Audibility of group delay at low frequencies

[Kvalsvoll]

Audibility of group delay at low frequencies:

https://www.kvalsvoll.com/…/audibility-of-group-delay-at-l…/

It is then confirmed and proven that group delay at low frequencies is audible.

This means timing - GD, phase - matter for sound in the bass range, it is not sufficient to tune towards a flat frequency response alone.

I have provided sound samples so you can try and hear yourself.

 

[pozz]

Shouldn't we distinguish between acoustic and signal-based sources of group delay?

If you increase the GD of a track's bass through a DAW and then listen to it through your soundsystem, aren't you compounding two sources and exagerrating the effects?

Edit: I put that badly. What I meant to say was: shouldn't we distinguish between the GD already inherent in your system and the GD you introduce?

 

[mansr]

Could you share the actual filters you used? That would let us test with other samples than the ones you made.

 

[Krunok]

Audibility of group delay at low frequencies:

https://www.kvalsvoll.com/…/audibility-of-group-delay-at-l…/

It is then confirmed and proven that group delay at low frequencies is audible.

This means timing - GD, phase - matter for sound in the bass range, it is not sufficient to tune towards a flat frequency response alone.

I have provided sound samples so you can try and hear yourself.

Interesting..

Can you please show phase measurement of your system?

 

[DonH56]

I couldn't access the link (will try tonight). If the group delay is off then integration with the mains will be off resulting in frequency response aberrations. Are you saying group delay alone is audible when isolated to the sub, or when group delay of the sub is varied with respect to the mains? Does this assume constant group delay (linear phase)?

 

[Kvalsvoll]

Article is now updated with spectrogram charts and subjective impressions form listening in Room2.

 

[Kvalsvoll]

I couldn't access the link (will try tonight). If the group delay is off then integration with the mains will be off resulting in frequency response aberrations. Are you saying group delay alone is audible when isolated to the sub, or when group delay of the sub is varied with respect to the mains? Does this assume constant group delay (linear phase)?

What causes the delay does not matter. Can be due to room acoustics, delay in subwoofers/bass-system, crossover between main/bass-system.

The objective here was to find out whether GD actually can be a problem, and the first thing to investigate then, is to determine if this is audible. This is what the article describes.

Frequency response is the same, it is only delay that changes across the frequency range, and to make that happen, there has to be non-minimum-phase phenomena present. In the experiment, I create it on software.

 

[Kvalsvoll]

Interesting..

Can you please show phase measurement of your system?

Phase measured in Room2, but it doesn't give much usable information - note that phase wraps at +-180 deg:


The spectrogram shows more relevant information on what is actually going on (100ms/40dB scale):

Compare to the 20ms processing:


When I listen to the 20ms, I listen to the Room2 on top of the 20ms. If my room is significantly worse than the 20ms, it is possible that what I hear from the 20ms processing will be masked by the room/playback system. This is briefly mentioned in the article.

 

[amirm]

Can you record the sound in the room with and without delay?

 

[Krunok]

How exactly did you measure GD, from which distance? Is the room anechoic?

The moment you put speaker in a "normal" room reflections will cause GD of more than 20ms up to several kHz.

 

[Krunok]

Phase measured in Room2, but it doesn't give much usable information - note that phase wraps at +-180 deg:
View attachment 31742

Well it should, as REW calculates GD from the slope of the phase trace. Can you please show unwrapped phase graph and adjust the Y scale?

That phase graph doesn't look particulalry flat and in order to have flat GD near 0 it should be flat.

 

[Kvalsvoll]

Shouldn't we distinguish between acoustic and signal-based sources of group delay?

If you increase the GD of a track's bass through a DAW and then listen to it through your soundsystem, aren't you compounding two sources and exagerrating the effects?

Edit: I put that badly. What I meant to say was: shouldn't we distinguish between the GD already inherent in your system and the GD you introduce?

What I - or you - listen to, will be the GD of our playback system added on top of the different processed sound samples. The system GD will also be present when listening to the original sample, and if this GD is very large compared to the added GD in the 20ms or 100ms, it may not be possible to hear much differences.

The objective of this experiment was to find whether group delay is audible, so we can determine whether this is something we need to pay attention to. Obviously, we need a playback system with GD less than the samples, to be able to do the test.

 

[KSTR]

Nice investigation!

I did a bunch of experiments on this a while back and the outcome was basically this:

The influence of group delay (and its equivalent phase response) on perception is multi-faceted.

There is a time-domain effect, excessive GD at low freq makes "the bass lag behind", eg smearing the "compactness" of bass drums and plucked upright string bass, ....

.... and there is also a timbre shift for quasi steady-state notes when they cointain "even order distortion" overtone profile with the proper phase so that the waveform is highly asymmetrical. It turns out that we seem to be sensitive to actual waveshape at low frequencies and shifting around phases of the harmonics by additional phase contribution from crossovers and bass alignments impacts the percieved timbre. Flipping the absolute polarity can have a very similar effect on timbre for its likewise apparant effect on harmonics phases.

