TranceAddict Forums (www.tranceaddict.com/forums)
- Production Studio
-- Doubling/Layering to get PHAT sounds?
Pages (2): « 1 [2]

thought this was pretty relavent to this discussion:

quote:

Taken from Tweakz To understand EQ is to understand limits. It is in it's essence an understanding that allows the whole (the song) to sound bigger by making the elements (the tracks) sound smaller, more narrow, and dare I say it in the world of phattness, thinner. This is because the audio bandwidth is itself limited. There is only so much space you have and if you fill it up with frequencies that overlap and conflict, you will not be able to hear the music through the sound. Stated positively, if you carve out a distinct sonic space for your instruments, you will hear each more clearly. That is the basic point and essential understanding that goes into developing a mix. Always remember, it is not how things sound in isolation that matters, it's how well they sound in the mix.

Ok, either nobody reads my posts, or I'm not making myself clear.

So here is a sample:

Single bass sound, recorded from Oberheim Matrix-1000:

http://www.trancestate.net/audio/OBBassSingle.wav

The same sound, recorded twice, both panned center:

http://www.trancestate.net/audio/OBBassDoubledMono.wav

The same sound, recorded twice, one panned hard left, and the other hard right:

http://www.trancestate.net/audio/OB...ubledStereo.wav

The same sound recorded three times, one panned center, the second hard left and the third one hard right:

http://www.trancestate.net/audio/OBBassTripled.wav

Obviously the volume increases as well, but if I compensated for the volume differences, you'd still hear the difference. I'm just being lazy.


Now during the 10 years I've been making music, I haven't found a single technique, plugin or hardware effect that does the same thing. The sound is extremely wide, but still 100% mono compatible. The stereo image is stable, and doesn't fluctuate or pulsate even a little bit.

Obviously, you have to actually record the sound twice, or three times, to get several slightly different takes. (Or use several instances of the same synth) That's what this is all about.

If anyone knows of a effect that can do this as well, I'm going to get it in an instant.

If you use the exact same sound twice, panned left and right, and put delay to left or right channel (this is what stereo widening effects do), it just doesn't sound as good. The attacks and decays of the sounds are on different times, and if the delay is too short you get a flanger effect, phase problems, and mono incompatibility. Also if the effect modulates the delay or some other parameters, the sound is going to fluctuate between wide and fat, and narrow and thin, or the flanging frequency will change. Well, this is all good if that's what you're after. But if you only want to make the sound wider and thicker while keeping it perfectly stable, there is no better way than doubling to do it.

I agree with you Mikk. My remarks were concerning the fact that the same sound is copied and panned left and right. That just creates mono. But, if you record a sound twice or more times and layer it, that definitely enhances the harmonic signature. Personally, I think it's a great effect...

quote:

Originally posted by Krispy Kreme I guess I got the idea from reading one of alphazones interviews. They said in one of their songs they used 3 vanguards for the main lead.

My guess would be that they layered three different sounds.

quote:

Originally posted by Mikk Ok, either nobody reads my posts, or I'm not making myself clear. [snip] If you use the exact same sound twice, panned left and right, and put delay to left or right channel (this is what stereo widening effects do), it just doesn't sound as good. The attacks and decays of the sounds are on different times, and if the delay is too short you get a flanger effect, phase problems, and mono incompatibility. Also if the effect modulates the delay or some other parameters, the sound is going to fluctuate between wide and fat, and narrow and thin, or the flanging frequency will change. Well, this is all good if that's what you're after. But if you only want to make the sound wider and thicker while keeping it perfectly stable, there is no better way than doubling to do it.

It actually is the same thing as just copying the audio track and using a time delay on different versions. Yes, if the delay is too short then you get phase cancellation, and if it's too long you get a cheap flanger - that's why you don't make it too long or too short.

But what you're actually getting from recording the sound over and over is one of two things: (a) phase differences from the synth, which is likely to cause phase cancellation when downmixed to mono on a club system, or (b) slight variance in the MIDI timing which causes a slight delay in different takes of the recording.

