Erins Review of the Yamaha NS-10

[goat76]

What does 'decay time' actually mean to a human subject, do you have data on preference as to establish a discrimination between 'long' and 'short'?

I'm not pretending to be an expert on loudspeakers, but the measurements shown in the paper I posted indicate that this speaker may have some strong points, that may or may not be part of the reason why people found them useful as a mixing tool.

I would expect that a loudspeaker with a very short decay time, such as the NS-10, should keep up a little bit better with the signal with less "overhang", "slugginess" or whatever you could call it, which should result in less masking of finer details in the mix, which should in turn be helpful when determining the actual decay time of the reverb in the programme material. At least that's what some people mention, that the NS-10s are highly revealing of both the amount and the time aspect of the reverb, making it instantly obvious if it's over the top, not enough, too long, or too short. If you have mixed music, you probably know how much this affects the overall feel of the music.

 

[Heinrich]

I got that you mentioned those "dual tone" results but answered something very different, namely that multitone distortion can also consist/result from doppler.

So you still think that Doppler distortion shows a different spectrum than amplitude modulation, at different frequencies?

Exactly, also to prove it a comparative measurement with closed port and equalisation to the identical response would be needed.

I don't think so. The argument was, that the port supports the driver in a narrow band only. It leaves a lot of excursion above the tuning frequency. That leaves a lot of excursion for modulating higher frequencies. Those intermodulation products are more severe because they have greater distance to the modulated tone, hence are masked less.

Scientific scrutiny would need comparative measurements. not assumptions, just install a port which you can close, EQ to the same response and we will see the truth.

I asked you for some data that would show by how much a port would benefit the subjective perception (preference), especially in regard to "important" intermodulation.

Wasn't it you to claim such a benefit before?

In the region of the port the distortion is significantly reduced compared to closed box which also reduces the important multitone distortion.

 

[Heinrich]

I'm not pretending to be an expert on loudspeakers, but the measurements shown in the paper I posted indicate that this speaker may have some strong points, that may or may not be part of the reason why people found them useful as a mixing tool.

Maybe language barrier, on my side of course, but I don't get the connections in your conclusive logical inference? Mixing tool, really?

I'm sure we can establish that the NS-10 has been one of the most commonly used mixing monitors of all time, is that even debatable?

Edit for fairness: https://dt7v1i9vyp3mf.cloudfront.net/assetlibrary/n/ns10m.pdf?jQWj8tYIeZeymRCNXitG9Qfwq9mLf1t0

This "paper" is full of measurements, but also stuffed with 'audiophile' misinterpretations. The 'timing' thing is not due to group delay and such. That's not plausible by any means, as even elder and more comtemporary investigations show. Group delay is not audible to a wide extent.

More plausible, test it yourself, is the sheer lack of (mid)bass doing all the magic. Because it is taken out, the mids get into focus, emphasized by the mid's bump, keeping it unmasked. The timing of the music lives in the 30ms range, see also the precedence (Haas) effect etc pp. The quest is for consistency of rhythm and rhyme. If the pick of the bass guitar matches the hit the drum set not in absolute terms, but over the series of repetitions. The bassy content only distracts. Again, you may want to test it yourself using an equalizer.

So, I doubt that the speaker was actually used as widely as it was present in the studios. If, then it might have been used for synchronizing instruments, which I would not subsumize under 'mixing'. Sorry, please take my apologies.

 

[goat76]

Maybe language barrier, on my side of course, but I don't get the connections in your conclusive logical inference? Mixing tool, really?

Yes, they are famously used as monitors for mixing, but not necessarily as the main monitors which makes them a tool among other studio tools. That doesn't exclude them from being used as tracking monitors as well, and most studios have more than just one pair of speakers.

Over the time course of the NS-10 history, it hasn't always been as common as nowadays that the mix was sent off to a separate mixing engineer. The mix where often done right there in the control room, so that makes it possible that the NS-10 was also used as tracking monitors.

 

[thewas]

So you still think that Doppler distortion shows a different spectrum than amplitude modulation, at different frequencies?

It shows multiple spectral lines, so for example not only at 900+- 30 Hz but also multiples of those:

Figure 5: Additional spectral lines, a.k.a. Bessel lines, generated by two frequencies, the first as before 30Hz with 7mm
membrane travel amplitude and the second one with 900Hz and the same SPL as the first one. Further weaker Bessel spectra appear at twice the second frequency and so on.

From: https://audioxpress.com/article/eliminating-speaker-doppler-distortion

I don't think so.

Your thought cannot deny the final measurement proof.

The argument was, that the port supports the driver in a narrow band only. It leaves a lot of excursion above the tuning frequency.

The excursion is 4 times(!) smaller with every doubling of the frequency.

Wasn't it you to claim such a benefit before?

The claim of mine you posted was about reducing the distortion, not about the audibility which as you know is very difficult to prove and literature in this case is still very limited. As a first step we should compare some real measurements till we move to that one.

