Originally Posted by sdurani
It is as you describe: stereo bass sounds more spacious, multi-mono bass makes it easier to get a smoother response.
Mono bass (even from multiple subs) tends to have that thumping in-your-chest feel, while stereo bass feels like it's all around you (like the bass you hear at live events). So, stereo bass isn't about directionality, since low frequencies aren't localizable, but more about externalization. It's not to everyone's tastes, since most people find it kinda phasey sounding.
There are some caveats to hearing the effect. The content itself has to have truly decorrelated low frequencies, not summed to mono (like 99% of the recordings out there), which is hard to find short of some good recordings by John Eargle on the Delos label. Also, the lower in frequency you go, the wider apart the speakers need to be in order to hear stereo. Placing the subs directly to the sides of the listening position puts them 180° apart, which is as wide apart as it gets, while placing them forward or rearward of the listening position makes that angle smaller.
For those wanting to delve deeper into it, check out the following papers:
Stereo_Bass_AESpaper.pdf 911k .pdf file
Stereo-Bass_Griesinger.pdf 148k .pdf file
In both cases, first read the conclusion of the paper (will take 30 seconds), then go back to the begining. Makes it much easier to get a grip on the explanation.
Excellent papers, sanjay (i think thats your name?). Thanks a lot for posting. i really enjoyed reading them this morning.
Building on what you say above, here is what I took away:
1. Can we spatially locate low frequencies? Unambiguously YES. Pretty clear qualitative and quantitative evidence that stereo subs can be "detected" by the human ear. To quote the AES paper "binaural detection by humans in the octave 45-90 Hz is physiologically possible". I was unaware that the evidence is scientific and compelling, yet so few audiophiles are aware of this or bother with stereo subs?
2. So fine, we can detect stereo subs. But is it more pleasant or real sounding? Pretty clearly YES as well. Author quotes the AES Banff conference in 2004 and his research with 6 experienced audio and music professionals. Both music/movie ambience was inevitably and reproducibly viewed by experts and larger audiences to be "more life like, natural, integrated with high frequency components..." when played binaurally through stereo subs. This appears to be a contradiction of your summary that most people find it less pleasant and phasey sounding? Am I missing something?
3. The bigger impact of stereo subs is not spatial localization of individual low frequencies but, as you say, externalization and the sensation of a concert/opera hall like sound. Both papers unequivocally state this. For higher frequencies, BTW, this is exactly why I bought MBL speakers. Delighted to see/hear that a similar spatially filling sound can also be achieved/created for 45-90 Hz by stereo subs.
4. Where should Subs in stereo be placed? Answer: Sides of the room and ideally the main speakers should on the long wall in a rectangle room. The latter point is counter intuitive to what most experts and laymen believe, or do in practice i.e. most of us place our main speakers (and our subs) on the short wall. Interetingly, again, MBL has always recommended that where possible place main L/R speakers on the long wall, symmetrically. In practice, this is sometimes hard. A practical solution of placing two subs in the front corners (same wall as L/R mains, but wider than them) or 4 subs, one in each corner, slightly assymetrically appears to be a practical and still audibly pleasant compromise.
5. Is there sufficient low frquency, stereo signal recorded? Yes. Plenty of low frequency stereo signal below 90 Hz recorded in stereo is already available, which higher end processors should decode in stereo. This also is in contradiction to your statement above, sdurani?
6. LFE below ~40 Hz does not need to be in Stereo.
7. How should 2 Subs be connected? Left sub should get L + SL + C + LFE signal. Right sub should get R + SR + C + LFE signal
8. How should 4 subs be connected? The paper does not talk about this. I wish it did. I would hypothesize that the rear two subs should mirror the signal of the front two subs. Ie Left Front and Left Rear Subs should get the same signal and vice versa. This is because the critical signal from 45-90 Hz is not to allow pinpoint sound localization but an overall enveloping, externalized and concert like sound and not just a thumping that presents exactly the same sound pressure to both ears at the seating place.
9. Finally, do Stereo subs create other problems especially by not making the room response as flat as an equal number of mono subs would? Frankly, I could not get a clear resonse from the papers but I also did not read them as critically and slowly as I should have (or would have in my younger days!).
Sdurani - you state what my original hypothesis was i.e. Mono subs would make the room response flatter than an equal number of stereo subs - thereby presenting us with a tradeoff. Do we want a flatter response from mono subs or the spatial envelopment of stereo subs? I quote one line from the AES paper which says: "use of two (or even number of) subwoofers resulted in less pronounced resonance effects from room modes and supports "positional EQ" of the listening space." Unfortunately, he does not clearly state which provides better positional or connectional EQ: 2 subs in mono or two subs in stereo?
In summary, I would say that any high end AV system MUST:
1. Connect 2 or 4 (or even number of) subs. This is of course old news
2. A high end processor must be able to run the subs in Stereo, as described above, for the 45-90 Hz frequencies. The listener will be rewarded with a spatially richer, concert like sound that is more pleasant than mono LF sound.
3. try to connect your mains on the long wall of a rectangle room, and your 2 subs in middle of the short wall i.e. on either side of the listener. (BTW - very nice pictures in the AES paper). Square rooms suck and must be avoided at all costs. Sitting in the dead middle of a rectangle room must alsoi be avoided.
4. if executing point # 3 is difficult, pretty good (nearly equal) results can also be achieved by placing 2 subs in the front two corners or 4 in the 4 corners, slightly assymetrically.
5. There may be atradeoff involved in that stereo subs will present a flatter room response than a single sub but perhaps not as flat as an equal number of mono subs. (This point remains unclear in my mind). in any event, the spatial advantages of stereo subs trump whatever little room node flattening advantages mono subs present. This last point is unequivocal from these papers.
Thanks again for posting the papers. Loved them. Hope Ive summarized them accurately and hopfeully usefully for others