Skip to main content
There have been questions about the high frequency response of the Sonos Amp. User hodgeal posted a very interesting plot showing the frequency response of the Sonos Amp using the software REW. It didn't show excessive high frequencies in that single test.



I had the opportunity to compare the Sonos amp to my Peachtree amp using the same software (which is very useful and easy to use - thanks hodgeal), and this provides a comparative test of the Sonos Amp frequency response.



The Peachtree Nova 125 SE amp is well-respected in the audio community. It is a class D digital amp, same as the Sonos Amp, but as far as I can tell uses different chips. The test uses Duntech Marquis speakers and a Studio Projects C1 microphone. The mic is setup on-axis at a distance of 120cm, which is pretty standard. But none of the details should matter because this is a comparison under identical conditions - I simply swapped amps and repeated the test.



And the results?







As you can see, the amps are almost identical. Ignore low frequencies - they vary from test to test even with the same amp below about 30Hz. At high frequencies, above about 4kHz, the Peachtree gets gradually brighter, ending up being about 1.5dB louder at 20kHz. This would be imperceptible to pretty much everyone.



Note that there is a slight increase in high frequencies above about 7kHz for both amps, so this could be the speakers. However, I think this is due to the frequency response of the microphone - the specifications show a similar slight increase.



Cheers, Peter.


I think you have discovered the Gibbs Phenomenon. Some people refer to this as "overshoot", and it is related to "ringing". Usually it is observed when low-pass filtering, but it can also happen with a high-pass filter. If the filter has too high a Q factor .


There is also this to consider (filter causing constructive effect).

https://www.soundonsound.com/sound-advice/q-why-do-my-mixes-clip-when-i-apply-high-pass-filter

FWIW I was using a 6db filter in Audaicity for a gentle rolloff on the HP filter like I would think a crossover would use.
Extra power is always useful - to not just run at max, but for better handling music with high dynamic range.

What's the science behind that idea when it comes to Class D amps? It might be true for more traditional amps where the distortion goes up as the volume increases. But I believe Class D amps go the other way (based on limited reading).



And when you say other amp manufactures could do the same, I don't suppose you are referring to the many that use electro mechanical volume controls.

I hadn't thought about whether the volume control method would make a difference or not. Perhaps non-Class D amps would be trickier, but I don't know enough about amps to really be able to offer an opinion.
Extra power is always useful - to not just run at max, but for better handling music with high dynamic range. At max, even a Connect Amp is pretty loud, but I wonder if it could not better use this feature. If Sonos ever gets around to giving it that.

And when you say other amp manufactures could do the same, I don't suppose you are referring to the many that use electro mechanical volume controls.
Does this then bring to light an undisclosed feature of the Amp? That it will deliver sound levels that are higher across the board by 20-30% if bonded with the Sonos Sub or with a third party Sub of similar power? That is no trivial feature to have. And I don't think that too many legacy HiFi amps can boast of this capability; all they can claim is what comes from the Sub - frequency extension at the lower end.

So, does anyone want to take a shot at answering this?


Sure. I think you're right. It's making optimum use of its 125W power. Of course, it's only relevant if running the Amp at maximum volume is not loud enough. And I have to say, it's already pretty loud at max! I suspect that all other amp manufacturers could choose do the same thing with the power they save by hiving off the low frequencies to a sub.
Does this then bring to light an undisclosed feature of the Amp? That it will deliver sound levels that are higher across the board by 20-30% if bonded with the Sonos Sub or with a third party Sub of similar power? That is no trivial feature to have. And I don't think that too many legacy HiFi amps can boast of this capability; all they can claim is what comes from the Sub - frequency extension at the lower end.

So, does anyone want to take a shot at answering this?
Something I find interesting... if you apply a 100Hz high pass filter to some modern popular music files, the peaks actually increase. So files with peaks just shy of 0dB go into clipping. If you attenuate the original file by 3dB, then HP filter it, you can see that the peaks actually increase by quite a bit. All of this convinces me that there is no reason for Sonos to a apply a 3dB bump with their sub crossover.

I think you have discovered the Gibbs Phenomenon. Some people refer to this as "overshoot", and it is related to "ringing". Usually it is observed when low-pass filtering, but it can also happen with a high-pass filter. If the filter has too high a Q factor (i.e. it is a steep filter), there can be overshoot for frequencies close to the filter cutoff. The first plot below shows the Sonitus EQ in Cakewalk by Bandlab (used to be Sonar, similar program to Audacity). It has a Q of 1.6 and simply cuts out low frequencies. The second plot pushes the Q up to 3, which boosts some frequencies while still cutting out most of the low ones.











Peter - sorry if we (especially me) are derailing your thread!

No worries - it's all interesting discussion.
Something I find interesting... if you apply a 100Hz high pass filter to some modern popular music files, the peaks actually increase. So files with peaks just shy of 0dB go into clipping. If you attenuate the original file by 3dB, then HP filter it, you can see that the peaks actually increase by quite a bit. All of this convinces me that there is no reason for Sonos to a apply a 3dB bump with their sub crossover.



