NHT Pro M-00 Powered Monitor Review

[Krunok]

Doesn't look like any anechoic GD graph I've ever seen, which always look like f(x)=1/x.

Yep, they do. FR also looks nice when measured in anechoic chamber, phase graph as well. Just remember you're not listening in anechoic chamber and that your room will generate so many reflections, it will have modes, SBIR effectw will occur - and the resulting GD is what I just showed you.

I'm just wondering if some measurements could help make those reports more exhaustive, nothing else.

You were just voicing your opinion, right? Well, problem has become with this forum that ratio of usefull/total posts is rapidly rushing to 0 because folks like you are voicing opinions although they have no or very little idea what they are talking about.

Did it occur to you that it is ok NOT to have an opinion about things that are out of your competence?
How about some reading and learning before voicing your opinion?

P.S. Nothing personal mate, I'm just voicing an opinion.

 

[amirm]

You were just voicing your opinion, right? Well, problem has become with this forum that ratio of usefull/total posts is rapidly rushing to 0 because folks like you are voicing opinions although they have no or very little idea what they are talking about.

Please dial it down. Nothing of the sort has happened. Without people challenging what we think we know, we don't advance the discussion.

 

[Krunok]

Please dial it down. Nothing of the sort has happened. Without people challenging what we think we know, we don't advance the discussion.

The way I see it is that instead of reading and learning from previous posts different folks are challenging the same things over and over again (sighted tests are not valid, double blind test with matched level is the only way to go, FR is what matters with speakers, THD and time domain stuff not really, and so on ..). But if you prefer to moderate me instead of folks voicing an "opinion" no wonder the usefull post ratio is rushing to zero and that folks which used to make competent opinions on loudspeakers are no longer active on this forum.

 

[amirm]

The way I see it is that instead of reading and learning from previous posts different folks are challenging the same things over and over again (sighted tests are not valid, double blind test with matched level is the only way to go, FR is what matters with speakers, THD and time domain stuff not really, and so on ..). But if you prefer to moderate me instead of folks voicing an "opinion" no wonder the usefull post ratio is rushing to zero and that folks which used to make competent opinions on loudspeakers are no longer active on this forum.

Again, nothing like that has happened. As I type this, I am reading the AES paper he pointed to. Did you bother to download and do the same?

 

[Juhazi]

Well Genelec uses that paper to marketing of their 2-way speaker S360. Unlike 3-way models it naturally doesn't have the GD step caused by W-M crossover.

Anechoic measured GD tells how steep and where crossover points are, basically. LR4 crossover makes a full cycle phase shift, LR2 half of it. When W-M xo is at 300Hz GD are 3ms vs. 1,5ms respectively (with ideal pysical time-alignment). The audibility of this kind of small GD is controversial, but also frequency-dependant, and it applies to on-axis sound only. Off-axis respones and total radiated sound power/room response are always plaqued with mix of varying GD differencies and reflected sound making it's own whoopies.

https://www.audiosciencereview.com/...ility-of-group-delay-at-low-frequencies.8571/

 

[thewas]

As written above group delay is in non-pathological loudspeaker constructions usually above known audability limits from Blauert & Laws https://en.wikipedia.org/wiki/Group_delay_and_phase_delay#Group_delay_in_audio in most frequency range and normally only a problem in low frequency crossed subwoofers, Neumann Monitors have written a nice article about it http://www.neumann-kh-line.com/neum...ing_knowledge_faqs_general-answers_question13

 

[briskly]

There is a definite gain-phase relation for minimum-phase systems, but the calculation is messy between the familiar (π/2 * d/df) asymptotes.

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 phase shift associated with vented speakers and that of the crossovers represent very different problems. Convention paper 10008 focuses on the latter. The testing is done with headphones which normally are more sensitive than speaker listening to phase shifts.

What does Klippel recommend here?

A microphone. Laser scanning is not about to tackle any of the various contributions of fluid non-linearity.

 

[vavan]

here is how uncorrected GD of Revel Performa3 F208 looks in a room

GD graph of my f208s doesn't look that nice for some reason, mostly around 20 ms

 

[lashto]

@amirm first of all, congratulations for your new toy and Thank You!

