[medium jim]

So if you keep the frequencies the subs are producing below, say, 80 Hz, you can get away with subs in mono. Why though? What's the advantage?

Try both ways and see what works the best for you. 

Jim

[andy_c]

So if you keep the frequencies the subs are producing below, say, 80 Hz, you can get away with subs in mono. Why though? What's the advantage?

That's a great question!  One aspect of the answer is the interaction of subwoofers with room modes.  If you look at the in-room frequency response of a single sub, it will have peaks and dips due to room modes that can be quite severe (say +/- 15 dB).  If you take another subwoofer supplied with the same signal but with a different location, the peaks and dips due to room modes will in general be at different frequencies.

If you were to try to EQ that first sub by itself, the dips would need to be either ignored or boosted.  If they are ignored, they will of course remain.  If the dips are boosted, this places some tough demands both on the subs and their amplifiers, as they are both being over-stressed in an attempt to "fill in" the suckouts.  Also, this attempt to fix the response can cause the response at other room positions to be worse than with no EQ.

Now suppose a second sub with the same signal applied is added in a different location.  Because of room modes, it will typically have reasonable output at the frequency for which the first sub has a suckout, so this sub can "fill in" the suckout of the first sub, at least partly, without having to send boatloads of power to the first sub.

It's kind of like the idea of modern portfolio theory from investing, in which many investments with different characteristics are combined to smooth out variations in the value of a portfolio with time compared to a single investment.  Only you have multiple subwoofers that interact with the room modes in different ways, such that they combine to have a flatter frequency response than any individual subwoofer.  Another beneficial side effect is that multiple subwoofers tend to make the bass response variation with listening position less than for a single sub.  There is actual data from Welti and Devantier and others showing this effect.

[poseidonsvoice]

That's a great question!  One aspect of the answer is the interaction of subwoofers with room modes.  If you look at the in-room frequency response of a single sub, it will have peaks and dips due to room modes that can be quite severe (say +/- 15 dB).  If you take another subwoofer supplied with the same signal but with a different location, the peaks and dips due to room modes will in general be at different frequencies.

If you were to try to EQ that first sub by itself, the dips would need to be either ignored or boosted.  If they are ignored, they will of course remain.  If the dips are boosted, this places some tough demands both on the subs and their amplifiers, as they are both being over-stressed in an attempt to "fill in" the suckouts.  Also, this attempt to fix the response can cause the response at other room positions to be worse than with no EQ.

Now suppose a second sub with the same signal applied is added in a different location.  Because of room modes, it will typically have reasonable output at the frequency for which the first sub has a suckout, so this sub can "fill in" the suckout of the first sub, at least partly, without having to send boatloads of power to the first sub.

It's kind of like the idea of modern portfolio theory from investing, in which many investments with different characteristics are combined to smooth out variations in the valuation of a portfolio with time compared to a single investment.  Only you have multiple subwoofers that interact with the room modes in different ways, such that they combine to have a flatter frequency response than any individual subwoofer.  Another beneficial side effect is that multiple subwoofers tend to make the bass response variation with listening position less than for a single sub.  There is actual data from Welti and Devantier and others showing this effect.

Andy!

Yes. I agree. Thank you for that.


  I was looking up various articles in my copy of Floyd Toole's book "Sound Reproduction" and happen to run across this one, Chapter 13.3.8, which is available from a google search regarding, ahem, stereo bass (sorry Jim!):

Stereo Bass: Little Ado about Even Less

I found it interesting that Dr. Toole said this:

Most of the bass in common program material is *highly correlated or monophonic* to begin with, and bass management systems are commonplace, but some have argued that it is necessary to preserve at least two-channel playback down to some very low frequency. It is *alleged* that this is necessary to deliver certain aspects of spatial effect. Experimental evidence thus far has not been encouraging to supporters of this notion (Welti, 2004).

But heh, what does Dr. Toole know  ?

Bye,
Anand.

P.S. I'm still going to call the studios though - to be thorough, I have a feeling that the devil is in the details.

[medium jim]

Anand:

You do realize what toole said corresponds to what I said?!  There are several recording techniques as Neekomax pointed out...each instrument typically is recorded on its own specific track and then the work begins to mix it down to the final 2 channel product, multi track in the case of surround sound recordings.

