Room modes

[Justdafactsmaam]

View attachment 342564

Waterfall:
View attachment 342568 View attachment 342570 View attachment 342571

Spectral Decay:
View attachment 342572 View attachment 342573 View attachment 342574

Morlet CWT:
View attachment 342575 View attachment 342576 View attachment 342577

*Absent equalization, the peaky bass frequency response in a highly dampened room decays more evenly -- EQ works better as well.

Looks to me like there is still plenty of ringing. As I expected.

 

[ernestcarl]

Looks to me like there is still plenty of ringing. As I expected.

EQ can work very well and not so well. Higher resolution EQ is not a panacea; and, left to the hands of the inexperienced human or not so optimally programmed “correction” computer algorithm could actually make things worse. Interestingly, installing just a few expensive “bass traps” in certain rooms may still prove highly insufficient (actual structural damping works even better for low frequencies): https://gearspace.com/board/studio-...dio-room-sound-post13225996.html#post13225996

 

[Justdafactsmaam]

EQ can work very well and not so well. Higher resolution EQ is not a panacea; and, left to the hands of the inexperienced human or not so optimally programmed “correction” computer algorithm could actually make things worse. Interestingly, installing just a few expensive “bass traps” in certain rooms may still prove highly insufficient (actual structural damping works even better for low frequencies): https://gearspace.com/board/studio-...dio-room-sound-post13225996.html#post13225996

Everything I have learned about room acoustics point toward bass traps as being the best solution for bass management but also the most inconvenient. It’s intrusive and expensive. But it is the most direct and comprehensive solution with no technical downside.

 

[NTK]

Looks to me like there is still plenty of ringing. As I expected.

The "ringing" is the natural impulse response of a system with a bandpass type filter behavior as shown by the frequency response curve.

 

[OCA]

Everything I have learned about room acoustics point toward bass traps as being the best solution for bass management but also the most inconvenient. It’s intrusive and expensive. But it is the most direct and comprehensive solution with no technical downside.

Most practical bass traps will trap everything else but not the bass though! I receive a large number of measurements from people and find it notoriously difficult to digitally fix phase issues in heavily treated rooms.

 

[Justdafactsmaam]

The "ringing" is the natural impulse response of a system with a bandpass type filter behavior as shown by the frequency response curve.

View attachment 342709

The posted waterfall plots show exactly what I was talking about. After flattening the initial attack with EQ the room mode frequencies have a substantially slower decay than the adjacent frequencies. Ringing

 

[Justdafactsmaam]

Most practical bass traps will trap everything else but not the bass though! I receive a large number of measurements from people and find it notoriously difficult to digitally fix phase issues in heavily treated rooms.

If they aren’t trapping bass they aren’t really bass traps are they? I acknowledge that effective bass trapping is intrusive, costly and less than practical. Most bass traps simply are inadequate. State of the art bass management is hard. No way around it. But true state of the art bass management will involve a great deal of acoustic bass trapping. Maybe active DSP systems such as the Trinnov Waveforming technology will supersede it. Maybe. But even the folks and Trinnov highly recomend doing as much as possible with acoustic traps first, then using their tech after.

I’m not saying EQ is a bad way to deal with bass management. Just that it is inherently compromised.

 

[NTK]

The posted waterfall plots show exactly what I was talking about. After flattening the initial attack with EQ the room mode frequencies have a substantially slower decay than the adjacent frequencies. Ringing

When you have a device that cannot reproduce DC and high frequencies, such as a subwoofer, that's the impulse response you get, in the best case. It is physics.

 

[OCA]

After flattening the initial attack with EQ the room mode frequencies have a substantially slower decay than the adjacent frequencies

That's correct but it's a flaw in the design of that filter not a direct consequence of an optimal filter.

 

[OCA]

Just that it is inherently compromised.