As Amir noted in this review. You don't need all that many graphs to separate the wheat from the chaff. This one chart will do that once you know what the various lines are showing you and how they fit together. You'll want an in room response too.
View attachment 45671
View attachment 45672

Thanks for posting the JBL-305 and NHT-M00 graphs near each other, something looks a bit strange there: the measurements between approx 6-11kHz look eerily similar and the dip in the 8-9kHz range is pretty much identical (see pic below). The probability that such different monitors have almost exactly the same response is quite low IMO.
@amirm could it be that the measurement setup has an issue in that area ?!

 

[3ll3d00d]

Here is the text export of one of the directivity plots if someone wants to take a shot at plotting it differently.

I notice the file contains 0-180 only, does the NFS assume the response is symmetrical and mirrors it or is this a partial export?

will the equivalent data for vertical be in a separate file?

 

[amirm]

I notice the file contains 0-180 only, does the NFS assume the response is symmetrical and mirrors it or is this a partial export?

You can tell it the speaker is symmetrical but I don't. It generates a report on whether the speaker is or is not and I always find that they are not.

In this case, I was told to not show negative angles so I took them out. Normally they would be there and I can repost if there is interest.

The export works on every graph and what is in it. So they would be separate file at this rate unless I create something custom.

 

[amirm]

@amirm could it be that the measurement setup has an issue in that area ?!

It could be an interference from the measurement rig causing it. In general, it is best to not zoom in so much in the measurements. That is why I draw general lines to indicate trends. Our hearing doesn't have that kind of resolution that high up in frequency.

 

[3ll3d00d]

I don't see any benefit to not including all the data. I mean it's ok to mirror if that's all you have (I tend to measure this way myself because it takes ages and I'm not a robot )

 

[amirm]

I don't see any benefit to not including all the data. I mean it's ok to mirror if that's all you have (I tend to measure this way myself because it takes ages and I'm not a robot )

Well, you all need to agree with each other was to which data you want. People thought the graphs looked too crowded. I am working on a 3-D representation to solve that.

 

[Krunok]

@amirm first of all, congratulations for your new toy and Thank You!



Thanks for posting the JBL-305 and NHT-M00 graphs near each other, something looks a bit strange there: the measurements between approx 6-11kHz look eerily similar and the dip in the 8-9kHz range is pretty much identical (see pic below). The probability that such different monitors have almost exactly the same response is quite low IMO.
@amirm could it be that the measurement setup has an issue in that area ?!

View attachment 46065

And how is your conspiracy theory explaining the fact that Kali measurement is not showing the same similarities?

 

[3ll3d00d]

Well, you all need to agree with each other was to which data you want. People thought the graphs looked too crowded. I am working on a 3-D representation to solve that.

fair enough, I've hacked in support for that format to https://github.com/3ll3d00d/pypolarmap

this gives a couple of static views




(the dark line is the -6dB point)

and one interactive one. You can see the crosshairs on the bottom chart, the other charts update as you move those around.



It lets you choose an angle on which normalise too.

Just a comment: the colour map that you have used (Jet) is not perceptually uniform, and I suggest that you use one that is if available

see above for an example

 

[lashto]

And how is your conspiracy theory explaining the fact that Kali measurement is not showing the same similarities?

With just two examples, I'd say my earlier post was barely a hypothesis ... and not even close to any sort of "theory".

And since you mentioned the new Kali measurements (which I just seen now): the same dip is present in the 8-9kHz range, it's just much deeper. And the whole 7-11kHz area looks quite similar again: deeper dips but the graph still has ~the same shape. Having 3 examples now (and considering that it's 3-out-of-3) this seems to be a pretty good hypothesis already ... one can almost call it a theory ... or "conspiracy theory" if you so wish.

It could be an interference from the measurement rig causing it. In general, it is best to not zoom in so much in the measurements. That is why I draw general lines to indicate trends. Our hearing doesn't have that kind of resolution that high up in frequency.

