quote:

Originally posted by S-Tune One question: Isn´t it "dangerous" to mess up with phase when it comes to vinyl mastering or playing the song in a club in mono? cheers, S-Tune

Yes and no.

When mixed down to vinyl, all predominately bass instruments will be centred and mono. Because if you look at the surface of the vinyl you can see that low frequency sound is represented by a long slow oscillating swiggly line. Its slow enough that you can see the oscillations.

When you increase the stereo separation of any instrument you will cause destructive phasing to occur where parts of one channel in a stereo sound dip periodically into anti phase with the other. When summed to mono - you will get silence. On bass, where the lines are long and drawn out you can cause the needle to skip into the next groove.

But sending a track to vinyl requires a pretty rigourous mastering stage where they sort stuff like this out so it does not happen.

Ptjor G is right in the sense that if you want a stereo sound, you should record it that way to begin with - with 2 mic sources, either in a spaced or coincident pair. It depends on what kind of sound you want - Coincident pairs are for when you don't want the sound in the middle of 2 spaced mics to disappear (same destructive phasing). So you arrange them in the same spot, facing away from each other to avoid this. But spaced pairs can be useful too (for instance you could have a spaced pair on the overheads of a drumkit and mix more of that in after its gone digital so you pick up more of the natural reverberation of the room).

But if you are working totally digitally (as alot of people do these days), you have to artificially recreate these kinds properties of stereo recorded sounds. You do this with stereo separation, reverb, delay and so forth. All of these are attempts to simulate aspects of sound playing in a 3 dimensional space.

Most of the samples you will work with will either be synthesized or they will be close mic recorded. Nearly everything I work with at the moment is synthesized - whether by myself or by someone else who then rendered it and cut it into samples. And they sound it because theres no sense of space to a mix if all you do is drop down a couple of these samples from your library.

Changing phase offset can result in destructive phasing to occur and the extent of it depends on the other instruments in the track and the length of the offset. But really, its no worse than going too far with the other 2 dimensions of a digital mix - you can hype up the amplitude of bass instruments so much that it drowns out your kick drum and the fundamentals of your lead. That sounds pretty terrible too.

quote:

Originally posted by Derivative But sending a track to vinyl requires a pretty rigourous mastering stage where they sort stuff like this out so it does not happen.

It's not good practice to try to let a mastering engineer fix your crap in the final stage. Get it right at the beginning and it can be much more in the end.

Furthermore, you are not limited to doing what I suggested to only a live- (mic)-recording situation. Double your instruments, tweak them differently, whatever. It's more lively than a stereo widener, and will not sound like total crap when you hit the mono button.

___________________
All rhythm evolves around a kick...

What you are saying is exactly what a stereo widener does - that is, double up a stereo sound and adds a time delay to one of them. Or you can do this on a single stereo pair using a channel delay like Voxengo AudioDelay.

I never said it was good practice to let an engineer fix your mistakes. However, Vinyl mastering differs from CD mastering in the sense that the engineer will be conscious of phasey bass, precisely because it can skip needles.

Recording from a mic in a stereo pair is completely different from using channel delays to simulate stereo recording.

And channel delays do exactly the same thing as stereo wideners - they increase stereo width. A stereo widener is a channel delay, only with a few bells and whistles on - and most likely a clumsy gui.

If you have found some way in which this process does not cause destructive phasing where a stereo widener does then please - enlighten me because it totally goes against the physics of sound.

I am NOT suggesting channel delays.

I am suggesting making left and right slightly different. And by different I don't mean the same, but set apart 20 ms. For instance use different effects on them, e.g. different set reverbs, different set echo's, go wild. Or different base sounds alltogether.

___________________
All rhythm evolves around a kick...

quote:

Originally posted by Pjotr G I am NOT suggesting channel delays. I am suggesting making left and right slightly different. And by different I don't mean the same, but set apart 20 ms. For instance use different effects on them, e.g. different set reverbs, different set echo's, go wild. Or different base sounds alltogether.

I do this a lot. I create wider sounds by using different instruments at different panning values.