I only disagree with that part of your reply. Most filters are designed to achieve a desired target curve in the room by inverting the room effect around that target curve ignoring the speakers true capacity (actual bass roll off). In fact, mathematically and otherwise only min phase responses can be inverted. When min phase speaker response at the mic position is inverted, you will end up with the speakers true (anechoic) response which will not include any room reflections. As long as room effects are left in the final response, you will have group delays (hence the sustained ringing we see in these graphs) because the additional bass frequencies are produced by wall reflections which are by definition "delayed" relative to the direct sound.

 

[Justdafactsmaam]

When you have a device that cannot reproduce DC and high frequencies, such as a subwoofer, that's the impulse response you get, in the best case. It is physics.

Are you saying it’s the speaker not the room?

 

[Justdafactsmaam]

That's correct but it's a flaw in the design of that filter not a direct consequence of an optimal filter.

Do you have an example of one that has no such flaw?

 

[Justdafactsmaam]

I only disagree with that part of your reply. Most filters are designed to achieve a desired target curve in the room by inverting the room effect around that target curve ignoring the speakers true capacity (actual bass roll off). In fact, mathematically and otherwise only min phase responses can be inverted. When min phase speaker response at the mic position is inverted, you will end up with the speakers true (anechoic) response which will not include any room reflections. As long as room effects are left in the final response, you will have group delays (hence the sustained ringing we see in these graphs) because the additional bass frequencies are produced by wall reflections which are by definition "delayed" relative to the direct sound.

Not sure I follow. Are you saying the ringing is an artifact of the measurement but not actually present in the room?

 

[OCA]

Not sure I follow. Are you saying the ringing is an artifact of the measurement but not actually present in the room?

No, I am trying to say if you EQ targeting the "spec" target response (anechoic response) of a speaker at the LP, you will improve ringing problems as well. In other words, digital filters are fully capable of achieving an optimal correction at a "single point" in the room.

 

[Justdafactsmaam]

No, I am trying to say if you EQ targeting the "spec" target response (anechoic response) of a speaker at the LP, you will improve ringing problems as well. In other words, digital filters are fully capable of achieving an optimal correction at a "single point" in the room.

How would such a filter do that unless it is sending a custom cancellation signal? You put the energy in the room the room takes over. Unless you are actively cancelling the waves at or near the room boundaries.

 

[OCA]

Do you have an example of one that has no such flaw?

Certainly, from real data:

This is the original response, the filter (I currently actively use) and the response after filter in my room at the MLP for my left speaker:

and these are the group delay, RT60 (T30) and musical clarity comparisons respectively before and after the filter:

-almost 10ms improvement in GD at still audible frequencies

 

[OCA]

How would such a filter do that unless it is sending a custom cancellation signal? You put the energy in the room the room takes over. Unless you are actively cancelling the waves at or near the room boundaries.

That's the magic of a digital filter. With the correct technique, it can do wonders but again ONLY at one point in the room. I have developed custom cancellation signals in the past, too but they need to be delayed as well and although they can produce a lot more bass than the speaker is designed to achieve (and some people prefer that), they still have flaws and delayed bass rains onto the following mid and high tones and cause muddiness I disliked. I haven't posted a video in months, I will probably do one explaining how to create the filter above when I have the strength to suffer video editing

 

[OCA]

To clarify the above, if I had targeted Harman curve to create the filter instead of the anechoic speaker response (which I could have given the response shape at the LP), it would have the artifacts you mentioned.

 

[Justdafactsmaam]

That's the magic of a digital filter. With the correct technique, it can do wonders but again ONLY at one point in the room. I have developed custom cancellation signals in the past, too but they need to be delayed as well and although they can produce a lot more bass than the speaker is designed to achieve (and some people prefer that), they still have flaws and delayed bass rains onto the following mid and high tones and cause muddiness I disliked. I haven't posted a video in months, I will probably do one explaining how to create the filter above when I have the strength to suffer video editing

Ok so this obviously isn’t just a little EQ. This is more a DSP correction.

 

[OCA]

Ok so this obviously isn’t just a little EQ. This is more a DSP correction.

Correct but generating it is even simpler than using REW's auto EQ feature and takes less than 5 minutes.