Stereo Bass using subwoofers

[Digital_Thor]

Boutique subs have room EQ such as ARC built in, but I'm not convinced it's worth paying the extra for. Give me "dumb" speakers and "dumb" TVs and I'm happy

My Denon, Chromecast, and PlayStation will be the brains of the operation.

True... you have to view it through individual glasses. I have a stereo system... and would never think of using a surround product. So I have to do my own processing to EQ my subs to fit my room with my given placement of speakers and subwoofers.

And I use subwoofers for everything. Music without subwoofers is really not good for me. I need that superb control and smooth extended deep bass to really enjoy music as well as movies. I see no difference in the movies of music, when it comes to high quality bass. And I have very rarely heard even remotely good bass from speakers without subwoofers or at least some kind of serious EQ.
Ones tried Magico Q7. They also need subwoofers - a lot. In my opinion, it is simply impossible to get excellent bass without EQ and subwoofers in a normal room.

 

[Chromatischism]

Agreed. I don't change my setup at all when switching between music and a movie. I guess one exception would be a movie that has very weak bass. I might bump up the sub levels for that.

I do think stereo-only systems are underserved when it comes to room correction solutions though. Normally you have to choose between a surround sound receiver with a whole bunch of extra amplifiers you'll never need, or a dedicated stereo device that has a diminutive feature list...

So I'll repeat here my obligatory wish that Sound United put out a desktop-sized stereo integrated class D amp with Audyssey.

 

[Digital_Thor]

I "simply" build my own 3 way amplifiers to match my mains and a hunk of a beast to drive the subs. Pre-amp is a digitally controlled 8 channel analog pre-amp with build in DSP, that both has individual correction for each channel and also each input. Very strong output(good SNR) and very quiet and super transparent.
And your correct, that many stereo products do not flash EQ features in any fashion like surround products.
I believe the Harmon Kardon HK 990 is an exception.... with it's room correction and dual bass management. Never heard it though - since I went fully active and did the correction manually instead.
Sound quality in any surround I've ever heard, could not match a good stereo product. I don't really know exactly why..... But I would only use a surround product for a really dedicated home cinema... never for stereo.

 

[GimeDsp]

Below around 80Hz you cannot localise the direction the sound is coming from, so stereo subs (one sent left and one sent right channel) are a tad pointless. So sum to mono. Im assuming your subs have left and right inputs? In which case send both channels to both subs.

2 subs are however very beneficial as you can use them to cancel room modes and provide a smother bass around the room. I would recommend properly crossing the subs over and filtering out what goes to the mains.

I have heard this my whole life and have never found it to be true. I am sure it might be true in some places BUT, I know from testing my floors resonate at 70 hz, walls a little higher, windows even higher. All of these influence LF response.

Non localization would only be the case if you had a low enough sound without harmonics or overtones and/or a super steep crossover slope on a sub that only does very low stuff. In the real world with kick drums, floor toms, and bass guitars that all have lower fundamentals but harmonics and overtones there is very prominent localization of the subs due to the higher frequency content/room interaction. The ear/mind knows how to locate the lower stuff even if it gets a hold of minor spatial cues.

Now, in a non treated room, with higher mass structures, and the normal mess of early and late reflections dominating the overtones and harmonics or LF content, then localizing may be harder.

The body knows a lot of things. Having a sub against the wall and a standard phase alignment/FR will sound one way. Now put that same sub right under you, and achieve the exact standard of phase/FR alignment and it will sound/feel different.

Low frequency content has many spatial cues from room integration and music content.

A simple and test is this.
1. set up symmetrical subs next to mains speakers.
2. Put a steep crossover on them at 80hz or even 50hz.

No put something like a drum set sound demo that isn't processed but raw
3. Play the subs in mono
4. Play the subs in stereo

In stereo you will feel the attack of the drum head, the location of the drum shell, and the very LF resonances will be in the proper place they were panned.

In mono only the attack of the head will be located right, the shell location will smear out in both directions, and the LF resonances will occupy base sound field I call LF fog, because it's there, heavy, but it's all blurry.

