• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

NHT Pro M-00 Powered Monitor Review

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,580
Likes
38,281
Location
Gold Coast, Queensland, Australia
To me a buzz in a powered loudspeaker is much worse than a hiss. I would want to be on notice of any hiss or buzz, but a buzz is much more likely to result in a DO NOT BUY based on my own preferences.

I have to agree with this. A buzz in a speaker that only occurs at a low level and at a particular frequency is one of my pet hates. I have several pairs of small and otherwise quite lovely little speakers that have those characteristics. They are perfectly fine 99% of the time, but when the resonance is found in some piece of music, you can't unhear it. Much time and effort is spent tracking down the source only to find sometimes it's simply not worth the effort.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,404
Likes
5,296
Location
Somerville, MA
At this you might as well just consider me a Toole Book Auto-Reply Bot, but to further elaborate on this point (Section 4.6/p78):
"Simple resonances are found in electronic, acoustical or mechanical devices and constitute “lumped elements” in a resonating system: a mass component (inductance), a compliance (capacitance) component and a damping (resistance) component. The mass and compliance determine the frequency of the resonance. The damping determines the amount of energy loss in the system, which defines the quality factor, or Q. This determines the bandwidth of the resonance in the frequency domain and the duration of the buildup and decay of energy in the time domain. High-Q resonances have a small footprint in the frequency domain (a narrow, sharpish spike in a frequency response) and a long buildup and decay in the time domain. As Q falls, the resonance gets wider in the frequency domain and occupies less space in the time domain. Eventually, the frequency response gets to be “flat,” there is no evidence of a resonance, and the time-domain misbehavior disappears. So it is inevitable that all systems designed to minimize resonances have flattish, smooth frequency responses."

It took me many years to appreciate the commonalities between acoustic, mechanical and electrical resonances. This is an amazing passage.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,424
Location
NYC
It took me many years to appreciate the commonalities between acoustic, mechanical and electrical resonances. This is an amazing passage.

Indeed, it was one of the biggest "a-ha!" moments in the book for me. All this talk of debate about CSDs aside, I also have to appreciate how digestible Dr Toole's book is for someone with minimal science background like me.
 

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
At this point you might as well just consider me a Toole Book Auto-Reply Bot, but to further elaborate on this point (Section 4.6/p78):

"Simple resonances are found in electronic, acoustical or mechanical devices and constitute “lumped elements” in a resonating system: a mass component (inductance), a compliance (capacitance) component and a damping (resistance) component. The mass and compliance determine the frequency of the resonance. The damping determines the amount of energy loss in the system, which defines the quality factor, or Q. This determines the bandwidth of the resonance in the frequency domain and the duration of the buildup and decay of energy in the time domain. High-Q resonances have a small footprint in the frequency domain (a narrow, sharpish spike in a frequency response) and a long buildup and decay in the time domain. As Q falls, the resonance gets wider in the frequency domain and occupies less space in the time domain. Eventually, the frequency response gets to be “flat,” there is no evidence of a resonance, and the time-domain misbehavior disappears. So it is inevitable that all systems designed to minimize resonances have flattish, smooth frequency responses."
I see, it makes sense now. So only group delay could matter in this case (yes, you can compute it from the LF roll-off, but that's not very readable). Has anyone read Genelec's paper about it? I don't have the required AES membership, sadly; I can foresee the answer, though, since all their designs are ported.
The other remaining question is: what is the audibility of high Q resonances? I remember the SOS review for the 705p mentioning an audible "pipe organ" effect, even if the resonance is very sharp; which was supported by some provided measurements confirmed by soundandrecording.de.
I might take a guess and say that the Q of the variation changes how it's perceived: as a time or frequency anomaly.

See
https://www.soundandrecording.de/equipment/jbl-705p-nahfeldmonitor-im-test/
JBL-705P-FRE.jpg

JBL-705P-SPC.jpg

and
https://www.soundonsound.com/reviews/jbl-7-series
JBL_7series_04-SQR7E7.PhmgMqsGahebvxuubKvH3aO1m.jpg

JBL_7series_05-ar2I68LxWTxROsf0DCt2W.B8iXsZr00T.jpg
 

CumSum

Active Member
Joined
Dec 10, 2019
Messages
106
Likes
116
Given powered monitors (and subwoofers) utilize ADC's, DSP and DAC's, they all sort of cheat their way into having a flat anechoic frequency response using said DSP.

So what is the best way of telling a good speaker apart from a bad speaker, given both are EQ'd to be flat from the factory?
 

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,463
Location
Land O’ Lakes, FL
So what is the best way of telling a good speaker apart from a bad speaker, given both are EQ'd to be flat from the factory?
If everything else is the same, the difference between a speaker that is naturally flat and one that is EQ’d to be flat would be distortion/ringing (high-Q)/SPL.

