Search for a compact subwoofer, SVS 1000 Pro or Kali WS 6.2?

[pehu63]

What, sinad depends a lot on the components. In fact Amir never showed sinad for speakers, because it is irrelevant

Yes, SINAD includes noise on top of harmonics. If harmonic distortion is irrelevant for speakers, why is it relevant for amps?

Please note that (in my mind), intermodulation distortion is much more critical. Much more critical than noise or THD.

 

[pehu63]

With a sub cutting at 60 Hz, you can obtain a better performances of the woofer in the speakers. They will sound cleaner, most of distortion come from the last 10 Hz of the lows.

Possibly. An 8 inch speaker is also SPL limited at 60hz.

 

[delta76]

Yes, SINAD includes noise on top of harmonics. If harmonic distortion is irrelevant for speakers, why is it relevant for amps?

Please note that (in my mind), intermodulation distortion is much more critical. Much more critical than noise or THD.

because speaker measurements are much more complicated than just a single SINAD. And as I said, Amir has a threshold for each component, for amp 85dB, DAC, 96dB, IIRC the higher the best yes but it's never the same threshold for different components.
You assumption of 70dB SINAD is bad is flawed.

 

[pehu63]

because speaker measurements are much more complicated than just a single SINAD. And as I said, Amir has a threshold for each component, for amp 85dB, DAC, 96dB, IIRC the higher the best yes but it's never the same threshold for different components.
You assumption of 70dB SINAD is bad is flawed.

Why? Because somebody thinks so? My logic is based on the fact that harmonic distortion is just harmonic distortion. From speaker, amp or dac, what does it matter. Just add it up.

 

[delta76]

Why? Because somebody thinks so? My logic is based on the fact that harmonic distortion is just harmonic distortion. From speaker, amp or dac, what does it matter. Just add it up.

There were researches on threshold of SINAD human ears can detect, and those were the threshold from my memory. If you just kind of assume SINAD must be the same for any component there is no point in continuing arguing.

 

[pehu63]

There were researches on threshold of SINAD human ears can detect, and those were the threshold from my memory. If you just kind of assume SINAD must be the same for any component there is no point in continuing arguing.

Would be interested to see the evidence, and understand what makes the THD behave different across the components in the chain.

Again, intermodulation is a different beast, with a much lower treshold, as it introduces frequencies not present in the music.

 

[delta76]

Would be interested to see the evidence, and understand what makes the THD behave different across the components in the chain.

Again, intermodulation is a different beast, with a much lower treshold, as it introduces frequencies not present in the music.

Audibility thresholds of amp and DAC measurements

Introduction A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC...

www.audiosciencereview.com

This is a start.
Nobody here is or was saying 8361 is a bad speaker. For speaker it is FR and directivity that are important and it excels in both.

 

[pehu63]

Audibility thresholds of amp and DAC measurements

Introduction A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC...

www.audiosciencereview.com

This is a start.
Nobody here is or was saying 8361 is a bad speaker. For speaker it is FR and directivity that are important and it excels in both.

So, the harmonic distortion in the speaker is not relevant? Remeber, the 83x1 has (Class-d) amps.
Once the amp is inside the speaker box, distortion is no longer relevant? Funny.

Tons of measurements info there, lots of opinions, cannot find any double-blind testing, like influence on preference rating.
Some quotes: "His guideline for distortion is <0.05% equivalent to -66 dBFS". For harmonics, I would agree.
Yet, totally funny that the threshold for intermodulation distortion is less strict. Intermodulation is more disturbing than harmonics.
I have no evidence, so checked with mr. Claude 4.5 on this:

• Audibility: IMD is generally more objectionable than equivalent levels of harmonic distortion. Even low levels (0.01-0.1%) can be audible as a loss of clarity or "cleanness" in complex passages."

On harmonics, in line with my current assumptions, finally I understand why even harmonics are "not that bad", octaves:

• Even-order harmonics (2nd, 4th, 6th) are musically consonant - they're octaves and harmonically related intervals. Many people find low levels of 2nd harmonic actually pleasing, giving a "warm" or "full" quality
• Odd-order harmonics (3rd, 5th, 7th) can sound harsh or dissonant at higher levels, though 3rd harmonic can add "richness" in moderation
• Audibility: THD below 0.1% is generally considered inaudible in most listening conditions, though this depends heavily on the harmonic structure and program material.

 

[delta76]

So, the harmonic distortion in the speaker is not relevant? Remeber, the 83x1 has (Class-d) amps.
Once the amp is inside the speaker box, distortion is no longer relevant? Funny.

Tons of measurements info there, lots of opinions, cannot find any double-blind testing, like influence on preference rating.
Some quotes: "His guideline for distortion is <0.05% equivalent to -66 dBFS". For harmonics, I would agree.
Yet, totally funny that the threshold for intermodulation distortion is less strict. Intermodulation is more disturbing than harmonics.
I have no evidence, so checked with mr. Claude 4.5 on this:

• Audibility: IMD is generally more objectionable than equivalent levels of harmonic distortion. Even low levels (0.01-0.1%) can be audible as a loss of clarity or "cleanness" in complex passages."