Generally, I agree with you, even if there is a problem it does not seem like a big/urgent one. OTOH, even the smallest issue might be worth checking & fixing. May be the rig, may be the room, may be the Klippel software, ...
Personally, I do have a serious highs sensitivity in the ~5-10kHz area, e.g. headphones like Sennheiser HD800 and Beyer DT880 put 'needles' into my ears. That is not representative for the general population (not that rare though) but makes a huge difference in one's listening habits & pleasure; it's why I zoomed so much into that range.
Additionally, anyone with tinnitus (not so rare either) might have a big interest in exactly that range too.

Anyway, many thanks for the response and hopefully you can find some time for a closer look at that possible interference.

 

[Krunok]

With just two examples, I'd say my earlier post was barely a hypothesis ... and not even close to any sort of "theory".

And since you mentioned the new Kali measurements (which I just seen now): the same dip is present in the 8-9kHz range, it's just much deeper. And the whole 7-11kHz area looks quite similar again: deeper dips but the graph still has ~the same shape. Having 3 examples now (and considering that it's 3-out-of-3) this seems to be a pretty good hypothesis already ... one can almost call it a theory ... or "conspiracy theory" if you so wish.

And that theory is indicating an error in Klippel's measurements? Wouldn't you say 3 is by far too small number of samples to form such theory?

Btw, I don't really share your opinion that Kali measurement share the same dip as JBL and NHT.

Luckilly for Amir Revel C52 measurement certainly doesn't have the same dip.

 

[lashto]

And that theory is indicating an error in Klippel's measurements? Wouldn't you say 3 is by far too small number of samples to form such theory?
Btw, I don't really share your opinion that Kali measurement share the same dip as JBL and NHT.
Luckilly for Amir Revel C52 measurement certainly doesn't have the same dip.

My bet would be on room and/or rig interferences that are not fully addressed by the Klippel kit. So yes, it could be a Klippel software problem. That isn't much of a wild hypothesis, software errors aren't exactly unheard of.
3 samples is indeed very small but 3-out-of-3 is not. And I do still share my own opinion . Yes the new RevelC52 graph is quite different at first sight. But still ~the same graph shape between ~7-11kHz, just much smoother.

Actually, had another look at the 4 graphs and I'd go even further: the whole treble range from ~3kHz to ~15kHz looks too similar .. at least for my taste.
I'd call the first two (JBL and MH) almost identical, the Revel is similar but much smoother and the Kali is still similar but deeper jagged.
See attached pic ... with some scaling this time so the graphs are (almost) aligned. Also marked the peaks (blue arrows) and the dips (green), just follow the arrows vertically. To me it looks that all four graphs have the same 6 peaks and 5 dips in the ~3-15kHz range. And that is a lot more similarity that I seen anywhere else (between anechoic measurements of different speakers).

P.S.
not looking to start an argument or some sort of conspiracy here, neither to annoy anyone, It's just a hypothesis. Maybe it's all in my head ... maybe I'm the only one who cares ... maybe there is a real problem ... maybe ...

 

[Krunok]

My bet would be on room and/or rig interferences that are not fully addressed by the Klippel kit. So yes, it could be a Klippel software problem.

Did you maybe read the part of C53 measurement where the Klippel's graph has been overlaid with a graph made in an anechoic chamber and they are still pretty much identical? So much for your room/rig interference/Klippel software problem.

Actually, had another look at the 4 graphs and I'd go even further: the whole treble range from ~3kHz to ~15kHz looks too similar .. at least for my taste.
I'd call the first two (JBL and MH) almost identical, the Revel is similar but much smoother and the Kali is still similar but deeper jagged.
See attached pic ... with some scaling this time so the graphs are (almost) aligned. Also marked the peaks (blue arrows) and the dips (green), just follow the arrows vertically. To me it looks that all four graphs have the same 6 peaks and 5 dips in the ~3-15kHz range. And that is a lot more similarity that I seen anywhere else (between anechoic measurements of different speakers).

Did it occur to you that all these 4 speakers have pretty linear on-axis/listening window response so no wonder why they look alike?

P.S.
not looking to start an argument or some sort of conspiracy here, neither to annoy anyone, It's just a hypothesis. Maybe it's all in my head ... maybe I'm the only one who cares ... maybe there is a real problem ... maybe ...

Yep, I'd say you're the only one who cares.