How can I give the impression of stereo from a mono song??

[bayswater]

1.- Messing with the track's M-S with a plug-in.
I just remembered we have a plug-in in DP that does just that (Spatial Maximizer). I'll try it, but, as noted by Stubbsonic, this won't work in this case because the M and S are the same (truly mono).
Is this the case? Can I still do something using this technique?

If it's not the case, I'm sorry, 'Nandito!

You have to modify the side in some way. But it's just easier to use any of the plugins mentioned, other than one that simply splits the signal into two channels and inverts one of them. Then you get silence on mono playback.

[monkey man]

Silly me. I thought most stereo-widening plugs used M-S techniques at their core.

Also, I thought Fernando was duplicating the track/s, meaning that he'd have had more than just a mono track to work with.

I thought I was covered, but my a$$ was bare?

[MIDI Life Crisis]

Another thing you can do, FM, is instruct the listeners to either:

1: Turn their heads really, really, fast;

2: Alternate covering their ears really, really, really fast.

[monkey man]

... or play back through two speakers, one of which is flipped upside-down.

[stubbsonic]

You have to modify the side in some way.

But to be clear, if the left & right channels are identical (i.e., it is a mono track), then if you were to run it into an M-S plugin, it would ALL be Mid, and there'd be nothing in the side to manipulate.

[bayswater]

My understanding:

Mid is the sum of Left and Right. Side is the difference between Left and Right. So if Left and Right are the same, Side=0, and you have a mono signal.

Going back to Stereo:

Left = Mid + Side
Right = Mid - Side

So if Side is zero changing the Mid / Side balance has no effect, and Left and Right will always be equal to Mid and each other.

But if Side is not zero, Left and Right will necessarily be different to the tune of 2 X Side.

The corollory: if Mid = 0, then Left and Right will be the inverse of each other, and if they are combined will zero out.

[stubbsonic]

My understanding:

Mid is the sum of Left and Right. Side is the difference between Left and Right. So if Left and Right are the same, Side=0, and you have a mono signal.

Going back to Stereo:

Left = Mid + Side
Right = Mid - Side

So if Side is zero changing the Mid / Side balance has no effect, and Left and Right will always be equal to Mid and each other.

But if Side is not zero, Left and Right will necessarily be different to the tune of 2 X Side.

The corollory: if Mid = 0, then Left and Right will be the inverse of each other, and if they are combined will zero out.

I'm not sure I follow your corollary. It sounds like you are suggesting that if the center is empty then the left and right channels must be identical and inverted phase. But clearly, I could have signals that are on the left side only, or on the right side only - - but not in the middle-- . Am I misunderstanding you?

[bayswater]

My understanding:

Mid is the sum of Left and Right. Side is the difference between Left and Right. So if Left and Right are the same, Side=0, and you have a mono signal.

Going back to Stereo:

Left = Mid + Side
Right = Mid - Side

So if Side is zero changing the Mid / Side balance has no effect, and Left and Right will always be equal to Mid and each other.

But if Side is not zero, Left and Right will necessarily be different to the tune of 2 X Side.

The corollory: if Mid = 0, then Left and Right will be the inverse of each other, and if they are combined will zero out.

I'm not sure I follow your corollary. It sounds like you are suggesting that if the center is empty then the left and right channels must be identical and inverted phase. But clearly, I could have signals that are on the left side only, or on the right side only - - but not in the middle-- . Am I misunderstanding you?

Yes, sorry, I didn't specify that completely. If all you have is Side, and Mid is zero, then the Left / Right would have been inverses of each other in the first place before conversion to Mid / Side. You would get a zero mid when they were added together. So to avoid the mysterious disappearing Moody Blues vocal track you'd have to create some non zero Mid.

If the Mid is truly empty, it means the sum of Left and Right is zero, so they must be inverses of each other (not identical). Like the old null test.

[stubbsonic]

Yes, sorry, I didn't specify that completely. If all you have is Side, and Mid is zero, then the Left / Right would have been inverses of each other in the first place before conversion to Mid / Side.

...

If the Mid is truly empty, it means the sum of Left and Right is zero, so they must be inverses of each other (not identical). Like the old null test.

Either this is incorrect, or maybe I am misunderstanding something fundamental.