On harmonics, in line with my current assumptions, finally I understand why even harmonics are "not that bad", octaves:

• Even-order harmonics (2nd, 4th, 6th) are musically consonant - they're octaves and harmonically related intervals. Many people find low levels of 2nd harmonic actually pleasing, giving a "warm" or "full" quality
• Odd-order harmonics (3rd, 5th, 7th) can sound harsh or dissonant at higher levels, though 3rd harmonic can add "richness" in moderation
• Audibility: THD below 0.1% is generally considered inaudible in most listening conditions, though this depends heavily on the harmonic structure and program material.

Not even sure what you are trying to say. For active speakers you care about their FR (and other things like directivity). Why does it matter how much distort its amp has if they used DSP to calibrate/balance what gets out of the speaker? I don't care if the amp has SINAD of 70dB if the FR is perfect.

 

[Miguelón]

Not even sure what you are trying to say. For active speakers you care about their FR (and other things like directivity). Why does it matter how much distort its amp has if they used DSP to calibrate/balance what gets out of the speaker? I don't care if the amp has SINAD of 70dB if the FR is perfect.

I suppose that, when audibility threshold of amplifiers non-linearities were measured, they used subjects listening the amps through speakers.

So, in this case, the numbers are qualifying an amp supposing a theoretically ideal speaker and room. Or perhaps anechoic chambers…

That’s a good argument to the most recommended first priority in audiophile: choose the speakers, secondary the amp.

I can imagine how bad a good amp will sound in a bad speaker, and I have relatively modest amps (the ones that Genelec included in my active speakers, cannot be state of the art given the price) that sounds wonderful due to the speakers FR and directivity.

I agree with you, although Genelec don’t publish separate SINAD or THD for active monitors, the main component is the speaker.

A just correct amp will be enough, if as I supposed, the tolerances were evaluated in a speaker-amp integration, when performing scientific studies.

 

[dominikz]

Some quotes: "His guideline for distortion is <0.05% equivalent to -66 dBFS". For harmonics, I would agree.
Yet, totally funny that the threshold for intermodulation distortion is less strict. Intermodulation is more disturbing than harmonics.
I have no evidence, so checked with mr. Claude 4.5 on this:

• Audibility: IMD is generally more objectionable than equivalent levels of harmonic distortion. Even low levels (0.01-0.1%) can be audible as a loss of clarity or "cleanness" in complex passages."

Note however that both IMD and THD are just two ways we measure the same underlying non-linearity in the device under test.
THD and IMD both result from the fact that at certain frequencies and certain signal levels the output of a device stops being proportional to its input - resulting in new frequency components being created at its output.

If you measure this non-linearity with a single tone, you will see the resulting distortion as multiples of the input tone frequency (i.e. harmonic distortion), and we express this as THD.

If you measure this non-linearity with multiple tones (typically two), you will see the resulting distortion as sum/difference components around the input signal frequency components, which we express as IMD.

Even the distortion seen in the multitone measurement is again just another measure of this same non-linearity.

So THD and IMD are linked, and you can't really have one without the other.

• Even-order harmonics (2nd, 4th, 6th) are musically consonant - they're octaves and harmonically related intervals. Many people find low levels of 2nd harmonic actually pleasing, giving a "warm" or "full" quality
• Odd-order harmonics (3rd, 5th, 7th) can sound harsh or dissonant at higher levels, though 3rd harmonic can add "richness" in moderation

This isn't entirely correct, however.

First: not all even-order harmonics are octaves to the fundamental.

E.g. if the fundamental is 1kHz then the first few octaves over it are: 2kHz (2nd harmonic), 4kHz (4th harmonic), 8kHz (8th harmonic).
Note that an octave is always a doubling of frequency, so e.g. the 6th and 10th harmonics are not octaves to the fundamental even though they are even-order harmonics (they are a perfect fifth and a major third, respectively - so still considered consonant if we look at music theory).

Next, some of the odd-order harmonics are what we'd consider consonant intervals as well: e.g. a 3rd harmonic is a perfect fifth, and the 5th harmonic is a major third; but the 7th and 9th harmonics would be considered dissonant.

So in that sense it is true that odd-order harmonics become dissonant earlier than the even-order ones.

However, you may notice that in most well-designed devices higher order harmonics are usually low in level, and the ones that are dominant are consonant (mostly up to the 5th harmonic, often just 2nd and 3rd).
But even so, a 3rd harmonic is less likely to be masked by the fundamental (because if is further away from it) so in that sense it might be easier to hear, assuming its level is high enough.

 

[pehu63]

Not even sure what you are trying to say. For active speakers you care about their FR (and other things like directivity). Why does it matter how much distort its amp has if they used DSP to calibrate/balance what gets out of the speaker? I don't care if the amp has SINAD of 70dB if the FR is perfect.

