Does using EQ degrade sound quality in Wiim Pro?

[canard]

I have seen no evidence of this advanced algorithm to avoid clipping and it has never been mentioned in the forum before. I wonder who wrote it.

It's especially easy to check in a few clicks ;-)

 

[slartibartfast]

It's especially easy to check in a few clicks ;-)

Yet nobody has and it still is a common request. WiiM have never replied to those requests to say that automatic gain reduction is already in use.

 

[Burnside]

Yeah, the faq also says you can use Alexa to run a preset, but that’s not accurate so unless WiiM confirm it here, I’d take that entry as open to question as well.

 

[jed1]

A happy Wiim Pro owner here. Having the Wiim volume set to fixed I have been somewhat reluctant to use the EQ though.

In this and other threads on this forum regarding EQ, the consensus seems to be that that clipping can/will occur in a situation like mine: having fixed volume (or at 100 %) and using positive gain in the EQ.

Wiim though, states the following:
View attachment 18386
https://faq.wiimhome.com/en/support...ow-to-use-parametric-eq-a-comprehensive-guide

Is it unnecessary to decrease the volume or having digital headroom when applying positive EQ gain, since with the mentioned “advanced algorithm” there will be no clipping or increase of distortion, or am I missing something?

Good finding! I calculate an impulse response in time domain and measure peak amplitude to estimate the headroom. I use MATLAB/Octave.

 

[Wimsey]

Yet nobody has and it still is a common request. WiiM have never replied to those requests to say that automatic gain reduction is already in use.

Yeah, the faq also says you can use Alexa to run a preset, but that’s not accurate so unless WiiM confirm it here, I’d take that entry as open to question as well.

Hmmm ... I have read and heard so many good things about WiiM, their customer support and their efforts in improving their firmwares and the app. I have had the highest regard for WiiM and their product support.

Now I must quote that tall dark guy in Star Wars: “I find your lack of faith disturbing”.

 

[slartibartfast]

Hmmm ... I have read and heard so many good things about WiiM, their customer support and their efforts in improving their firmwares and the app. I have had the highest regard for WiiM and their product support.

Now I must quote that tall dark guy in Star Wars: “I find your lack of faith disturbing”.

You haven't been here long have you?

 

[Burnside]

You haven't been here long have you?

Both of me neither

 

[slartibartfast]

It's especially easy to check in a few clicks ;-)

What's the best way to test this? I was thinking of a +12dB filter at around 18kHz and listening for a reduced volume.
That wouldn't work for the Amp if the algorithm only limits the maximum volume if clipping is detected. It would work for the Ultra if the volume was set to 100%.

 

[canard]

What's the best way to test this? I was thinking of a +12dB filter at around 18kHz and listening for a reduced volume.
That wouldn't work for the Amp if the algorithm only limits the maximum volume if clipping is detected. It would work for the Ultra if the volume was set to 100%.

a sound card and rew etc etc.. hihi......
;-)
1khz 0db no correction
1khz 0db but correction +1db Q5, or in gec for example etc etc... ;-)
I was just thinking of a verification at the pro pro plus ultra level ;-)

 

[canard]

4mn... time to turn on the laptop

for the sake of conscience...just did it...overload...


like on the first day of the gec in spring 2022... ;-)

 

[slartibartfast]

4mn... time to turn on the laptop

for the sake of conscience...just did it...overload...


like on the first day of the gec in spring 2022... ;-)

So as we thought the algorithm doesn't exist.

 

[dominikz]

What's the best way to test this? I was thinking of a +12dB filter at around 18kHz and listening for a reduced volume.
That wouldn't work for the Amp if the algorithm only limits the maximum volume if clipping is detected. It would work for the Ultra if the volume was set to 100%.

I was intrigued so I did some tests with my WiiM Mini today.
First I created a severe EQ preset:


Then I measured WiiM Mini Aux output (2Vrms output level setting) with a 0dBFS peak signal - first without EQ (i.e. EQ disabled) at level 100%, then with the above EQ and volume level set to 100%, 90%, 80%, 70% and 60%:


As we can see, WiiM doesn't decrease the volume when you add a boost - it instead limits the output.

Note: I got the same behaviour if I kept volume at 100% and changed sweep level. Equivalent behaviour can also be seen if you use other output levels (e.g. 200 mVrms instead of 2 Vrms).

Looking at distortion plot for the above measurements we can see that this is indeed not clipping, because THD doesn't skyrocket in any of the measurement around 1kHz where the boost is:


My guess would be WiiM implemented some kind of soft-knee brick wall limiter post-EQ to avoid digital clipping - IMHO an indication of this is the slight rise in THD prior to 1kHz; I'd say this is where the rising digital signal starts to hit the limiter knee.

This means that as you come closer to maximum digital output any EQ boosts will progressively have less and less effect until they have no effect. At lower playback and/or lower volume levels all filters work as expected. This way no headroom is lost, but tonality can change at highest playback levels (i.e. peaks close to 0dBFS) since EQ boosts will no longer have as much effect.

