidence that shows when a coaxial driver is moving, its acting as a variable waveguide modulating the tweeter response. If you have many drivers, such as in an active four-way monitor, each driver is playing less frequensies and they do that without disturbing the other drivers. To measure IMD distortion with a loudspeaker, you have to use measurement signals thats a bit like real music, with information in the whole freq spectra from 40 -20000 Hz at the same time, and measure this at a normal listening level.

One can measure those artefacts in fullrange coaxials, but its not showing up with the klippel measurements system. The artifacts are very audible with music though.

This shows what happens in a full range coaxial design with the cone moving 3 mm. Its acting as a variable waveguide, modulating the treble. Listening to music with both bass-playing and cymbals at the same time gonna produce this results.

 

[Sancus]

If IMD didnt exist, then a singel coaxial driver should sound exactly the same as a four-way active loudspeaker, even at rather high volumes.

HD alone would prevent any hope of this.

Thats not the case. Theres measurement evidence that shows when a coaxial driver is moving, its acting as a variable waveguide modulating the tweeter response.

Why are you talking about doppler distortion as if I don't know it exists when I literally referenced it in my post. IMD and doppler distortion are two different things, btw. IMD occurs on *all* speakers, not just coaxials. Doppler distortion is unique to coaxials(or any other design where a driver's movement would modulate another driver's somehow).

Posting your personal thoughts is not the same thing as research showing audibility thresholds and what any of the measurements actually mean to the listener, sorry.

 

[tuga]

HD alone would prevent any hope of this.

Why are you talking about doppler distortion as if I don't know it exists when I literally referenced it in my post. IMD and doppler distortion are two different things, btw. IMD occurs on *all* speakers, not just coaxials. Doppler distortion is unique to coaxials(or any other design where a driver's movement would modulate another driver's somehow).

Posting your personal thoughts is not the same thing as research showing audibility thresholds and what any of the measurements actually mean to the listener, sorry.

Would you expect IMD to be audible "in isolation" from other types of distortion?
If that is not possible, would it not still have a detrimental audible effect?

 

[Frank Dernie]

This shows what happens in a full range coaxial design with the cone moving 3 mm. Its acting as a variable waveguide, modulating the treble. Listening to music with both bass-playing and cymbals at the same time gonna produce this results.

Did you model this with the main driver moving 3mm right up to the crossover frequency of 2kHz?

 

[amirm]

The main problems with IMD to me are:
1) No standardized way of testing this.

I have not even seen any standardization of frequency pairs. Even if they were, they may need to be changes due to which frequency range the coaxial driver is playing.

I spent days trying to see if I can get meaningful IMD results with my AP analyzer which is far more capable than Klippel but could not.

At high level, the speaker system has certain linearity. How that comes out with one or two tones is identical It can't act different under one measurement than another. To wit, if you have the phase information for distortion products, you can sum independent distortion measurements to get IMD. So the only reason to run IMD is because it can be more revealing as a measurement, not because it is testing something that is not otherwise.