I guess you are confusing things. DSP cannot correct harmonics, or prevent it. Harmonic distortion is just harmonic distortion, whether the speaker does that, or the amp does, or the amp in the speaker does that. Harmonic distortion just adds up. Again, why bother about 0.01% in your amp, when the speaker easily does 0.5%? But odd harmonics are more disturbing.

Somehow difficult to get understanding on intermodulation. I write it out now. Intermodulation is a totally different beast. It _adds_ frequencies to the content, called INTER-modultation. Suppose you have 10khz and 400hz. With intermodulation, you also get 9.6Khz and 10.4Khz. And this is much more disturbing to the human ear than harmonics. So, in my mind, harmonic distortion is not so relevant, but odd harmonics are a bit more harmful. Remember, in music, the harmonics are there already. And that is the difference between harmonic distortion and intermodulation distortion.

In classical class-AB amplifier, limited slew rate is a big contributor to intermodulation distortion. Class-D amplifier stand out for two reasons, (1) easy to design/build with very low intermodulation, and (2) highly insensitive to complicated loads, like the passive cross overs. With the same spec on paper, class-D will sound better...... (oops ;-)

In active speakers, intermodulation is limited by using separate amps.....

 

[pehu63]

Note however that both IMD and THD are just two ways we measure the same underlying non-linearity in the device under test.
THD and IMD both result from the fact that at certain frequencies and certain signal levels the output of a device stops being proportional to its input - resulting in new frequency components being created at its output.

As far as I know, intermodulation is a totally different beast. Much more disturbing. See my previous post.
Non linearity just add harmonics.
(and the slew rate concept, I know, I was electronics engineer 40 years ago....)

 

[dominikz]

As far as I know, intermodulation is a totally different beast. Much more disturbing. See my previous post.
Non linearity just add harmonics.

As I mentioned in the previous post, IMD and THD are just two ways we measure the same underlying non-linearity in the system - if you don't believe me, have a look at the first sentence in chapter 1.1 of this Hypex application note, and the explanations provided there. This should also be covered in undergraduate-level courses on signal processing and control theory.

But in short:

• If you pass a single pure tone through a non-linear transfer function you will measure THD at the output.
• If you pass two pure tones through the same non-linear transfer function you will measure IMD at the output.

So every non-linear system will produce both THD and IMD to some extent. This can even be demonstrated mathematically for some simpler transfer functions (this is also mentioned in the Hypex application note I linked above).

 

[pehu63]

As I mentioned in the previous post, IMD and THD are just two ways we measure the same underlying non-linearity in the system - if you don't believe me, have a look at the first sentence in chapter 1.1 of this Hypex application note, and the explanations provided there. This should also be covered in undergraduate-level courses on signal processing and control theory.

But in short:

• If you pass a single pure tone through a non-linear transfer function you will measure THD at the output.
• If you pass two pure tones through the same non-linear transfer function you will measure IMD at the output.

So every non-linear system will produce both THD and IMD to some extent. This can even be demonstrated mathematically for some simpler transfer functions (this is also mentioned in the Hypex application note I linked above).

Thank you for your persistence, and thank you for tolerating my ignorance. Real interesting read, not sure if I understand all implications.

Article does argue that for class-D amplifiers (at least Hypex) THD meaurement is sufficient. For class-AB, things can be more interesting. I will re-read a few times, will take a week, and get back.

Anyway, arguing against the hypex guy offers little chance of success ;-)

 

[dominikz]

Thank you for your persistence, and thank you for tolerating my ignorance. Real interesting read, not sure if I understand all implications.

No problem at all, glad you found it interesting! After all, we're all here to learn and share.

Article does argue that for class-D amplifiers (at least Hypex) THD meaurement is sufficient. For class-AB, things can be more interesting.

Truth be told, my opinion is that most audio measurements have value. When choosing them the question is only which detail we want to focus on the most.
When talking about non-linear behaviour, if I had to choose only one measurement method, I personally find that multitone measurement gives the most informative overview.
The standard equal-level ("white" spectrum) multitone signal is a great torture test, and the "pink" spectrum multitone is a good way to represent real-world performance. Here's an example of those for a class-D amp (full measurements here).

 

[Miguelón]

No problem at all, glad you found it interesting! After all, we're all here to learn and share.

Truth be told, my opinion is that most audio measurements have value. When choosing them the question is only which detail we want to focus on the most.
When talking about non-linear behaviour, if I had to choose only one measurement method, I personally find that multitone measurement gives the most informative overview.
The standard equal-level ("white" spectrum) multitone signal is a great torture test, and the "pink" spectrum multitone is a good way to represent real-world performance. Here's an example of those for a class-D amp (full measurements here).

This is my post, you two! Get a room...

Thinking about a new sub for a smaller setup (Genelecs 8020 in my other office): I'm thinking about the WiiM Sub because easy to move without cables.

Any issue from WIFI sync in subs? Reading here It seems the major complaint but don't understand why...