Different approaches to managing bass

[NTK]

I'll have to say I am skeptical. IMHO, there are plenty of misleading and questionable information in the YT vid.

Jon Herron said that the "traditional" multi-sub approach "flood" the zone with multiple sources and no control creates "overlapping modes". What exactly does that mean? Does MSO or Harman's SFM (sound field management) counts as control? I'd think so.

Jon was suggesting that multi-sub works because of "cacophony", and with MSO (and SFM),I beg to disagree. The MSO approach is based on measurements and mathematical optimization to arrive at the sub EQ/level/delay settings. As Dr Toole said in his Audioholics article, SFM is done usually with 4 subs, fewer than the numbers show in many of the configurations shown in the vid. Dr Geddes' approach favors "different" (as opposed to random) placements more. And I wouldn't call averaging, which is a totally valid method of evening things out, "cacophony".

Also, room modes are the characteristics of the room and the room only, and not of the source(s) or listener(s). You don't "add" a mode by adding a source. You excite a mode if a source is placed near a mode peak, for that mode frequency. You can also "cancel" the mode when you place 2 sources at mode peaks of opposite polarity. More sources has the chance of exciting more different modes at different frequencies, which usually is good for averaging.

I also find the idea of steering the wavefront of subwoofer frequencies away from walls in a residentially sized room dubious, especially with a small number of subwoofers randomly distributed (i.e. unspecified numbers at unspecified locations).

Here is my simulation of a 4-driver linear array, trying to steer the wavefront to angle 30 deg towards the top. The room is 6 m by 6 m, and the span of the drivers is 4 m. To show the undisturbed wavefront, the simulation is done with no wall reflection (all walls are perfect sound absorbers). The width of the lobes and number of the lobes are both frequency dependent. I am not convinced that this will work well in real life, and Trinnov hasn't shown any mathematics or data on the effectiveness of steering that show how much better their approach works.

 

[neRok]

Here is my simulation of a 4-driver linear array, trying to steer the wavefront to angle 30 deg towards the top. The room is 6 m by 6 m, and the span of the drivers is 4 m. To show the undisturbed wavefront, the simulation is done with no wall reflection (all walls are perfect sound absorbers). The width of the lobes and number of the lobes are both frequency dependent. I am not convinced that this will work well in real life, and Trinnov hasn't shown any mathematics or data on the effectiveness of steering that show how much better their approach works.

Was this done with increasing delays on each driver away from corner -3?

The graphic on the right looks like it could/is being run as a cardioid array. Would it work that way?

 

[Alice of Old Vincennes]

According to @René - Acculution.com:

Many different multi-sub approaches could now be tried out but let us focus on the so-called Double Bass Array method. Here, the strategy basically is to turn the room into a tube. Tube acoustics is a little more involved than room acoustics (and a good topic if you want to really understand room acoustics better), but there is one major thing to understand: “The plane wave will always propagate”. Tube modes differ from room modes in that the former type really exist at all frequencies at once and will either propagate if excited above an associated modal frequency or die out as an evanescent wave below it. The only wave that will always propagate is the plane wave, and one way to achieve a propagating plane wave is to excite both the front wall and the rear wall in a pistonic-like fashion as shown in the animation below.

The front wall has its particular displacement, and the rear wall has a modified displacement to account for the phase shift coming from the wave travelling the room distance. You now effectively sit in an infinite tube with only a travelling wave and no reflection.



I don't really care what "problematic" means to one person. You need prior knowledge of the goals set forth for any given project/application.



So that "revolutionary" EQ approach again? People have been equalizing without "VBA" successfully for a good while now.

Rene's description sounds like goofy pseudoscience.

 

[OCA]

I recognize the concept that inversion can produce comparable outcomes to a VBA filter (though this is somewhat diminished with classical least squares optimized IIR filters). While I comprehend the disdain for imitative filters, I'm seeking clarity on the extensive enthusiasm surrounding a filter, akin in principle, employed by a prominent figure with exclusive control over multichannel FIRs.

 

[NTK]

Was this done with increasing delays on each driver away from corner -3?


The graphic on the right looks like it could/is being run as a cardioid array. Would it work that way?