If you do a mono sub in asymmetrical location it will alter music too much.
Even a single mono sub in center will screw the energy profile so that floor toms sound like they are horizontal.

In pro audio settings like churches and events, in non treated rooms, and in treated studio set ups, it's been like this for me.

 

[Chrispy]

True... you have to view it through individual glasses. I have a stereo system... and would never think of using a surround product. So I have to do my own processing to EQ my subs to fit my room with my given placement of speakers and subwoofers.

And I use subwoofers for everything. Music without subwoofers is really not good for me. I need that superb control and smooth extended deep bass to really enjoy music as well as movies. I see no difference in the movies of music, when it comes to high quality bass. And I have very rarely heard even remotely good bass from speakers without subwoofers or at least some kind of serious EQ.
Ones tried Magico Q7. They also need subwoofers - a lot. In my opinion, it is simply impossible to get excellent bass without EQ and subwoofers in a normal room.

Stereo encompasses more than mere 2ch, tho. 2ch is just a limitation of technology at the time it expanded from mono with existing tech/playback stuff. Not ideal, not the greatest, just what we got.

 

[GimeDsp]

Bassiusness(lol) diary day 20.

We've ran out of food and the attitude has shifted
What started out as good hearted adventure to find the perfect bass has now turned ugly. the crew is mutinous and I don't know how much time I have left at the helm.
We started with 1, then 2 subs. I pushed for 3 and at 4 my crew began to turn. Coffee and snack sare the last thing holding them back and I am running low.

what we have learned so far is that pushing subs past the extension of the listening triangle will results in unnatural widening of low frequency location.

With dual subs close to mains we have found this:

2 subs in mono sounds like a rock concert
2 subs in stereo sounds like a live event with no PA

 

[Pdxwayne]

For my stereo setup, I am currently using a pair of subs in stereo. The subs have the ability to do room eq. So far listening to songs with lots of sub bass, I don't feel like I am missing anything. The body feels are great. My daughter still complaint from upstairs when I played certain songs. ; )

 

[DonH56]

I have gone back and forth over the years and settled on mono subs with fairly steep 24 dB/oct L-R crossovers. Just not enough stereo content to bother, especially in a room not all that large, and as said above the directional info comes above the sub frequencies so a steep roll-off takes care of that. I find a group of mono subs with placement and integration optimized provides much smoother in-room bass than a stereo pair (or quad with front and back pairs) with much less hassle trying to integrate them.

When I had one sub, years ago, I centered it between the main L/R speakers. When I went to two subs, much later, I ran in stereo for years though went back and forth a few times. When I went to four subs, I ran them several ways, but except for the evenness (or not) of the response couldn't really tell if they were stereo or not and so they are all mono now.

To each his own - Don

 

[Chromatischism]

We've ran out of food and the attitude has shifted
What started out as good hearted adventure to find the perfect bass has now turned ugly. the crew is mutinous and I don't know how much time I have left at the helm.
We started with 1, then 2 subs. I pushed for 3 and at 4 my crew began to turn. Coffee and snack sare the last thing holding them back and I am running low.

This is hilarious, but...

what we have learned so far is that pushing subs past the extension of the listening triangle will results in unnatural widening of low frequency location.

With dual subs close to mains we have found this:

2 subs in mono sounds like a rock concert
2 subs in stereo sounds like a live event with no PA

I'm not making sense of this. In my experience, and that of many others more knowledgeable than me (Welti, Toole, etc), a few important things. Mid-wall placement gives smoother bass response than if a sub is near a corner in most cases. And, we can not localize frequencies above about 80 Hz in a room, and our accuracy in pinpointing sounds in a room is fuzzy until maybe 150 Hz where our head/ears can begin to effectively detect level differences from side to side. It is not until 800 Hz or so when our ear spacing can start to be used to detect timing differences in order to locate the direction of higher-frequency sounds.

With that in mind, after lots of testing the best place for my front sub is always right in the middle of the front wall, and the system is crossed between 60-80 Hz. Everything works in unison in this arrangement and I haven't found better.