A test signal for this can easily be constructed: take a eg 80Hz fundamental, then add equal level 2nd harmonic but de-tuned by 0.5Hz, so 160.5Hz.
This gives two distinct and stable lines in an FFT plot but perception tells, beside the slightly out-of-tune harmonic, there is a cyclic slight change of timbre at a 2 seconds rate going from "fatter" to "leaner" and back, exactly linked to a realtime oscilloscope display of the waveform where we can see the 2nd harmonic "cycling" slowly on top of the fundamental. The nice feature of this test that it works with any speaker or headphones(preferred) no matter what their phase response is. Low distortion is required, though, therefore headphones at low signal levels should be preferred for listening.
The de-tuning makes it not a good choice for an actual ABX. For this, a set of files is required, with the exactly pitched H2 content aligned in 30° or 45° steps vs the fundamental, through the whole 360° cycle. There will be one specific set -- depending on playback system total phase -- of offsets spaced 180° apart that will show the strongest timbre shift when compared, and another 180° set, this time rotated by 90° against the first which shows almost no timbre change when switched. The "strong" set can then be successfully used in ABX.

A "semi-scientific" paper somewhat relevant in this context: https://www.researchgate.net/public...akes_auditory_sense_a_new_paradigm_in_hearing

 

[DonH56]

I'm confused. A constant broadband group delay just adds latency so should be inaudible. If you add delay only to the low frequencies and not the higher frequencies (or vice-versa) then you introduce all sorts of problems by not time-aligning the signal. In that case steady-state frequency response, assuming no crossover or the like that causes interaction among frequencies, would not show anything. Time-domain analysis like step or impulse response will show problems as group delay is changed for one set of frequencies and not the others. That introduces a phase discontinuity, natch.

 

[Krunok]

Here is how reflections influence GD (and everything else related to time domain):

red - no windowing (sweep taken at LP app 4m from speakers)
green - 2ms IR window
blue - 5ms window
beige - 10ms window

 

[Kvalsvoll]

Could you share the actual filters you used? That would let us test with other samples than the ones you made.

I used Audacity, and the Phaser effect to create the GD samples.

Settings like this:

This creates the 20ms. For 100ms, simply apply the filter 4 times.

 

[Kvalsvoll]

Can you record the sound in the room with and without delay?

f
No, because I do not have suitable mics for that.

But why? It is much better to listen to the samples yourself. You could try headphones, I did not do that.

 

[Krunok]

The influence of group delay (and its equivalent phase response) on perception is multi-faceted.

Would you say that this GD trace is derived from this phase response? It doesn't look that way to me..

Phase:

GD:

 

[Kvalsvoll]

Nice investigation!

I did a bunch of experiments on this a while back and the outcome was basically this:

The influence of group delay (and its equivalent phase response) on perception is multi-faceted.

There is a time-domain effect, excessive GD at low freq makes "the bass lag behind", eg smearing the "compactness" of bass drums and plucked upright string bass, ....

.... and there is also a timbre shift for quasi steady-state notes when they cointain "even order distortion" overtone profile with the proper phase so that the waveform is highly asymmetrical. It turns out that we seem to be sensitive to actual waveshape at low frequencies and shifting around phases of the harmonics by additional phase contribution from crossovers and bass alignments impacts the percieved timbre. Flipping the absolute polarity can have a very similar effect on timbre for its likewise apparant effect on harmonics phases.

A test signal for this can easily be constructed: take a eg 80Hz fundamental, then add equal level 2nd harmonic but de-tuned by 0.5Hz, so 160.5Hz.
This gives two distinct and stable lines in an FFT plot but perception tells, beside the slightly out-of-tune harmonic, there is a cyclic slight change of timbre at a 2 seconds rate going from "fatter" to "leaner" and back, exactly linked to a realtime oscilloscope display of the waveform where we can see the 2nd harmonic "cycling" slowly on top of the fundamental. The nice feature of this test that it works with any speaker or headphones(preferred) no matter what their phase response is. Low distortion is required, though, therefore headphones at low signal levels should be preferred for listening.
The de-tuning makes it not a good choice for an actual ABX. For this, a set of files is required, with the exactly pitched H2 content aligned in 30° or 45° steps vs the fundamental, through the whole 360° cycle. There will be one specific set -- depending on playback system total phase -- of offsets spaced 180° apart that will show the strongest timbre shift when compared, and another 180° set, this time rotated by 90° against the first which shows almost no timbre change when switched. The "strong" set can then be successfully used in ABX.

A "semi-scientific" paper somewhat relevant in this context: https://www.researchgate.net/public...akes_auditory_sense_a_new_paradigm_in_hearing

I did experiments for absolute phase once, and could not hear any difference at all. I tested using both music signals and transient test signals.

 

[Krunok]

I did experiments for absolute phase once, and could not hear any difference at all.

You added phase shift in a similar way as you now added GD?