• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Neumann KH 80 DSP Monitor Review

ctrl

Active Member
Joined
Jan 24, 2020
Messages
279
Likes
959
Location
DE
use the - 10 degree vertical angle as the reference?
Yep, the vertical -10° would do much better.

Compared with the on-axis measurement of the KH 80 by Neumann and Sound&Recording it does nearly the job:
KH80_diff_measurements_3.jpg

The reason for the difference at low frequencies we know.

Whats a bit bothering is the "small comb filter effect" between 3kHz and 10kHz. Could this be some kind of reflection from the Mic/Measurering-gear?
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,808
Location
Zg, Cro
Whats a bit bothering is the "small comb filter effect" between 3kHz and 10kHz. Could this be some kind of reflection from the Mic/Measurering-gear?
This "comb filtering" is in the amplitude range of +/- 1dB. Don't you think this can be due to sample to sample variation? Or difference in curve smoothing?
 

carlob

Senior Member
Forum Donor
Joined
Aug 4, 2019
Messages
452
Likes
685
Location
Roma, Italy
Perhaps it would be a better service to the forum to move all this endless stuff in a new thread for the people interested. I have to agree with @amirm and am massively uninterested in splitting hairs or obsessing on 1dB, that doesn't add anything to the review.
 

edechamps

Addicted to Fun and Learning
Forum Donor
Joined
Nov 21, 2018
Messages
595
Likes
2,368
Location
London, United Kingdom
If you want to dig in, how about asking the manufacturer what "interpolated" means on their graph? What samples were used and whether customer units are tested to comply with that spec and if not, they accept returns. Ask them what the frequency resolution is. What lab and technique was used to make those measurements.
Oh come on. @napilopez posted a comparison of measurements from 3 other sources, not just the manufacturer. These three measurements all agree with each other, and they all consistently disagree with yours in a very clear, hard-to-ignore way. Are you seriously suggesting that it's them we should investigate?

Collectively, you all are taking what started as a fun hobby to measure speakers into a miserable back and forth over things that don't matter. You are welcome to keep going but know the costs as you do so....
Again, I don't see the point of going through all that trouble with acquiring and setting up the NFS at great expense if you're not willing to investigate possible accuracy issues with the test protocol. And yes, as I previously explained, frequency response anomalies in the form of broad low-Q tilts do matter a lot.

Such debates are to be expected when creating a test methodology from scratch using novel hardware, especially for something as complicated (and exciting!) as measuring loudspeakers. Let me remind you of what you wrote yourself when you embarked on this whole speaker measurement journey:

Anything worth doing, is worth doing well.
And no, just because it works fine with other speakers doesn't mean the protocol is fine. It just means other speakers are less pathological cases. For example, in the case of the Revel C52, I'm guessing it would take more to make it compress, and that its reference axis is aligned with the twitter; therefore, such shortcomings in the measurement procedure would not make a difference there, but they would result in anomalies on other speakers such as the one we're discussing here. The reason why the case of the NH 80 is so interesting is because it's the first speaker you reviewed that has (relatively) many third party measurements to draw comparisons from.

You are welcome to keep going but know the costs as you do so....
If the outcome are measurements that we can actually trust and that agree well with other sources, then the cost is worth it, in my opinion. It would really be a shame if every time I see a speaker review and I see something strange like recessed bass or bright treble, I have to wonder if that's because the speaker really has issues or if it's just a measurement artifact.

If you end up reviewing something like a Genelec 8030 for example, and you use the same procedure as you used here (high output level, measured at tweeter height), I'm willing to bet you'll spark the exact same debate all over again, because, just like the KH 80, it's a small speaker that is prone to bass compression and its acoustical axis is not the tweeter. And once again, people will point out that your measurements don't agree with anyone else's, and they'll be right to do so.

But anyway, it's your hardware, your time, your rules, so I rest my case. And just to be clear, I still think what you're doing is highly valuable, that you should be commended for it, and that you should continue doing it. I'm just sad that you seem to be going for quantity over quality, but I'd still take these reviews over "buying blind" any day :)
 

Absolute

Senior Member
Forum Donor
Joined
Feb 5, 2017
Messages
382
Likes
575
On-axis is not interesting, listening window averaging is. As far as I can see, the Neumann measured as expected with a flat listening window per Dr. Toole's recommendation. The only non-flat area is around the crossover at 1,8 khz, which also happens to be around the same area where you get the typical stereo cancellation in a common 30 degree equilateral triangle. It should make for an interestingly dull experience in that particular area if not adjusted for.