So I'm not sure what you mean by an "unnatural widening of low frequency location" if our anatomy is not able to detect such a thing?

 

[GimeDsp]

"And, we can not localize frequencies above about 80 Hz in a room, and our accuracy in pinpointing sounds in a room is fuzzy until maybe 150 Hz"

I am in the fortunate position to be trying out all the different set ups now.
I have a sound check sheet with everything for LF content and material to listen to and take repeatable notes.

My last test today just disproved this for both me and the person switching the subs between 1, 2, and 1+2

Reference is a raw drum recording with 3 toms and kick, and full kit.

With kick/toms playing ONLY through subs crossed over at 48hz per oct. at 40hz we were both easily able to tell the sub that was playing, every time 100%
we tested this both at the normal LP in a 60" triangle and far out to the other side of the room.

We also put music on and turned on the mains and that made it somewhat harder to localize but we still could reliably.

So to sum it up.

20-60hz is easily localized to which sub is playing, both at exact same level and frequency response. (frequency and crossover confirmed by Systune)

Music and it's layering/complexity may mask the ability to localize subs BUT this is program material dependent and would result in a statement that "music layering and complexity makes the localization of LF content harder"

This test was done with matched subs just to the inside of mains, if we put the subs in the corners or outside as REL recommends the localizing would be very much more pronounced.

I wouldn't say you can pinpoint LF sounds but definite locate them just as in real life.

I am also using drum tracks and older songs that aren't squashed and over processed.

"With that in mind, after lots of testing the best place for my front sub is always right in the middle of the front wall, and the system is crossed between 60-80 Hz. Everything works in unison in this arrangement and I haven't found better."

I think that is a great set up too. and for finished music is a good way to go.

 

[Chromatischism]

20-60hz is easily localized to which sub is playing, both at exact same level and frequency response. (frequency and crossover confirmed by Systune)

You aren't locating the source of 20-60 Hz sound in a room. It isn't possible for a human to do. Toole, Geddes, and many others are unanimous on this.

Instead, you are hearing other sounds that are correlated with those lower frequencies, and that is tipping you off to their direction.

With kick/toms playing ONLY through subs crossed over at 48hz per oct. at 40hz we were both easily able to tell the sub that was playing, every time 100%

Maybe it's the way you wrote this, but it sounds like you knew the subs were playing because you knew only the subs were playing.

 

[abdo123]

I have my subwoofers low passed in 180 Hz and i still don’t ‘hear’ them.

most mixes have the bass below 200Hz downmixed to mono.

 

[GimeDsp]

I am not sure what causes the localizing of LF but it's there. And the different sound fields for LF created by using single or dual subs and symmetrical vs asymmetrical is also pronounced.

As a studio tech and mix tech I don't see much use in white papers and theories. The people I help are friends with those and talk to them weekly, I the top acousticians/physicists disagree on a great deal of things. To me results matter, not which white paper is right.

The fact of the matter is that every sub set up has sounded different.
Below is a representation of the sound field of each tom drum.
Hi tom has nothing below 100hz.
2nd and floor tom do. which each sub set up the drums sound different, the sound matches exactly what the crossover and location behavior says they would.

with stereo subs the tom position is maintained and not skewed. the attack/shell/ and sustain all stay correct.

With a single mono sub in center, the floor tom and 2nd tom get pulled into the sub location, even at 50hz this localizing is pronounced.
Future studies will look at structure born sound and vibration as possible elements of localizing but for know here is a very rough sketch of the different LF sound fields.

Stereo Drums reference, 5 piece kit.

This sound field takes place with both raw drums (like one would deal with when mixing/mastering) and also other content like EDM music.

Now, you can say I'm wrong, doesn't matter, what matters is that each client hears it and that it helps them with their goals.

 

[abdo123]

I am not sure what causes the localizing of LF but it's there. And the different sound fields for LF created by using single or dual subs and symmetrical vs asymmetrical is also pronounced.

As a studio tech and mix tech I don't see much use in white papers and theories. The people I help are friends with those and talk to them weekly, I the top acousticians/physicists disagree on a great deal of things. To me results matter, not which white paper is right.