EQ is adjusting level, so if you have a 3dB dip that you EQ up, what you basically did is lower everything else down by 3dB, so you now have less max SPL as you are sending more wattage into that dip.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,424
Location
NYC
I see, it makes sense now. So only group delay could matter in this case (yes, you can compute it from the LF roll-off, but that's not very readable). Has anyone read Genelec's paper about it? I don't have the required AES membership, sadly; I can foresee the answer, though, since all their designs are ported.
The other remaining question is: what is the audibility of high Q resonances? I remember the SOS review for the 705p mentioning an audible "pipe organ" effect, even if the resonance is very sharp; which was supported by some provided measurements confirmed by soundandrecording.de.
I might take a guess and say that the Q of the variation changes how it's perceived: as a time or frequency anomaly.

See
https://www.soundandrecording.de/equipment/jbl-705p-nahfeldmonitor-im-test/
JBL-705P-FRE.jpg

JBL-705P-SPC.jpg

and
https://www.soundonsound.com/reviews/jbl-7-series
JBL_7series_04-SQR7E7.PhmgMqsGahebvxuubKvH3aO1m.jpg

JBL_7series_05-ar2I68LxWTxROsf0DCt2W.B8iXsZr00T.jpg

On audibility of Q, Toole cites a 1988 paper:
Snag_f723586.png


Figure 4.11: Deviations from flat caused by resonances of three different Qs that were just detectable when listening to pink noise (all three frequencies) and to close-miked, low-reverberation, pop/ jazz recordings (200 Hz only). Also shown is a ± 3 dB tolerance, which can be seen to be too tolerant of medium- and low-Q resonances and unnecessarily restrictive of some high-Q resonances. Adapted from Toole and Olive (1988).


These were done in anechoic conditions and thresholds may be lower in a real room. So audibility will vary with music content, of course, but it's interesting to note just how much more sensitive we are to low q resonances than high Q ones. We do hear the high Q ones, they're basically just more picky about the frequencies and durations to energize them.
 

Attachments

  • Snag_f6fa3a1.png
    Snag_f6fa3a1.png
    230.5 KB · Views: 124

thefsb

Addicted to Fun and Learning
Joined
Nov 2, 2019
Messages
796
Likes
657
The graphing capabilities in Klippel are very limited. It can export data but then it is a lot of work to plot them externally. If such data was useful, I would put in the work. But they are not really.

I can provide some exports for people to play with however. Would you like that?

I think we should concentrate on the useful.

Otoh, I'm all in favor of a database of results that people can do cool stuff with. For example, side-by-side comparisons.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,395
Location
Seattle Area
So what is the best way of telling a good speaker apart from a bad speaker, given both are EQ'd to be flat from the factory?
EQ cannot fix the off-axis issues. So if there are problems, we will see them there.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
Advanced Speaker Measurements
Folks like pretty and colorful graphs so here we are with contour plots in both horizontal and vertical axis:

View attachment 45649


View attachment 45650

The more uniformity and gradual change, the better in these graphs. Vertical axis performance tends to be worse in many speakers and such is the case here. So absorbing those if your room is not already too dead may be advised.

I took feedback to trim down these measurements some. Let me know if they are easier to read:
View attachment 45653

They tell us what the contour graphs do but without the fancy colors.
I prefer the contour plots. For me it's easier to see how good a speaker's dispersion is. And they are used by others as well.

As an EE I should prefer the contour graphs because it's easier to detect the actual level but with lots of grey lines crossing each other it's almost impossible to get a safe reading.

What I would like to see (yeah I know THD is not audible, but still ...) is something like the following plots (Stereoplay, german mag). They show frequency response and THD for 4 different SPLs (85, 90, 95 and 100 dB):

B&W 804 D3 (floor stander):
bw804d3.png


Dynaudio Excite X18 (small 2 way) shows much higher THD in the lows.
dynaudio-excite-x128.png



Cabasse Stream (streaming box) shows strong compression (active limiter) below 500 Hz:
cabasse-stream.png


Another possibility is to show max SPL at 1m for 1% and 3% THD.

Neumann KH310:

neumann_kh310_max_spl_510_2.gif


Neumann KH420:

KH420_MAX_510.gif


Maybe an even better possibility is the IMD test with multi tone as done by Sound & Recording (german mag).

Neumann KH80:
kh80-mt.png


Neumann KH420:
kh420-mt.png
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,395
Location
Seattle Area
What I would like to see (yeah I know THD is not audible, but still ...) is something like the following plots (Stereoplay, german mag). They show frequency response and THD for 4 different SPLs (85, 90, 95 and 100 dB):
Do we have any information that these distortion measurements are correct? What is the distortion in their microphone for example? I performed a bunch of headphone distortion tests and no matter what headphone I put on, I would see the same signature which pointed to it being a microphone problem, not headphone.