IMHO it is a very reasonable compromise - average level of most content will not be near 0dBFS anyway, and it is similarly unlikely that most users would keep volume at 100%/fixed level all the time.
For the rare cases where these conditions do happen, WiiM implementation of EQ still avoids clipping by using a graceful form of limiting output, without having to decrease playback level. Even in such cases the change in tonality due to this implementation might not be very audible, since it would likely only affect waveform peaks in the content.

Have to say it is a very interesting and clever approach! I'm impressed, nice work @WiiM Team!

 

[canard]

I was intrigued so I did some tests with my WiiM Mini today.
First I created a severe EQ preset:
View attachment 18403

Then I measured WiiM Mini Aux output (2Vrms output level setting) with a 0dBFS peak signal - first without EQ (i.e. EQ disabled) at level 100%, then with the above EQ and volume level set to 100%, 90%, 80%, 70% and 60%:
View attachment 18404

As we can see, WiiM doesn't decrease the volume when you add a boost - it instead limits the output.

Note: I got the same behaviour if I kept volume at 100% and changed sweep level. Equivalent behaviour can also be seen if you use other output levels (e.g. 200 mVrms instead of 2 Vrms).

Looking at distortion plot for the above measurements we can see that this is indeed not clipping, because THD doesn't skyrocket in any of the measurement around 1kHz where the boost is:
View attachment 18405

My guess would be WiiM implemented some kind of soft-knee brick wall limiter post-EQ to avoid digital clipping - IMHO an indication of this is the slight rise in THD prior to 1kHz; I'd say this is where the rising digital signal starts to hit the limiter knee.

This means that as you come closer to maximum digital output any EQ boosts will progressively have less and less effect until they have no effect. At lower playback and/or lower volume levels all filters work as expected. This way no headroom is lost, but tonality can change at highest playback levels (i.e. peaks close to 0dBFS) since EQ boosts will no longer have as much effect.

IMHO it is a very reasonable compromise - average level of most content will not be near 0dBFS anyway, and it is similarly unlikely that most users would keep volume at 100%/fixed level all the time.
For the rare cases where these conditions do happen, WiiM implementation of EQ still avoids clipping by using a graceful form of limiting output, without having to decrease playback level. Even in such cases the change in tonality due to this implementation might not be very audible, since it would likely only affect waveform peaks in the content.

Have to say it is a very interesting and clever approach! I'm impressed, nice work @WiiM Team!

we had the use of a drc ( at -1db) which is possible on the dac ti of the mini and pro in the fall of 22, , to try to get around the situation,..then removed..
I just did it in coax output on a plus..overload

 

[dominikz]

we had the use of a drc ( at -1db) which is possible on the dac ti of the mini and pro in the fall of 22, , to try to get around the situation,..then removed..
I just did it in coax output on a plus..overload

Thanks for sharing!
It seems like a pretty graceful handling of overload though, given that THD is still -87dB.
If it was hard clipping, THD would be much higher.

So I think this is showing the same kind of limiting behavior I measured on the Mini analog aux out - just that in my case the artefacts are largely masked by Mini's analog output noise.

 

[canard]

 

[dominikz]

 

[canard]

Thanks for sharing!
It seems like a pretty graceful handling of overload though, given that THD is still -87dB.
If it was hard clipping, THD would be much higher.

So I think this is showing the same kind of limiting behavior I measured on the Mini analog aux out - just that in my case the artefacts are largely masked by Mini's analog output noise.

just sine 1khz... not complex sound
just 2 frequency (imd din).... same correction
etc
;-)
good night...

same situation with no autogain since 2022, in gec , presets gec , (or pec now)

 

[dominikz]

just sine 1khz... not complex sound
just 2 frequency (imd din).... same correction
;-)
good night...

same situation with no autogain since 2022

Nobody mentioned autogain, at least not that I've seen - the current EQ overload protection implementation looks like soft knee brickwall limiting to me.

I appreciate that some might prefer autogain or manual pregain to this, but note that this kind of implementation has its benefits too.

Either way I hope we can agree it is much better than hard digital clipping (which is what I expected prior to testing).

 

[slartibartfast]

Thanks for sharing!
It seems like a pretty graceful handling of overload though, given that THD is still -87dB.
If it was hard clipping, THD would be much higher.

So I think this is showing the same kind of limiting behavior I measured on the Mini analog aux out - just that in my case the artefacts are largely masked by Mini's analog output noise.

Why is flat topping below 0dB different from a distortion perspective than hard clipping at 0dB? The waveform must be very similar.

 

[canard]

Nobody mentioned autogain, at least not that I've seen - the current EQ overload protection implementation looks like soft knee brickwall limiting to me.

I appreciate that some might prefer autogain or manual pregain to this, but note that this kind of implementation has its benefits too.

Either way I hope we can agree it is much better than hard digital clipping (which is what I expected prior to testing).

I don't understand what you're saying...it requires gain adjustment in many cases...that's why autogain comes with correction functions, etc., and the proof is here...

ps
ps personally does not concern me the manipulation that I showed in fr with chirp allows me to test determine the gain adapted to a given correction in a few clicks..but the autogain would concern the majority of users.... and now even more with the rc mode which determines a "peq"
(volume "32bit")