By all means call as many studio's as you can and faithfully report back...the devil is in the details as you noted.

Just to clarify, bass will be recorded onto its own track (usually) which technically is mono (symantics) and then during the mix down is placed into the playback soundstage, or stereo (more symantics). 

The key is the final product.


Here's a good primer on the recording mix down process:

http://en.wikipedia.org/wiki/Audio_mixing_(recorded_music)

Jim

[andy_c]

I think it's worthwhile to make a distinction between what instruments are recorded in stereo, and the spectrum of the resulting recorded sound.

Medium Jim is conflating instrument category with spectrum.  Suppose the bass guitar is panned hard left.  So what?  From the POV of mono subwoofer signal processing, what matters is the spectrum of the signal below the crossover frequency, which is a completely different thing.  If said spectrum below 80 Hz can be made mono with no perceptible change in localization (and there is no evidence I'm aware of that this is not so), then all the benefits of multiple subs can be realized without changing the artist's intent regarding localization - which relates to harmonics that appear in the stereo mains.

[neekomax]

So, do we know for sure that two or more subs run in mono cancel room modes to a greater extent than if run in stereo?

[JohnR]

So, do we know for sure that two or more subs run in mono cancel room modes to a greater extent than if run in stereo?

Well, I think the point is that you are optimizing for a different thing with subs than with mains. If you insist on "stereo" subs then a lot of the optimization opportunities are lost.

[Freo-1]

Andy!

Yes. I agree. Thank you for that.


  I was looking up various articles in my copy of Floyd Toole's book "Sound Reproduction" and happen to run across this one, Chapter 13.3.8, which is available from a google search regarding, ahem, stereo bass (sorry Jim!):

Stereo Bass: Little Ado about Even Less

I found it interesting that Dr. Toole said this:

Most of the bass in common program material is *highly correlated or monophonic* to begin with, and bass management systems are commonplace, but some have argued that it is necessary to preserve at least two-channel playback down to some very low frequency. It is *alleged* that this is necessary to deliver certain aspects of spatial effect. Experimental evidence thus far has not been encouraging to supporters of this notion (Welti, 2004).

But heh, what does Dr. Toole know  ?

Bye,
Anand.

P.S. I'm still going to call the studios though - to be thorough, I have a feeling that the devil is in the details.

Well played, my friend.  I think a single sub of high quality well placed can sound better than two subs of more modest quality.

[AJinFLA]

Jim postulated that if you drive dual subs from mono signal soundstage collapses but doesn't if you supply stereo signal to separate subs, left and right.
That is why it doesn't make sense to say its about distortion.

Ah, I misunderstood the context, my apologies. I was referring to localization cues not present in the input signal, but self generated by the transducer.

[AJinFLA]

Yes, it looks like Earl is concentrating on sealed boxes now.

Indeed. Upon glancing at his site, it appears even his mains are now sealed(?) as well. Earl has been paying attention.
Now I wonder how much Greisinger et al has crossed his radar...

cheers,

AJ

[AJinFLA]

Most of the bass in common program material is *highly correlated or monophonic* to begin with

I have a feeling that the devil is in the details.

Hi Anand,

Indeed it is. I highlighted Toole's caveat. If one is a typical audiophile listener, as I have encountered by the hundreds (thousands[?]) at shows, meets, events, demos, etc, etc. over the years, then read no further.
If not, http://www.davidgriesinger.com/asa05.pdf.
I've been hesitant to weigh in on localization and "stereo" bass, etc., because "it depends" and is so largely inapplicable.

cheers,

AJ

[AJinFLA]

I think a single sub of high quality well placed can sound better than two subs of more modest quality.

Absolutely. But statistically, the opposite is much more likely.

[poseidonsvoice]

Hi Anand,

Indeed it is. I highlighted Toole's caveat. If one is a typical audiophile listener, as I have encountered by the hundreds (thousands[?]) at shows, meets, events, demos, etc, etc. over the years, then read no further.
If not, http://www.davidgriesinger.com/asa05.pdf.
I've been hesitant to weigh in on localization and "stereo" bass, etc., because "it depends" and is so largely inapplicable.

cheers,

AJ

AJ,

Thanks for the link, a great read so far. Earl stopped making ported mains before 2007 fwiw.