Are these anechoic measurements? If not, how do they know the effect of room modes and long decay of impulses?

How do I prove any of my measurements are correct and defensible?
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,424
Location
NYC
Do we have any information that these distortion measurements are correct? What is the distortion in their microphone for example? I performed a bunch of headphone distortion tests and no matter what headphone I put on, I would see the same signature which pointed to it being a microphone problem, not headphone.

Are these anechoic measurements? If not, how do they know the effect of room modes and long decay of impulses?

How do I prove any of my measurements are correct and defensible?

Yeah, sound and recording performs their measurements in an anechoic chamber - I believe the same one neumann uses? - so it's not quite comparable.
 

bobbooo

Major Contributor
Joined
Aug 30, 2019
Messages
1,479
Likes
2,079
It is possible to generate the CSD such that it is in reference to Time 0, meaning the it starts out completely flat; that way it is much easier to identify resonances and uneven decay.

Like these graphs of headphone measurements by Reference Audio Analyzer: https://reference-audio-analyzer.pro/en/report/hp/audio-technica-ath-a1000x.php#rw36 ?

They definitely make it easier to visualize resonances and ringing. Maybe something similar for speakers would be a good idea.

Here's what RAA say in their explanation of these graphs (translated from Russian by Google):
You can see that all frequencies do not fall too fast during the first millisecond. There are separate "long-falling" frequencies. In the presence of non-linear distortions in the headphones and / or amplifier, some frequencies will receive an additional emphasis. In the presence of resonances in the high-frequency region, depending on the choice of amplifier, there will be a different degree of noticeability of the “sibilants”.
 
Last edited:

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
@amirm , I want to kindly suggest that you publish a core set of the data as well, not just the graphical output. Even one singe item would do great: impulse response on-axis, in any commonly used format (.WAV with 32-bit floats preferred). Data itself is so much more useful that printouts, notably for a simple rough guess of actual sound (throw the IR in a convolver).
That would be a great gift for the community...
 

JIW

Senior Member
Joined
Nov 11, 2019
Messages
378
Likes
551
Location
Germany
Do we have any information that these distortion measurements are correct? What is the distortion in their microphone for example?

According to an older article, S&R uses a Brüel & Kjær Type 4135 which has the following specs:
Screenshot 2020-01-14 at 11.03.28.png

Further up the measurement chain, they use a B&K Type 2669,
Screenshot 2020-01-14 at 11.06.07.png

a B&K Type 2610
Screenshot 2020-01-14 at 11.08.58.png

and finally a RME Multiface II for AD and DA at 24 bit/96 kHz.
Screenshot 2020-01-14 at 11.11.20.png


I can't quite make sense of all the specs but it seems transparent to me compared to 1% or greater distortion of the speakers.
 

kschmit2

Active Member
Joined
Oct 8, 2018
Messages
159
Likes
208
Do we have any information that these distortion measurements are correct? What is the distortion in their microphone for example? I performed a bunch of headphone distortion tests and no matter what headphone I put on, I would see the same signature which pointed to it being a microphone problem, not headphone.

Are these anechoic measurements? If not, how do they know the effect of room modes and long decay of impulses?

How do I prove any of my measurements are correct and defensible?

Hi Amir,

Stereoplay and Audio in Germany use the same measurement lab run by their publishing house Weka Media.

Some of their methodology is described here in articles they translated to English:

Speaker measurements: https://static1.squarespace.com/sta...o+testet+AUDIO+Lautsprecher+Audio+2017-05.pdf
According to that article their measurement system uses Brüel & Kjaer 4191 microphones (https://www.bksv.com/en/products/transducers/acoustic/microphones/microphone-cartridges/4191)

Amplifier measurements:
https://static1.squarespace.com/sta...c8dd7957a/1489563712942/Verstärker+Audio.pdf

Headphone measurements:
https://static1.squarespace.com/sta...rt+Kopfhörermessungen+stereoplay+2017-04.pdf

Turntable measurements:
https://static1.squarespace.com/sta...41acad/1489563683682/Plattenspieler+Audio.pdf



Some more descriptions for other types of measurements are linked to on their website:
https://www.connect-testlab.com/measurement-capabilites
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,780
Location
Oxfordshire
Thanks for this, didn't know about this model, but I'll still play the mean ingrate and say that CSD is needed to know if you monitor rings like a bell or comes with a free church organ.
I tend to agree though my horns have a waterfall that looks more like gentle rapids. It doesn't sound bad though perhaps because there aren't any long resonant ridges
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I see, it makes sense now. So only group delay could matter in this case (yes, you can compute it from the LF roll-off, but that's not very readable).

Group delay is calculated as 1st derivation of phase response and can't possibly help you to identify resonances.
 
Top Bottom