I guarantee you that if you actually investigate this, you'll find one of those two scenarios (or possibly both). Essentially you're relying on randomness and timing inaccuracies to create "phatter" sounds which I suppose is fine if it works, but simply stated there are better ways to do this (see Effero's second post for example) and, as Derivative mentioned, you are killing your headroom. You'll never need this if you've got a good synth.

quote:

Originally posted by DigiNut But what you're actually getting from recording the sound over and over is one of two things: (a) phase differences from the synth, which is likely to cause phase cancellation when downmixed to mono on a club system, or (b) slight variance in the MIDI timing which causes a slight delay in different takes of the recording. I guarantee you that if you actually investigate this, you'll find one of those two scenarios (or possibly both). Essentially you're relying on randomness and timing inaccuracies to create "phatter" sounds which I suppose is fine if it works, but simply stated there are better ways to do this (see Effero's second post for example) and, as Derivative mentioned, you are killing your headroom. You'll never need this if you've got a good synth.

I took a closer look at the recorded audio and you are right. There are slight random timing errors that make the notes start time vary. Usually it's less than one full wave cycle.

But there is another thing. Because of the way the patch is programmed, or the unstability of the synth, the waveform is constantly changing. So even when the timing is perfect the waveform is different between the takes, creating the effect.

It's true that doing this is relying on randomness and luck that the waveforms aren't cancelling out (not very likely with the usual asymmetric waveforms), or in perfect sync amplifying the sound. It's strange though that I've been using this a lot, and it never sounds wrong. Maybe when I get my new MOTU midi interface this will change..

As long as it works, I'll use it whenever I need the feel to. It's just another technique to get the sound I want.

Whatever technique that sounds good, is good. Rock (and probably not only) has been using the technique for ever - play a riff twice (important to have different versions - not just copying the track) and panning each hard left/right. It makes for a nice wide and fat sound and you can stick some vocals straight bang in the middle where there's room.

I was thinkin alot about what Vizay said about 2 tracks panned left and right being the same as 1 louder 1 centered......and it seems true....i never even though of it that way....its very interesting......but then i thought about how there must still be a difference........because lets say you have 1 panned left with 1/4 delay and 1 right with 1/8.......your ears hear that difference...so thats obviously not the same as having 1 louder synth right down the center.........something to think about. rubbish in my head

quote:

Originally posted by Emperor because lets say you have 1 panned left with 1/4 delay and 1 right with 1/8.......your ears hear that difference...so thats obviously not the same as having 1 louder synth right down the center

Yes, that's correct - as we've mentioned, if the left and right channels are actually different (most typically by having a small time delay on one) then it will sound different from one centered.

That's what the stereo wideners do, though; changing the "stereo separation" on those just adjusts the time delay between the channels.

quote:

Originally posted by Four_On_Four-er As for panning the equal parts left and right to achieve "phatness"... all you have to do is apply some stereo widening effects to achieve the exact same effect without the duplication nonsense.

How a stereo enhancer works.

First it duplicates whatever sound it is playing, then it adds a short delay to one of the "channels" accordiingly to what u do with the knob. Most likely you would also use the phase offset, which on 0 is the same as the sound sorce, and at 127 is the exact opposite (phase wise).

So it's not exactly the same effect as duplicating the sound and panning it hard to left/right.

When doing rock/metal etc, you often duplicate the guitar track and panning one to each side of the sound image, to leave room for bass, synths, drums etc, without having to sit there for two weeks fine eqing everything.



Oh and yeah, Mikk is right. Its a very classic trick. Analogue synth+tape recorder=fun

any nice stereo widener plugins or hardware available?

quote:

Originally posted by Krispy Kreme any nice stereo widener plugins or hardware available?

loads.. i like he fl stereo enhancer.. but thats me..

You dont need a stereo enhancer plugin.

Go and get Voxengo Delay (freeware). Render whatever sound you want to widen. Put Voxengo delay in the first Mixer Effect slot thingy. Increase the delay of either the left or the right channel - you can do this in the plugin. Leave the opposite channel with 0 ms delay.

Up to a point, the longer the delay the greater the extent of the widening effect. If the delay gets too long, then your ear will start to hear the left and right channels as separate sounds. That means you have gone way too far.

You can use Voxengo delay for all kinds of things too - not just widening. Great little freebie.

I hope I didn't seem like an idiot insisting how good of a technique doubling is. It's just another way to get the sound you want. The more unstable the sound source is, the better doubling works. For vocals and acoustic instruments it's sometimes just what's needed.

Adding delay to either channel like Derivative said is often basically the same thing, at least in the case of my poor Midiman midi-interface with horrible midi timing. Much easier to do obviously and it probably morks even when doubling doesn't. I still like to have the two sounds slightly different, which is possible when using analog/va synths or patches designed to change over time. But use whatever technique is most suitable for the task at hand.

Powered by: vBulletin
Copyright © 2000-2021, Jelsoft Enterprises Ltd.