Audibility of group delay at low frequencies

[dc655321]

Top : linear phase parametric bass boost. Group delay = 0. Some pre-ringing is visible.

How does one get zero group delay from a linear phase filter?

 

[Pio2001]

How does one get zero group delay from a linear phase filter?

By setting the time reference on the peak of the impulse response.
In this comparison, the overall delay is not interesting for us, as it doesn't change the sound in itself.

A better example is preringing. It is called "pre" ringing because it occurs before the reference time (although it occurs after the original signal is fed into the filter).

 

[dc655321]

By setting the time reference on the peak of the impulse response.

But it's not zero group delay - it is simply time shifted.
Maybe it's a small semantic thing, but it feels misleading to characterize the effect you've demonstrated as "zero group delay".

 

[mansr]

But it's not zero group delay - it is simply time shifted.
Maybe it's a small semantic thing, but it feels misleading to characterize the effect you've demonstrated as "zero group delay".

Call it constant rather than zero if you prefer. Same result, but zero is easier to work with.

 

[QMuse]

Call it constant rather than zero if you prefer. Same result, but zero is easier to work with.

That is not what he meant. I think there is a confusion here.

Here is impulse response with start adjusted to 0:

While this brings GD to 0 at HF it certainly doesn't bring it to 0 at all freqs:

 

[mansr]

That is not what he meant. I think there is a confusion here.

Here is impulse response with start adjusted to 0:

You're never going to get perfect results once speakers and rooms get involved.

 

[QMuse]

You're never going to get perfect results once speakers and rooms get involved.

That's for sure. I was merely trying to clear the cconfusion here, not to claim otherwise.

 

[QMuse]

A better example is preringing. It is called "pre" ringing because it occurs before the reference time (although it occurs after the original signal is fed into the filter).

Pre-ringing has nothing to do with this - it is an artifical effect coming from convolving a linear phase filter.

 

[mansr]

That's for sure. I was merely trying to clear the cconfusion here, not to claim otherwise.

Now I'm confused.

 

[QMuse]

Now I'm confused.

What @dc655321 correctly stated is that adjusting ref time of the impulse response to 0 will not result in GD being 0 at all frequencies, which may be implied from the way @Pio2001 has put it.

 

[dasdoing]

what would a natural group delay, let's say of a waterfall, look like? all 0, or rising at bass?

 

[hyperplanar]

what would a natural group delay, let's say of a waterfall, look like? all 0, or rising at bass?

A waterfall makes a continuous sound and I don’t think group delay can be used to describe a sound generating source (not a transducer) anyways. But to entertain the idea, an impulsive sound like a gunshot in an anechoic chamber might be better. I would expect a flat group delay if you were to treat a recording of the sound as an impulse response and analyze it.

 

[QMuse]

A waterfall makes a continuous sound and I don’t think group delay can be used to describe a sound generating source (not a transducer) anyways. But to entertain the idea, an impulsive sound like a gunshot in an anechoic chamber might be better. I would expect a flat group delay if you were to treat a recording of the sound as an impulse response and analyze it.

Phase and group delay are measured as speaker output relative to the input signal. It makes no sense to measure phase and GD of the input signal.

 

[mansr]

What @dc655321 correctly stated is that adjusting ref time of the impulse response to 0 will not result in GD being 0 at all frequencies, which may be implied from the way @Pio2001 has put it.

With a perfectly symmetrical impulse response, that is exactly what it does.

 

[dasdoing]

A waterfall makes a continuous sound and I don’t think group delay can be used to describe a sound generating source (not a transducer) anyways. But to entertain the idea, an impulsive sound like a gunshot in an anechoic chamber might be better. I would expect a flat group delay if you were to treat a recording of the sound as an impulse response and analyze it.

so, a perfect group delay for our speakers in our room would be 0 excess group delay? since a flat group delay seams to be only possible with totaly flat frequencie response

 

[QMuse]

With a perfectly symmetrical impulse response, that is exactly what it does.

As I explained in my previous post, phase and group delay are measured as speaker output relative to the input signal (usually sine sweep over the freq range). Only ideal speaker would output perfectly symmetrical IR with phase and GD being 0, but something like that doesn't exist in a real world. @Pio2001 certainly wasn't speaking about phase and GD of "ideal" speaker.

 

[hyperplanar]

so, a perfect group delay for our speakers in our room would be 0 excess group delay? since a flat group delay seams to be only possible with totaly flat frequencie response

Perfect would be both zero minimum and excess group delay. The minimum group delay is caused by the FR as you’ve alluded to, which is a possible reason one would want to extend FR as low as possible (even if the max SPL achievable is not that high at these low frequencies). And the excess is from other sources. But excess is not correctable without preringing.

 

[mansr]

As I explained in my previous post, phase and group delay are measured as speaker output relative to the input signal (usually sine sweep over the freq range). Only ideal speaker would output perfectly symmetrical IR with phase and GD being 0, but something like that doesn't exist in a real world. @Pio2001 certainly wasn't speaking about phase and GD of "ideal" speaker.

Now I'm even more confused.

 

[dasdoing]

But excess is not correctable without preringing.

you can to a certain point if you have a subwoofer

 

[hyperplanar]

you can to a certain point if you have a subwoofer

Well yes, if you high pass the speaker where its excess group delay is high and have a subwoofer with a lower excess group delay that would help. This mainly helps lower minimum group delay due to the frequency response extension though. It’s worth mentioning the crossover between the subwoofer and speakers itself also adds excess group delay.

I forgot to consider, setting the delay incorrectly for the speakers/subwoofer can add excess group delay. But assuming everything has been set up correctly, the remaining excess group delay can’t be solved without preringing.