Δt (time delay of each successive driver) is calculated as shown below (in our case c = 343 m/s). The Trinnov system cannot turn standard subwoofers into cardioids.
(Reference)

 

[OCA]

has the multi-sub aprouch actually been mathematicly proven, or is it all anecdoticly?

In a simple simulation in REW for example I fail to find benefits for 3 people on a sofa:

2 subs



View attachment 305746


1 sub

View attachment 305747

I frequently experience second sub to be of minor benefit and not worth all the calibration overhead in my room. I remember having read somewhere that they really start adding up after the third and four is optimal but I am not entirely certain.

 

[neRok]

The Trinnov system cannot turn standard subwoofers into cardioids.

If it can control each sub individually, then why not? AFAIK cardioid system/config can be set up with 2 normal subwoofers, and the sound is projected in the direction of the "forward" subwoofer (as the below graphic calls/shows it). So if you rotate this set up and put it on a wall, like in the Trinnov graphic, doesn't that mean the sound will project more in 1 direction than the other? Maybe it's not ideal because the drivers aren't in the proper orientation like the attached image, but if the bass is omnidirectional, then does it really matter?

 

[Andrej]

If it can control each sub individually, then why not? AFAIK cardioid system/config can be set up with 2 normal subwoofers, and the sound is projected in the direction of the "forward" subwoofer (as the below graphic calls/shows it). So if you rotate this set up and put it on a wall, like in the Trinnov graphic, doesn't that mean the sound will project more in 1 direction than the other? Maybe it's not ideal because the drivers aren't in the proper orientation like the attached image, but if the bass is omnidirectional, then does it really matter?

View attachment 306617

What is the bandwidth of the directional sound? At what frequency should one optimize the most directional radiation? Surely it only works properly at one wavelength, unless one can create varying phase shifts/time-delays for each frequency so that the distance from the front to the back driver works (they will be out of phase as needed). Assuming that it is practical to do it.

 

[neRok]

What is the bandwidth of the directional sound? At what frequency should one optimize the most directional radiation? Surely it only works properly at one wavelength, unless one can create varying phase shifts/time-delays for each frequency so that the distance from the front to the back driver works (they will be out of phase as needed). Assuming that it is practical to do it.

I'm not an expert, so I don't know those answer, and I don't have 2 subs to test with. But I found this simple simulator: https://jahudka.github.io/subsim/



Cardioid seems to pretty evenly cancel a lot of rearward sound across the subs full spectrum. I reckon that would help room modes build up slower. It does seem that the lowest frequencies might suffer a bit, but things are likely very different in room when considering room gain etc.

You can also load "end fire" and see that it targets a specific frequency.

 

[neRok]

Since posting in this thread I've learned that room modes behave as high-Q IIR EQ filters, so a peaking room mode can be corrected. Yesterday I was going through a backlog of bookmarks I've made and found the comment "the above-mentioned "Selective Mode Cancellation" technique, in Toole's 2002 presentation, would also be an opportunity to contrast it with Geddes' preference for maximizing modes.". A comment from Floyd Toole follows. I have yet to read up on either document/technique/preference mentioned.

But that got me thinking, perhaps "Geddes' preference for maximizing modes" is so that they can be simply corrected with an EQ filter?

 

[AudioKC]

I did some room treatment with furniture and panels.

Biggest game changer for me was Dirac Live Bass Control with multiple subs in system.

 

[Flaesh]

DBA. This can be considered a combination Cancel the reflections and active Room treatment

Rene's description sounds like goofy pseudoscience.

Why do you think so?
I have been living in a tube with a [roughly approximated] traveling plane wave for ~7 years.

 

[youngho]

But that got me thinking, perhaps "Geddes' preference for maximizing modes" is so that they can be simply corrected with an EQ filter?

No, you can read a little about Geddes' approach at http://www.gedlee.com/Papers/Philosophy.pdf or https://mehlau.net/audio/multisub_geddes/

 

[Justdafactsmaam]

In another thread, @ernestcarl posted a video with an interview with Trinnov's Jon Herron. Most of the discussion was about Trinnov's "waveforming" technology - for me it was a fantastic video and I recommend watching the whole thing. However, this slide was of particular interest and I would like to see what ASR members thought of it.