Best,
Anand

[bdiament]

After a PM discussion with Jason and Anand on the subject of how bass is recorded and bass in recordings in general, I figured I'd post one of my replies:


In short, most pop recordings are collections of multiple monaural recordings, sometimes with dual-mono thrown in (such as when they might use two mics on a piano).

The arrangement of these sounds using pan pots does not create stereo in my opinion but something more accurately described as dual-mono... or, if I can invent a term, "panoramic mono".  We can localize a mono sound on the left side of a stereo pair of speakers because the pan pot makes it louder on that side (or exclusive to that side, if panned all the way).  But this should not be confused with real stereo -- of which, unfortunately, there are not really that many recordings.

All those faders on the console of a typical studio production carry mono signals.  Instruments are picked up by a single microphone or a single direct feed.  Often, a pop drum set will be picked up with several microphones but they are all mono signals, not correlated with each other and not arranged in a way that provides meaningful stereo cues to the listener.  So a drum set might use something like seven different *mono* signals, arranged left to right by the engineer using pan pots.

As to basses, in most pop recordings, they are picked up via direct feed from the pickups.  Sometimes an additional mono signal is derived from a microphone.

Pop mixers tend to pan bass instruments to the center.  The origin of this is not artistic but in fact is a matter of convenience:  It is *much* easier to cut a lacquer for vinyl record production if the strongest groove modulations are monaural.  Many vinyl cutters will, in fact, introduce a matrix which monos out the bass, sometimes from as high as ~140 Hz down. (!)  And then they'll filter out the bottom octave altogether.  It is not only easier but allows for more level to be packed onto the disc.

It takes a courageous (and skilled) vinyl cutter to leave the real bass intact and in stereo.  They need to forego the Loudness Wars and aim for sonic Quality instead of sonic quantity.  Nowadays, such are very, very few and very, very far between.

I now of almost no recordings where the above is not true, certainly in the pop or jazz fields and all to often, in the classical fields too. 

If I may say so, all of the Soundkeeper Recordings are made in real stereo, with the entire event picked up by a single pair of microphones arranged in a stereo array.  The mic array itself provides all the types of cues we use to localize sounds (not just the amplitude differences of pan pots but also timing differences and frequency differences).  No pan pots are used in the production.  I've used this same technique for classical music, jazz and rock.  Acoustic bass (unamplified) is picked up by the same stereo pair that "hears" everything else.  In the pop/rock recordings, the electric bass amp is picked up by the same stereo pair.  Again, no pan pots.  In fact, no electronic mix at all because the "mix" is achieved acoustically *before* the Record button is pressed.  Instead of pushing faders, we move the players (and amps) around the stage to achieve what we want. 

Hope this helps.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

[medium jim]

After a PM discussion with Jason and Anand on the subject of how bass is recorded and bass in recordings in general, I figured I'd post one of my replies:


If I may say so, all of the Soundkeeper Recordings are made in real stereo, with the entire event picked up by a single pair of microphones arranged in a stereo array.  The mic array itself provides all the types of cues we use to localize sounds (not just the amplitude differences of pan pots but also timing differences and frequency differences).  No pan pots are used in the production.  I've used this same technique for classical music, jazz and rock.  Acoustic bass (unamplified) is picked up by the same stereo pair that "hears" everything else.  In the pop/rock recordings, the electric bass amp is picked up by the same stereo pair.  Again, no pan pots.  In fact, no electronic mix at all because the "mix" is achieved acoustically *before* the Record button is pressed.  Instead of pushing faders, we move the players (and amps) around the stage to achieve what we want. 

Hope this helps.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

So on one hand you say it is too hard to do, but then tout your studio for recording in pure stereo, interesting....BTW, many studios mike as you suggest....go to the JackShit Video I posted as an example....I agree that some less well equipped studio's will short cut and pan, but these days most recordings are digital and transfer to CD format and not vinyl, with the recording/mixing techniques being different.   Loudness wars was and still is a reality, but that just means more compression and less dynamic rises within the music. 

Jim

[bdiament]

Hi Jim,

So on one hand you say it is too hard to do, but then tout your studio for recording in pure stereo, interesting....BTW, many studios mike as you suggest....go to the JackShit Video I posted as an example....I agree that some less well equipped studio's will short cut and pan, but these days most recordings are digital and transfer to CD format and not vinyl, with the recording/mixing techniques being different.   Loudness wars was and still is a reality, but that just means more compression and less dynamic rises within the music. 