In either case, lowering the reference axis will, if anything, make the listening window tilt a little bit more downward because the new on-axis reference will be flatter, but it will still keep the same dip in the crossover. As far as I can see, this review is quite perfect.

Bass compression is an issue, though. But then again, if we're only going to see one measurement of any given speaker, I'd like to know if that speaker can reach normally loud volume levels without compressing. Maybe a comment about such issues in a review can off-set this concern.
 

ctrl

Active Member
Joined
Jan 24, 2020
Messages
279
Likes
959
Location
DE
...due to sample to sample variation? Or difference in curve smoothing?
I don't think so, because smoothing will not flatten the "comb filtering" (enough) and in Post#319 edechamps posted the measurement of his KH 80 without such an effect (okay, this is not really proof of the contrary) - seems to me some kind of interference by reflections or calibration issue.
 

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
1,120
Likes
1,235
Location
France
I already know what it is capable of. It produced the flattest response of any speaker I have tested with this sample. That is what the manufacturer promotes, and that is what the results are. There is just no way anyone should try to focus so minutely on speaker measurements. The spin data for example sums many graphs as to get rid of such nuances.

If you all want to dig in, you are welcome but please don't involve me. I have spent more time dealing with protests on the measurements than remotely measuring speakers. It is not fun and promotes a strategy that is just wrong with it comes to evaluating speakers based on measurements.

If you want to dig in, how about asking the manufacturer what "interpolated" means on their graph? What samples were used and whether customer units are tested to comply with that spec and if not, they accept returns. Ask them what the frequency resolution is. What lab and technique was used to make those measurements.

Collectively, you all are taking what started as a fun hobby to measure speakers into a miserable back and forth over things that don't matter. You are welcome to keep going but know the costs as you do so....
I'm pretty sure the "complaints" are only numerous because it's the beginning, everyone is very excited and since this is an important opportunity, nobody wants a mistake that can be avoided by reading the manual to be present in all the following measurements. Besides, there's no need to rush, your readers won't get sick (that fast) without their measurement fix.

Well, everything so far is pretty clear, the acoustical axis should be used and some SPL at the listener position (thus handling near/mid/far field matters) should be chosen once and for all. I don't think it's useful or at least urgent to re-measure the Neumanns, we know they're as good as their maker says they are.

If this can make it happier to you, your effort has the potential to change the industry; or at least provide us some ammo to convert some audiofools not too deep in it. But you already know this.
 
Last edited:

LarsS

Addicted to Fun and Learning
Forum Donor
Joined
Mar 23, 2016
Messages
959
Likes
851
Location
Sweden, Stockholm
Perhaps it would be a better service to the forum to move all this endless stuff in a new thread for the people interested. I have to agree with @amirm and am massively uninterested in splitting hairs or obsessing on 1dB, that doesn't add anything to the review.
I strongly support this. @Thomas savage , if you’re not in a nice nearby PUB, something you could intervene in?

I’d be sad to see this go any further.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,808
Location
Zg, Cro
Maybe a comment about such issues in a review can off-set this concern.
Maybe a repeated measurement done at reference axis recommended by manufacturer and avoiding LF compression would do a better job off-setting this concern.

12-15 measurements were performed during setting up NFS and that was certainly a great effort, but doing one more with KH80 may be a good idea to blow away any doubts and to establish a good measurement procedure.
 
Joined
Sep 18, 2019
Messages
48
Likes
23
Lots of discussions about the Klippel system, but I can't find any reference on the measurement mic Armim is using.
 

Absolute

Senior Member
Forum Donor
Joined
Feb 5, 2017
Messages
382
Likes
575
Maybe a repeated measurement done at reference axis recommended by manufacturer and avoiding LF compression would do a better job off-setting this concern.

12-15 measurements were performed during setting up NFS and that was certainly a great effort, but doing one more with KH80 may be a good idea to blow away any doubts and to establish a good measurement procedure.
Well, if Amir knows the many hours of repeating the test will produce the same result, why bother? If there's compression going on, it should have been indicated in the speaker. If not, the indicator is useless, which is also beneficial for me to know. In any case, as long as we make sure to include estimated preference rating both with and without bass extension, it's all a bit nit-pickety.