Ultimately, I think of phase as a subset of the panning, or rather the stereo dimension. They both deal with the track's position on the left to right spectrum. Panning just creates one position, while phase offset provides two.

___________________

www.jexmusic.com - My website

quote:

Originally posted by Pjotr G I am NOT suggesting channel delays. I am suggesting making left and right slightly different. And by different I don't mean the same, but set apart 20 ms. For instance use different effects on them, e.g. different set reverbs, different set echo's, go wild. Or different base sounds alltogether.

+ 1

This is a much more valid approach, even if its just as simple as EQing the two sides differently, or putting a different filter setting on the two sides of the bass. The idea of layering two different drum samples so they trigger at the same time is good too. We do this with guitar, use two different sounds to get more "balls" (with guitarists its always about "balls" lol) I think that layering diverse sounds may be a substantial part of the "pro sound" with or without panning. You look at most major producers tracks in cubase or whatever and the trackcount is like in the 60s. Thats coz they layer EVERYTHING. Its quite scary for those of us who are computer only, because we just think about our CPU. Those guys just put all those layers back off onto their hardware.

Just changing the delay, or using stereo widening or whatever is quite a lazy way of doing it, and it really doesnt save that much time. I guess it could be good for pads tho, if they dissapear in a club it might sound quite interesting.

I don't think you quite understand. What you are doing - cloning a stereo pair, then adding a time delay to one of those pairs is widening.

Are you people smoking crack or something?

Thats what widening is. Thats what a Stereo Widener does if it isn't a an independant L/R channel delay like Voxengo Audiodelay is. They are all derivatives of the same basic idea - taking 2 identical or similar stereo pairs or 2 identical or similar channels in a single stereo pair and adding a time delay to one of them.

Otherwise known as increasing phase offset.

You can process both pairs or both channels independantly but that has nothing to do with phase offset and nothing to do with widening. Thats something completely different.

Using stereo wideners is absolutely not lazy. Its a tool you use to change stereo width. Simple as that. You can do this any number of ways or reason that your method is 'better' but the effect is the same. You will get destructive phasing. The extent of it will vary massively depending on the extent of the offset but theres no avoiding it when theres a phase difference.

If using stereo wideners is being lazy then thats just like saying using a compressor is lazy because it dynamically changes amplitude so you don't have to do it manually. I mean, what the fuck are you talking about?

Last edited by Derivative on Dec-12-2006 at 14:13

No no you misubderstand me, I mean NOT using any delays on the channels. Like for example grab two completely seperate hi hat sounds and pan one hard left one hard right, set them to trigger together. It gives a "fatter" sound, its like layering pads or leads or something like that.
At any rate, what we're trying to achieve is a little bit more decorrelation between the two channels, stereo wideners/ delays are a way to do this, but they don't sound as fat and can often be just plain poor sounding. Not always, but a lot of people use them as a substitute for layering because they're quicker. For once, fatness and lazyness dont go together! (ok it wasn't that funny but its past midnight over here)

I am not smoking crack. I could quote the first line of my previous post but if that didn't get my point across, I doubt anything more I write will.

___________________
All rhythm evolves around a kick...

Thats because it doesn't make any sense. Making left and right slightly different? What does this mean? This is about as vague as you can get on the subject.

Do you mean panning 2 different instruments hard left and right? Because that has nothing to do with phase difference. That is just hard panning 2 different instruments left and right.

Hard panning the same instrument left and right and using no channel delay means nothing when summed to mono. You will simply get the same sound at a higher amplitude.

Hard panning the same instrument or a very similar instrument left and right and adding a delay of a few milliseconds to one channel in the pair is called widening. Every guitar rhythym track for like past 10 years near enough has exploited this trick.

Panning the same instrument left and right and tuning one of those channels out changes the period of that sound against the other. Thats called detuning and this can cause a flanging/phasing effect to occur as destructive phasing occurs periodically. When summed to mono, parts of this sound will disappear.

