Virtual Bass Array

VBA, DBA and MIMO:

A DBA is set up as a source and sinc configuration, whith a front that ideally covers the front wall so well that it more or less operates as a "rectangle" source. Not a point source, not a line source, more like a two-dimensional line source that produces a uniform wave from side wall to side wall and from ceiling to floor. The rear subs absorb the bass, so that the back wall becomes virtually non-reflective for low frequencies. It's like playing bass in a room with the back wall removed.

While DBA is implemented as mono bass, MIMO is a more general solution that stands for "multiple in - multiple out". You can e.g have a stereo bass arrangement where you use left and right front as the sources, and as many subs around that operates as bass absorbers for left and right. And the left could absorb the right bass and vice versa. Very related to the above, but more complex ... active bass absorbation.

A VBA is an attempt to do some of the same, except that the sound absorbation is only done by the front speakers who play a delayed pulse in opposite phase of the original. A VBA has limitations compared to a (true) VBA since the front speakers still have to deal with the return sound from the back wall. And even a successful implementation will not have as uniform waveform as a true DBA, since it's only one or two point sources involved.

The potential problem of VBA unless it is done with great care is that excess phase problems can be magnified. These excess phase problems occur on frequencies where the sum of reflected sound is out of phase and stronger than the direct sound from the speaker. So instead of getting cancelled, these may be amplified by the virtual sinc operation and make things worse. This problem does not occur in a real DBA, where the back wall really has been silenced.

The TTD correction in Audiolense cancels room reflections and corrects the low frequencies to the degree allowed by the user (TTD window size), with the possible moderation of skipping those frequencies that have excess phase problems (the selective pre-ringing option). It literally does the the same as a VBA solution could possibly do, but perhaps with even better control and moderation. TTD is technically capable of producing a close to perfect phase and decay pattern in the deep bass in most rooms, but this will often lead to audible correction artifacts. The same would be the case for a successful VBA. Bottom line is I do not see a principal difference between VBA and TTD. Also, remember that the listening seat measurement includes all the room reflections.

A couple of things to round off with: In a decent room the TTD correction can give you excellent bass in the listening seat(s) even from a regular pair of stereo speakers or a single subwoofer ... the same as you could theoretically get from a VBA approach. I've seen excellent measurements of DBA and the listening verdicts are very good. But I have also seen at least as good results in well damped rooms with huge bass capacity installed. MIMO is still new. Trinnov and Dirac have their own versions, but I haven't heard clear verdicts that these solutions produces better bass. Dirac has been on it for years, and I suspect a practical MIMO implementation can involve acoustic challenges, especially in the transition region where the sinc's are phased out. The math needed ... various versions ... have been ready for years..But regardless of whether the speakers are corrected as single entities or by a MIMO-approach, more woofers can always be set up to produce a more uniform bass across the room than fewer. And more membrane areal is always better than less.

maxand said:

Thanks Bernt, that makes it a lot clearer but just wondering if you could expand on the audible artifacts you mention.

Click to expand...

With Audiolense, you can get audible artifacts in the form of pre-ringing and/or reduced clarity if you try to time-correct uncorrectable frequencies. So the software is designed to avoid that. We're not talking about problems that are frequently in your face here. Audible pre-ringing typically only occurs on certain tunes from certain types of music. I never heard it with my own music before I deliberately started to searc for it and it took two or three year before the problem surfaced. But when I found a working solution it turned out that it also led to a cleaner response also in systems where the pre-ringing problem was not heard before.