The Mid channel only contains signal is common to both channels, so if the Mid channel is silent (at zero), that must means that there isn't any audio on the left or right channels that is common to both sides. But that wouldn't meant that everything on the left & right sides is simply phase-reversed.

But here's where my brain gets a little fuzzy. So if I have a signal on the left side only, then a signal on the right side only, how does the M-S de-matrix know to put those signals back to their correct sides? Time for me to visit the M-S laboratory and do some experiments.

[bayswater]

Maybe you're right. Mid is often stated as the common signal. Maybe I'm getting Mid / Side confused with Vertical / Horizontal on vinyl records.

But here's here I'm coming from:

The basic Mid-Side equations are:

Left = Mid + Side
Right = Mid − Side

If you rearrange these equations you get

Mid = (Left + RIght) / 2
Side = (Left - Right) / 2

That leads to the conclusion that Mid is the sum (or more strictly the average (sum-6db?)) of left and right, and Side is the difference between left and right.

That's not the same thing as Mid being common and Side being unique.

[mikehalloran]

Forger M/S as a solution to the OP. Bad idea and true M/S does not apply.

The Mid channel is a mono signal. The Side channel is recorded by either a figure 8 at a 90 degree angle or two cardioids back to back aimed 90 degrees from the source.

In a 2-mic setup, the figure 8 signal is split in 2. The signal is combined with the mono for the right; the other side has its polarity reversed and combined with the mid to create the left (or vise versa -- the MS plugin lets you don't both ways). So L=M-S and R=M+S or vise versa.

In a 3-mic M/S array, you record to three tracks. R=M+R; L=M+L.

The Sides reduce to mono combined with the M. Without the M, the Sides reduced to mono will null any common information.

A number of years ago, I recorded a chamber music concert in a church in San Francisco using an M/S array. I worked fairly close and made certain that no performer was dead center.

Shortly after intermission, it sounded as if all hell had broken loose. A local radio station had a massive fireworks show from a barge in the bay (the KFOG Kaboom!) and this church was only a few blocks away. Well, the performers kept singing and playing and I kept recording.

This was the very first time I used DP, btw. I load the recording in and start trying to figure out how the hell I am going to redraw all those wave forms to minimize the noise. Even then, I told the client it was going to cost big time and the recording was not going to be very good. Digital could do things that analog couldn't but we were talking miracles on the order of loaves and fishes here...

Anyway, in full stereo, it was just as bad as I feared with many of the explosions drowning out the soloists. Then I removed the M and panned both the sides to mono. The explosions, being external to the church, null'd out almost completely while the players, none of whom were centered, came out perfectly. The only thing missing was the stereo image -- small price to pay.

Anyway, I told the client that the second half was mono while the first was stereo. He liked that I came in as budgeted and considered me a miracle worker for having rescued his concert. I was damned lucky that I chose M/S as I always do in a good sounding room. It couldn't have worked out had I chosen any other way of tracking.

[bayswater]

So L=M-S and R=M+S or vise versa.

RIght.

In a 3-mic M/S array, you record to three tracks. R=M+R; L=M+L.

R=M+R ??? Not unless M=0.

[mikehalloran]

So L=M-S and R=M+S or vise versa.

RIght.

In a 3-mic M/S array, you record to three tracks. R=M+R; L=M+L.

R=M+R ??? Not unless M=0.

Of course one pans the M to 0, same as a 2-mic M/S array using a figure 8 and a decoder.

Three cardioid mics. One pointed to the source. That is your M. One pointed right; the other pointed left, back to back. The L/R mics should be identical. Pan M zero. R to hard right and L to hard left. This takes three channels and does not require an M/S decoder. In the 1970s, a decoder was an exotic piece of equipment that I couldn't afford. With DP, the decoder was built in but that was 20 years later.

R=M+R; L=M+L to get 2 channel stereo.

[bayswater]

Mike, the explanation is fine. The math is the problem. M=M+R. That cannot be. Subtract M from both sides of the equation, basic algebra, and you get R=0.

[mikehalloran]

Mike, the explanation is fine. The math is the problem. M=M+R. That cannot be. Subtract M from both sides of the equation, basic algebra, and you get R=0.

Uhhh... I never wrote M=M+R.

I haven't had my coffee yet either.