Jim

I'm sorry if I wasn't clear but that isn't at all what I'm saying.

I never said "it is too hard to do".  I said "it is easier" for less skilled mastering engineers - or those seeking volume over fidelity - to simply mono out whatever bass they do not filter into oblivion.  (Note I was talking about *convenience* for the engineer, not what is possible for those with goals other than convenience.)  On the other hand, a skilled mastering engineer, like the late George Piros would *never* matrix the bottom of remove any of it, simply for the sake of making his job easier or to put more level on the disc.  George was into Quality, not sheer quantity.  If true fidelity is the goal, there is no reason why bass... in real *stereo*, unfiltered, cannot be cut onto vinyl or encoded in digital bits.

I don't know which studios you've worked in but in my experience, having done a lot of both over the years (recording, mixing, mastering in both analog and in digital), there is not the slightest need to record or mix differently for analog vs. digital.  While some engineers will definitely take a different approach, from my conversations with many colleagues, their intent is an attempt to "compensate" for flaws they perceive in each medium.  As I have never heard audible evidence to support the contention that such flaws can indeed be successfully compensated, while having heard a great deal of evidence to support the contention that they can't be, I've found I can get the best out of either format by recording and mixing with a single technique.

By the way, my studio is used only for post-production.  I never record in a studio because I much prefer to record in spaces where the music would best be heard by an interested listener.  Depending on the specific constellation of instruments and/or voices, I work in an auditorium, gallery or church.  I record *only* with a stereo array of no more than one microphone per playback channel.  (This does *not* preclude overdubs but I don't use them for Soundkeeper, which is always "live", in one take, with no post-production mix.)  The array provides all three types of cues humans use to localize sounds so there is no need for pan pots.  And the bass is never mono'd or filtered, as I prefer to hear what the player and instrument contribute to the music, unimpeded.  This approach is very different from how most records are made, as I explained in my initial post above.
 
I hope this clarifies my intent a bit.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

[medium jim]

Barry, thanks for the clarification...

Actually, analog tape has less usable bandwidth and the process of miking and such is different...it is best to move the players around on the set to get the best results.   Most in the day would push the mids a bit.   Agreed that the digital age has changed the rules as it was too easy to short cut and record directly off of the board and then mix to the final product.  Then enter home based software and things really got simple, but not better...the feel and dynamics were lost. 

Most major labels and many independent's take the time to do it right, although there is still those who practice packing as much as possible into the signal.  Not sure how your playback room is set up, but most that I've encountered play back two channel as it will be by the end user and not with multiple subs in cancellation mode.  The artist wants to hear it as they think the end user will be.  This doesn't mean that they pander to the masses and do want the best sounding product.  After all, it is their money on the line. 

As I have said over and over, ultimately it is the end listener who only has to please themselves and it is their choice as to what bass management works best for them.  However, knowing how records or CD's are made/recorded/mastered can't hurt on how they ultimately set up their systems.

Jim

[andy_c]

AJ,

Thanks for the link, a great read so far. Earl stopped making ported mains before 2007 fwiw.

Yes, +1 on that link AJ.  That will take a while to digest.

[medium jim]

The Griesinger paper was drafted about 12 years ago, it is a nice treatis, but accounts only for cone speakers to the best of my knowledge and not for planars, electrostats, and horn speakers which load and radiate within a room differently.

Jim

[TRADERXFAN]

Hi Anand,

Indeed it is. I highlighted Toole's caveat. If one is a typical audiophile listener, as I have encountered by the hundreds (thousands[?]) at shows, meets, events, demos, etc, etc. over the years, then read no further.
If not, http://www.davidgriesinger.com/asa05.pdf.
I've been hesitant to weigh in on localization and "stereo" bass, etc., because "it depends" and is so largely inapplicable.

cheers,

AJ

I know this is very old thread... but for various reasons I was revisiting it... and finally clicked that link. Thanks very much!!

I think this is really important extract:
"The statement that low frequencies cannot be localized is easily shown to be
true when a sine tone as used as a signal. However it is equally easy to show that an
broadband – such as low-pass filtered noise or a low-pass filtered click – is easily
localized in most rooms."