Of course, personally I'd like to check and double check each and every speaker just to make sure, but if all the testing and calibrating with the system has already been done, why keep repeating and wasting the time? I have no doubt about the Klippel's accuracy and if Amir says he has tested and re-tested to double check the results, we're just gonna have to trust him until someone else with the same system shows different results.

Like I wrote earlier, I would like to keep the on-axis reference point between tweeter and midrange because in my head it makes sense acoustically, but I have no basis to estimate better or different results on average. This Neumann's listening window is essentially picture perfect flat save from the crossover dip, so not sure why people are baffled about the result.
 

NTK

Senior Member
Forum Donor
Joined
Aug 11, 2019
Messages
360
Likes
715
Location
US East
Maybe a repeated measurement done at reference axis recommended by manufacturer and avoiding LF compression would do a better job off-setting this concern. ...
I don't know what Klippel's software will allow, but the mathematics of near field acoustical holography (NAH, which the NFS is based on) allows for the reconstruction of the acoustic field anywhere in 3D. The NFS scanner measures in 2 concentric cylinders, process the data to compute a set of fitting function (spherical wave expansion functions) coefficients, and calculates the SPL's at the CTA-2034 spinorama coordinates based on the fitting functions and coefficients.

To shift the reference axis, all that needs to be done is to recalculate the SPL's to the new set of coordinates. No new measurements should be required.
 

ctrl

Active Member
Joined
Jan 24, 2020
Messages
279
Likes
959
Location
DE
On-axis is not interesting, listening window averaging is.
I have to agree with @amirm and am massively uninterested in splitting hairs or obsessing on 1dB, that doesn't add anything to the review.
It's not about the KH 80 or splitting hair, it's about using a top notch measuring gear to its full extend.
It's like using a super-computer for sum up someone's shopping.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,808
Location
Zg, Cro
I don't know what Klippel's software will allow, but the mathematics of near field acoustical holography (NAH, which the NFS is based on) allows for the reconstruction of the acoustic field anywhere in 3D. The NFS scanner measures in 2 concentric cylinders, process the data to compute a set of fitting function (spherical wave expansion functions) coefficients, and calculates the SPL's at the CTA-2034 spinorama coordinates based on the fitting functions and coefficients.

To shift the reference axis, all that needs to be done is to recalculate the SPL's to the new set of coordinates. No new measurements should be required.
Good point. Not so with LF compression though..
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,808
Location
Zg, Cro
That is correct.

I see little reason to deviate from CTA-2034-A, which clearly specifies the output levels for the spinorama tests. That, IMO, should be the easiest defensible position.
I fully agree. Calibrated mic would also come handy to ensure those levels are being followed.
 

carlob

Senior Member
Forum Donor
Joined
Aug 4, 2019
Messages
452
Likes
685
Location
Roma, Italy
It's not about the KH 80 or splitting hair, it's about using a top notch measuring gear to its full extend.
It's like using a super-computer for sum up someone's shopping.
Feel free to discuss the methodology for another 20 pages, or 200 or 2000. I am just suggesting to do it elsewhere and not in this thread. I for one am interested in the review but unfortunately couldn't care less about obsessing on the methodology details. Is the parquet in the room red oak, what's the relative humidity of @amirm ears during measurements? Did he drink red or white wine before measuring? Have you seen that 0.0001 dB dip at 47.23 hertz?

You can discuss it at will (in another thread please) and when you guys decide that the methodology is good enough let us know :)
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
2,808
Location
Zg, Cro
Feel free to discuss the methodology for another 20 pages, or 200 or 2000. I am just suggesting to do it elsewhere and not in this thread. I for one am interested in the review but unfortunately couldn't care less about obsessing on the methodology details. Is the parquet in the room red oak, what's the relative humidity of @amirm ears during measurements? Did he drink red or white wine before measuring?

You can discuss it at will (in another thread please) and when you guys decide that the methodology is good enough let us know :)
Review is in the post #1. I didn't noticed anybody forcing to you read anything beyond that post.
 

carlob

Senior Member
Forum Donor
Joined
Aug 4, 2019
Messages
452
Likes
685
Location
Roma, Italy
Review is in the post #1. I didn't noticed anybody forcing to you read anything beyond that post.
yes and most of this thread is off-topic if you have ever participate in an internet forum. Off-topic, the magical worlds of the forums. You are discussing the methodology not the review. Open a thread and name it "discussion on @amirm measurement methodology" so I can skip it.
 
Top Bottom