View attachment 305481

For some reason ASR won't let me embed the video, so here is a link: Link to video (timestamped 33:49)

The traditional approach of "flooding the room" includes strategies by MSO and Harman, where multiple subwoofers are used to even out the bass response. I am not sure, but I think that Room Shaper (@mitchco review here) also uses the same approach. "Control the zone" refers to Trinnov's new DSP, and although it is not mentioned - probably also Dirac ART and VBA's (he does mention DBA's though). This approach is wasteful of subwoofer power because half the output goes into cancelling reflections. There is also bass equalization which he does not mention, where you simply lop off the peaks. And of course room treatment is the last option in my poll. My understanding is that it is intrusive and less effective.

Jon Herron says that Trinnov's approach minimizes standing waves in the room, reduces the "cacophony of reflections" that obscure bass detail, whilst also minimizing seat to seat variations in bass. Of course, he works for Trinnov and he may be biased. So I was wondering whether this was a true discussion of the pros and cons of different approaches to bass management.

I have been working on my own VBA. Although it works, it is no better than bass equalization. I have measured in multiple positions in the room, and differences are quite minor. There is no difference in the spectrogram in terms of bass decay time. Part of me wonders whether this was because my VBA was poorly designed, and whether I could get a better result by refining it. But the lazy part of me says that the result with bass equalization is already satisfactory, and I should just call it a day.

So: what approach do you use, and what are the pros and cons of each approach?

My approach was and will continue to be neither. I was already all about bass trapping and my next move is to take it further with a room rebuild.

Flooding is IMO a half way solution.

I was seriously considering the Trinnov system. But for two channel stereo it presents one big issue. It’s mono. Being mono it’s only going to work for 20hz- 40hz if that. Cross over higher and you get into messing up spatial cues.

So for me it’s going to be bass traps, a massive wide band trap, targeted diaphragmatic traps, active traps and DSP room correction to finish the job.

 

[Justdafactsmaam]

I'll have to say I am skeptical. IMHO, there are plenty of misleading and questionable information in the YT vid.

Jon Herron said that the "traditional" multi-sub approach "flood" the zone with multiple sources and no control creates "overlapping modes". What exactly does that mean? Does MSO or Harman's SFM (sound field management) counts as control? I'd think so.

Jon was suggesting that multi-sub works because of "cacophony", and with MSO (and SFM),I beg to disagree. The MSO approach is based on measurements and mathematical optimization to arrive at the sub EQ/level/delay settings. As Dr Toole said in his Audioholics article, SFM is done usually with 4 subs, fewer than the numbers show in many of the configurations shown in the vid. Dr Geddes' approach favors "different" (as opposed to random) placements more. And I wouldn't call averaging, which is a totally valid method of evening things out, "cacophony".

View attachment 306529

Also, room modes are the characteristics of the room and the room only, and not of the source(s) or listener(s). You don't "add" a mode by adding a source. You excite a mode if a source is placed near a mode peak, for that mode frequency. You can also "cancel" the mode when you place 2 sources at mode peaks of opposite polarity. More sources has the chance of exciting more different modes at different frequencies, which usually is good for averaging.

I also find the idea of steering the wavefront of subwoofer frequencies away from walls in a residentially sized room dubious, especially with a small number of subwoofers randomly distributed (i.e. unspecified numbers at unspecified locations).

View attachment 306530

View attachment 306531

Here is my simulation of a 4-driver linear array, trying to steer the wavefront to angle 30 deg towards the top. The room is 6 m by 6 m, and the span of the drivers is 4 m. To show the undisturbed wavefront, the simulation is done with no wall reflection (all walls are perfect sound absorbers). The width of the lobes and number of the lobes are both frequency dependent. I am not convinced that this will work well in real life, and Trinnov hasn't shown any mathematics or data on the effectiveness of steering that show how much better their approach works.

View attachment 306532

They’ve shown data. Before and after waterfall plots. Their system works

 

[Elder]

Went with room treatment on the survey since only one could be selected, but I will be using treatment alongside bass eq and "flooding the room" with multiple subwoofers.

Hoping the combination of 3 yields some good results, have wanted to improve my bass accuracy for years.

 

[ViperDom]

My current preferred method is:
- Multiple subs / LF sources,
- Adjust delays to get max/preferred summation,
- EQ