 

[ZolaIII]

@Heinrich it's time/space and it can not be any other way around. It's RT60 to RT20 (linear) or RT30 (non linear). Put into space perspective ISO 3382-1 large performance spaces, ISO 3382-2 ordinary rooms or ISO 3382-3 open offices. Long time ago Manfred Robert Schroeder (Bell Labs) established method for measuring integrated energy instead of power of the sound.
It impacts everything! From speach intelligibility to note main tone and harmonic masking. You can do little about it in refractive spaces (which most rooms are) and reason why accustic treatment exists but again some things are close to impossible to get from it mainly bass area, good ratio of back to front refractions (refer to ISO's). So as it's problematic to cope with it it's good at least to take care that system doesn't add to it more than it have to (make less problems that you can not solve). Reason for dead full treated studios is removing space element (as much as possible), dereverberation so that you have it dry and simulate different spaces (quasy anesthetic chamber is another example of it and so is active noise cancelling, closed back headphones and so on). Reverb is oldest effect there is but (most) people here never done production disregard it.

How Humans Conquered Echo

A brief history of artificial reverb in music

www.theatlantic.com

 

[Heinrich]

It shows multiple spectral lines, so for example not only at 900+- 30 Hz but also multiples of those:

@thewas, you said: "Multitone distortion is not only IMD but also doppler distortion so it doesn't have to be just multiples, also with the non-linearities in the displacement thing get even more complicated."

I concluded from that, Doppler distortion was something else. Now you say the contrary? May be it's me. Case closed.

Your thought cannot deny the final measurement proof. ...

You stated a fact in a debate, hence I asked you for some, at least, plausibility, but you put it on me to prove you wrong. I was under the impression it was the other way round. Measurement needs a plan. See below also..

The claim of mine you posted was about reducing the distortion, not about the audibility which as you know is very difficult to prove and literature in this case is still very limited. As a first step we should compare some real measurements till we move to that one.

You said it would be "important". In my book that says there is priority to it, because of audibility? What would a measurement look like? Did you do it, was your hypothesis falsified?

Of course bass reflex aka ported provides some advantages. At least an increase in efficiency by 6dB on the tuning frequency. That says only a quarter of amp power is needed, which is not too shabby. But in reality the port's support is limited to a quite narrow frequency band, which cools down my enthusiasm just a smidge. Maybe I'm overthinking. Today's drivers are that capable at very low prices, though, I recommend to replace the port by a second driver in the same enclosure. Frequency response is corrected for room integration anyway.

 

[thewas]

@thewas, you said: "Multitone distortion is not only IMD but also doppler distortion so it doesn't have to be just multiples, also with the non-linearities in the displacement thing get even more complicated."

The frequency band of a port is not a singularity so it ends in practice not having just single lines.

You stated a fact in a debate, hence I asked you for some, at least, plausibility, but you put it on me to prove you wrong. I was under the impression it was the other way round. Measurement needs a plan. See below also..

Let's get it clear, I hope you agree that the multitone distortion will be reduced when porting a loudspeaker and having it equalised to the same response, so what remains is how much that reduction will be which finally only measurements can exactly reveal.

Of course bass reflex aka ported provides some advantages. At least an increase in efficiency by 6dB on the tuning frequency. That says only a quarter of amp power is needed, which is not too shabby.

Not only that, but this reduction in power and thus also excursion leads also to a reduction of the distortion.

But in reality the port's support is limited to a quite narrow frequency band, which cools down my enthusiasm just a smidge.

That cannot be generalised as it depends on the Q factor tuning of the port which in the end is a resonator.

Today's drivers are that capable at very low prices, though, I recommend to replace the port by a second driver in the same enclosure.

That is an apples vs pears comparison as a second driver is not something without significant cost like an opening/port, needs different drivers (TSP parameters to work similarly well in the same enclosure volume) and has different amplification needs. And also in this case still a reduction in the power, displacement and distortion could be achieved with a BR, that's why practically almost all significant loudspeaker engineers and companies use it.

 

[Heinrich]

That is an apples vs pears comparison as a second driver is not something without significant cost like an opening/port, needs different drivers (TSP parameters to work similarly well in the same enclosure volume) ...

Sure, if one is following alignment recommendations by the book. Who does today, though? And as D&D or Kii show, it doesn't need boutique drivers likewise. In case one cannot take my humble comment as a little bit of friendly nudging into questioning the ancient wisdom, that's o/k with me.

 

[thewas]

Sure, if one is following alignment recommendations by the book. Who does today, though? And as D&D or Kii show, it doesn't need boutique drivers likewise. In case one cannot take my humble comment as a little bit of friendly nudging into questioning the ancient wisdom, that's o/k with me.

Where did I mention boutique drivers? I only said different TSP parameters and by advising me to just correct response per EQ you are carrying coals to Newcastle, but this was not the topic here. It should be also kept in mind though that the more the response is flattened by EQ instead of by TSP tuning the more power reserves are reduced and distortion increased.

 

[Heinrich]

... the more power reserves are reduced and ...

That's what reserves are made for, use them?

 

[thewas]

That's what reserves are made for, use them?

Not if you want to get the max SPL or minimum distortion. By the way the same could be said by a port which is even for free.

In the end of course everything is a compromise and the only thing that matters is that everyone finds the least worse one for the set of his individual needs and priorities, I hope we agree there.

 

[Heinrich]

Not if you want to get the max SPL or minimum distortion. By the way the same could be said by a port which is even for free.

A reserve cannot be spent to get to a max or a min? O/k , a port is free as long one doesn't invest even a thought into it. I leave it at that.