You can do any of these techniques with 2 stereo pairs if you want, provided you can output enough channels. It always has the same effect and it will always cause destructive phasing on parts of the sound. It usually isn't a problem but it can be sometimes. It depends on the extent of the phase difference.

A Stereo Widener plugin does this for you in rather the same way that a subtractive analogue VCO generates a simple saw saw wave instead of you having to create a saw waveform additively out of multiple sine waves.

And you can hardly call using a VA lazy when real men apparently use additive synths. Pffff.

quote:

Originally posted by Derivative Do you mean panning 2 different instruments hard left and right? Because that has nothing to do with phase difference. That is just hard panning 2 different instruments left and right.

Yes. And it can lead to great, wide sounds, without annoying phase cancellation artifacts when summed.

quote:

Originally posted by Derivative Hard panning the same instrument left and right and using no channel delay means nothing when summed to mono. You will simply get the same sound at a higher amplitude.

however, panning the same instrument left and right using no channel delay, BUT using other, independent, different processing (...NOT a delay), can lead to interesting, wide results. And while different EQ settings summed together (i.e. mono) can lead to minor phase issues, it is nothing compared to stereo widener issues.

By the way, there is something called the panning rule, or panning law, that automatically quiets down the signal when panned. The result of this is that the sum will be the same sound at the same amplitude, instead of your mentioned higher amplitude. I believe this rule is pretty much implemented in all usual DAW's and mixing consoles. This means mono-ing the mixdown won't fuck up your entire mix.

quote:

Originally posted by Derivative Hard panning the same instrument or a very similar instrument left and right and adding a delay of a few milliseconds to one channel in the pair is called widening. Every guitar rhythym track for like past 10 years near enough has exploited this trick.

No, they have exploited recording the same phrases twice, and hard panning them, without the delay. The fact that a human being doesn't play millisecond-tight, makes the sound come out the different sides at very slightly different timings. This is something your delay method tries to emulate, but effectively fails at, because of the machine-exact difference between left and right, instead of random differences.

quote:

Originally posted by Derivative Panning the same instrument left and right and tuning one of those channels out changes the period of that sound against the other. Thats called detuning and this can cause a flanging/phasing effect to occur as destructive phasing occurs periodically. When summed to mono, parts of this sound will disappear.

True. However I don't think the effects are as dramatic as the stereo widener mono-ed. A supersaw can still sound nice in mono, it's not a big signal-dropout-fest. The flanging effect you are referring to is otherwise known as "beating". Its pattern becomes more and more complex as differently detuned waveforms are added. Besides, it's most notable with very basic sounds, played dry. With effects such as reverb in the mix, the 0 points are obscured by the reverb noise.

quote:

Originally posted by Derivative You can do any of these techniques with 2 stereo pairs if you want, provided you can output enough channels. It always has the same effect and it will always cause destructive phasing on parts of the sound. It usually isn't a problem but it can be sometimes. It depends on the extent of the phase difference.

It could pose problems, and yes you should watch out. But as you said, it usually isn't a problem. Unlike, dare I repeat myself, the stereo widener.

quote:

Originally posted by Derivative A Stereo Widener plugin does this for you in rather the same way that a subtractive analogue VCO generates a simple saw saw wave instead of you having to create a saw waveform additively out of multiple sine waves. And you can hardly call using a VA lazy when real men apparently use additive synths. Pffff.

Real men drink beer. Va's rock. Additive synths rock. Analog synths rock. Use them all to your benefit.

___________________
All rhythm evolves around a kick...

Derivative, you're the one who first attributed stereo widening to what PjotrG is talking about. He never was. He's saying that it's better to combine two different sounds in different stereo locations than to simply widen the sound.

Ultimately, the choice is the effect one wants to achieve. If you want to make an enveloping sound that evolves atmospherically while managing the stereo element of mixing, then you should go for what PjotrG is suggesting. If you want a broader panning range, you go for stereo phase offset. If you just want a wide sound, detune.

The thing about what PjotrG is suggesting is that it creates a new instrument that is neither of the summed parts. And the effect is in fact made more dramatic if they come from the same place.

___________________

www.jexmusic.com - My website