Peter - sorry if we (especially me) are derailing your thread!
I was comparing peak amplitude before and after the filtering. The link I provided shows how to do this in Audacity. I was using jazz and classical music samples from 2L. I wouldn't be surprised if other types of music gave different results.
Perhaps a stupid question: how would you see this headroom unless the volume control was recalibrated as Sonos seem to have via software for the Sonos Amp? But I agree that we need to see other responses.
But isn't this before the volume control has been recalibrated as Sonos has apparently done - I suppose your finding above is with some other amp?

What I was doing with audacity was just filtering a music file, nothing really specific to the amp. I was just trying to see if high-pass filtering creates enough headroom to justify bumping gain with a sub crossover. I'd be interested to see what others find.
But isn't this before the volume control has been recalibrated as Sonos has apparently done - I suppose your finding above is with some other amp?
This is interesting, so I imported some music into audacity to see how much headroom is created by high-pass filtering. I'm finding that it isn't much, only a few tenths of a dB typical even with a 100hz filter. So I am not convinced that there is good reason for the 3dB bump that Sonos uses with the crossover. Try it yourself; here is how to find the peak level in audacity:



http://think3x.com/audio/2014/09/02/audacity-effects-amplify-5-finding-peak-amplitude/
Which lead me too: this feature would be even more welcome in the Connect Amp, which is just 55 wpc, and could better use a boost in overall sound levels. I am pretty sure it isn't implemented there, from memory; my Sub is with the play 1 these days. I see no reason why it can't be implemented there, other than the usual prioritisation constraints. And of course, the fact that it is a superseded product.
Does this then bring to light an undisclosed feature of the Amp? That it will deliver sound levels that are higher across the board by 20-30% if bonded with the Sonos Sub or with a third party Sub of similar power? That is no trivial feature to have. And I don't think that too many legacy HiFi amps can boast of this capability; all they can claim is what comes from the Sub - frequency extension at the lower end.
Also, a question of the Sonos Amp for confirmation: If music content minus frequencies below 80hz is played, will it sound noticeably louder every time the Sub is toggled on in the app with no other change in how the music sounds?

I think this is correct. It is demonstrated by the measurements hodgeal and I made.
Why is it wrong to say it is 250W into 8 ohms for a 80-20k measurement?

Given an 8 ohm impedance, I think the Amp can only put out 125W - this would be determined by the design of the amp, in particular its ability to push current through an 8 ohm load. If that power (ability to push current) is spread over 20-20k, you get a certain sound level. If that power is limited to 80-20k, then frequencies above 80 Hz can have a 3dB increased sound level (i.e. more current), but the total power is still 125W. The power saved from not having to reproduce the low frequencies is now able to be utilised for frequencies above 80 Hz. But the total available power is fixed at 125W by design.



Cheers, Peter.
Also, a question of the Sonos Amp for confirmation: If music content minus frequencies below 80hz is played, will it sound noticeably louder every time the Sub is toggled on in the app with no other change in how the music sounds?
There is a better set of questions to ask:

Suppose the same Sub was to be attached to the Sub out of a traditional HiFi stereo amp of the same power spec of the Sonos amp, 125 wpc into 8 ohms across 20-20k.

1. Would that combo now be 250 wpc also?

2. The 3dB boost seen here, isn't typical for such amps, as far as I know. So how does this ability to drive the 80-20k to higher sound levels than before achieved? And if it never gets achieved, with unused headroom even with amp volume at 100%, does it mean that the combo is NOT 250 wpc, but 125 wpc as it was minus the Sub?
I think it is more accurate to say that it is 125W into 8 ohms for a reduced range of frequencies compared to the normal 20-20k measurement.

p.s. The Amp plus powered sub combo might now be 250W.


Why is it wrong to say it is 250W into 8 ohms for a 80-20k measurement? Isn't that what the p.s. also conveys, if the Sub delivers 250 wpc from 20 to 80 hz, that the combination is 250 wpc from 20-20k?
I think it is more accurate to say that it is 125W into 8 ohms for a reduced range of frequencies compared to the normal 20-20k measurement. Frequencies below 80Hz (or whatever the crossover is set at) are excluded from the power rating. The power supply doesn't magically get any beefier. A good fraction of an amp's power is used driving low frequencies.



Cheers, Peter.



p.s. The Amp plus powered sub combo might now be 250W.
I had not thought of it that way; but if it is a doubling of the power then that is a lot gained by just adding a Sub. As much as is gained by the Sonos Amp over the Connect Amp.

The Sonos Amp becomes 250 wpc into 8 ohms?!
A small snippet of information gleaned in reading about power:

3 dB is a doubling of power, but it takes 10 dB to achieve a perceived doubling of loudness.

So 3 dB changes are noticeable but not big.
Presumably, since the Amp's gain is entirely under software control, Sonos can calibrate it however they wish for different volume settings.

Agreed. They could have no boost at low volumes and larger boost at high volumes. Although, it is a bit of a coincidence that both hodgeal (in the original LS50 thread) and I found an exact 3 dB boost at whatever volume we tested (I was at 30%, I don't know about hodgeal).



Cheers, Peter.
Of course, this will mean the lowest possible volume is also increased.

Presumably, since the Amp's gain is entirely under software control, Sonos can calibrate it however they wish for different volume settings.
p.s. I missed reading page 3 of this thread when posting just now. I can buy the argument that Sonos decided to use the additional headroom available when low frequencies are removed to make the Amp a bit louder. Of course, this will mean the lowest possible volume is also increased.