The fact of the matter is that every sub set up has sounded different.
Below is a representation of the sound field of each tom drum.
Hi tom has nothing below 100hz.
2nd and floor tom do. which each sub set up the drums sound different, the sound matches exactly what the crossover and location behavior says they would.

with stereo subs the tom position is maintained and not skewed. the attack/shell/ and sustain all stay correct.

With a single mono sub in center, the floor tom and 2nd tom get pulled into the sub location, even at 50hz this localizing is pronounced.
Future studies will look at structure born sound and vibration as possible elements of localizing but for know here is a very rough sketch of the different LF sound fields.

Stereo Drums reference, 5 piece kit.
View attachment 154286View attachment 154288View attachment 154289
This sound field takes place with both raw drums (like one would deal with when mixing/mastering) and also other content like EDM music.

Now, you can say I'm wrong, doesn't matter, what matters is that each client hears it and that it helps them with their goals.

Do you have crossovers set?
If yes what are the slopes of the crossover? how many subwoofers do you have?

 

[Chromatischism]

As a studio tech and mix tech I don't see much use in white papers and theories.

What I posted weren't "white papers and theories", but the results of controlled testing. Your examples, while interesting, are uncontrolled and therefore less reliable. Sorry, I don't mean any offense by that. Just using the correct terms.

The fact of the matter is that every sub set up has sounded different.
Below is a representation of the sound field of each tom drum.
Hi tom has nothing below 100hz.
2nd and floor tom do. which each sub set up the drums sound different, the sound matches exactly what the crossover and location behavior says they would.

Blind testing found that people are not able to locate the source direction of sounds below about 80 Hz in a room. It's hard to say your results translate when you know the location of the sub and you know details about its setup.

Stereo Drums reference, 5 piece kit.

I'm not sure how to read this. Are you trying to say it's supposed to sound like the drums are around you? Unless you are seated less than one meter in front of the drum kit where the extent of the kit reaches 45 degrees to your left and right, it's not. In real life, a drummer is entirely in front of you, in the center of the music. Now, we can do all kinds of recording and mixing tricks to come up with neat sounds, and I enjoy that, but you can't tell me your expectations are depicting some semblance of accuracy to the real world.

With a single mono sub in center, the floor tom and 2nd tom get pulled into the sub location, even at 50hz this localizing is pronounced.

Since most recordings put those frequencies to mono to avoid phase interference between Left and Right speakers (and I would guess for a little more acoustic power), it is even less possible to hear left/right differences at those frequencies in a room.

But where the rubber meets the road is this. The drums you describe consist of a far broader spectrum of harmonics than their fundamental frequencies, and that spectrum is split between your subs and your speakers by the crossover. If the crossover is too high you will in fact hear the sub play some of those higher frequency sounds. If the crossover is low, you will also hear them – but coming entirely from the speakers. In either case, the higher-frequency content that accompanies these drum hits are what tip us off to their source direction.

A more accurate test of localizability would be with sine waves.

 

[GimeDsp]

Do you have crossovers set?
If yes what are the slopes of the crossover? how many subwoofers do you have?

Yes, for the test I wanted the least overlap so we used 48db per octave, I'll have to check tomorrow if it was on .707Q or 1.

Also after test we ran some pink in Systune to verify DSP was working right. We have 4 subwoofers but the tests I have talked about so far are with more basic symmetrical set ups.

1. single sub in center facing LP
2. Dual subs just to inside in both stereo and mono (as pictured in my profile pic)

I did subs in corners and also many other positions but the examples included fit most playback/mixing systems. Plus I wanted the stereo subs as close as possible to give the "non-localizing" the best chance to be the case.

Next up to try is ASC rec for subs behind LP, 4 subs at 25% positions

I think we have patched in for up to 5 subs on the Nion DSP now.

My favorite for punchy sound was the dual mono subs, this allowed all the LF content to feed into both subs creating a punchy and strong energy that was different than stereo sub set up. Stereo was more realistic and tracked drums sound just like they should.

 

[abdo123]

Yes, for the test I wanted the least overlap so we used 48db per octave, I'll have to check tomorrow if it was on .707Q or 1.

Also after test we ran some pink in Systune to verify DSP was working right. We have 4 subwoofers but the tests I have talked about so far are with more basic symmetrical set ups.

1. single sub in center facing LP
2. Dual subs just to inside in both stereo and mono (as pictured in my profile pic)

I did subs in corners and also many other positions but the examples included fit most playback/mixing systems. Plus I wanted the stereo subs as close as possible to give the "non-localizing" the best chance to be the case.

Next up to try is ASC rec for subs behind LP, 4 subs at 25% positions

I think we have patched in for up to 5 subs on the Nion DSP now.

My favorite for punchy sound was the dual mono subs, this allowed all the LF content to feed into both subs creating a punchy and strong energy that was different than stereo sub set up. Stereo was more realistic and tracked drums sound just like they should.

For me the best sub-bass reproduction i have heard was with the speakers making an equilateral triangle with one subwoofer on my right and one subwoofer on my left making a semi circle with the triangle.

this would be a great setup choice for your next experiment since you’re equal distance to all 4 sources and at the same time you don’t see the subwoofers while you’re listening.

What you will experience will be similar to a sphere of bass all around you. Rather than a pulsating subwoofer or a wall of sound coming from the two speakers.

I would be really interested to hear your thoughts on whether stereo and mono are distinguishable in this case, with some listening controls ofcourse.

 

[GimeDsp]

"Since most recordings put those frequencies to mono to avoid phase interference between Left and Right speakers (and I would guess for a little more acoustic power), it is even less possible to hear left/right differences at those frequencies in a room."

having LF in mono doesn't fix the issue. The sound will still be screwed

The sketches I posted show the sound stage as looking forward at the a drum set. and the slopes show how much energy and it's position as it emanates. Very hard concept to sketch up.

The problem with the tests summarized at your link is that in real world mixers, mastering engineers, and end users are hearing real sounds, so thats what need to be used, not just test signals.

• Tests where bass transient signals were used support 60Hz being the limit of localizability.
• Tests where 2nd order slope crossovers are used support 60Hz.
• Tests where 4th-8th order slope crossovers are used support 80Hz.
• The early research on woofer localization used bass transients, bass sine waves, and music for blind testing as well as for measurement. Conclusion: bass transients over 60Hz can be localized. For sine waves, it is higher.
• The AES Technical Council notes in the Multichannel surround sound systems and operations document AESTD1001.1.01-10 that in small rooms, bass transients under 80Hz cannot be localized to the subs when steep crossovers are used.

If you have the article for the actual test it would be interesting to see the weighting of test signal vs music.

BTW half the songs I have checked so far have a GOOD amount of non mono'd LF.

 

[GimeDsp]

For me the best sub-bass reproduction i have heard was with the speakers making an equilateral triangle with one subwoofer on my right and one subwoofer on my left making a semi circle with the triangle.

this would be a great setup choice for your next experiment since you’re equal distance to all 4 sources and at the same time you don’t see the subwoofers while you’re listening.

What you will experience will be similar to a sphere of bass all around you. Rather than a pulsating subwoofer or a wall of sound coming from the two speakers.

I would be really interested to hear your thoughts on whether stereo and mono are distinguishable in this case, with some listening controls ofcourse.

That sounds cool. make up a quick
if you could mark up the location you are talking about that be great.

 

[GimeDsp]

I cant' post the audio here but I took a screen grab of my process to test songs for LF content.

1. Take sound into audacity
2. Split to mono, center pan each
3. Invert one side, usually the main VOX goes away unless they have a stereo widener used or near-panned backups.

4. Low pass at 120, see what's left, low pass at 80, see what's left, low pass at 60 see what's left

After each low pass I left a section of full audio so I can hear the relative difference.
Low pass being used in audacity is 48db per octave. As you can see from the screengrab, 7th Nation Army has a ton of LF still left at 120 and also at 80, the LF still left at 60 would is audible but would be masked to much