What I've noticed in many electronic productions (club, trance, electro) that the many of them sound amazingly wide/spacious. Even tiny details are audiable but the track itself does not loose the focus on the core parts (drums, bass etc).

I'm guessing that partly it is done by:
1. placing the instruments in stereo field in the right way
2. eq-ing parts to avoid frequency clashes
3. using stereo wideners on certain instruments (eg. pads)
4. using descent reverbs
5. offseting L and R channels by a few milliseconds on some parts.
6. autopanning or panning automation (leads, sequences or hats).
7. putting a dash of a reverb on the muster bus

But it seems that the above do notdo the trick. I'm getting a descent "space" but it is not even close to the commerical productions.

How do they do it? What am I missing?

For example in trance or electro tracks bass seems to be so wide but it still delivers in the mono area.I have tried many times to split bass sound into hi and low elements and widen the hi part. But never get the desired result.

I'm wondering if partially the reason for that is that I'm using small monitors. Have been using bigger ones for a few times and it gave a bit of improvement but again not that big I'm thinking of.

Any suggestions? Tips?

You're mentioning all the right techniques so you are on the right track. It's really just a matter of experimenting and using what you know in creative ways.

Some techniques:

Inter Aural Time Difference
You mentioned it in point 5. "offseting L and R channels by a few milliseconds on some parts". This can usually be done with any delay that allows you to specify in milliseconds who long to delay each channel. Use from between 1-10 milliseconds.

Splitting Frequencies
Basically just split the sound up in to the two channels and put a slightly different EQ on each channel. I think this is what you are talking about in point 2.

Stereo Widening
You can then enhance either of these two techniques by using a widener. I think there are a bunch of plugins to do this, but you don't need a plugin. Ableton has a device called "Utility". This allows you two either a) Mix a stereo sound down to mono, or b) remove the mono portion of the sound (leaving only the difference between the two channels). I like to create a return channel (bus I think in Logic), throw a utility on there and cut the mono signal (leaving only the stereo), then compress the @#$# out of the sound. In Ableton, I end up being able to send more or less of the sound to this return in order to create more or less width. You can even bounce the track and then use this technique for mastering if you are game.

I learned this stuff from Mr. Bill:

http://www.youtube.com/watch?v=wdBO_2EHS9c

I just have to say about #7: putting reverb on your master bus will do nothing for widening your mix, and will only make it muddy and have the opposite effect.

You need some good monitors, room treatment, a great set of ears and some 3d software.

Offsetting left and right channels has worked for me so far. I have noticed that delaying one of the channels by 11 msec works the best - more than that, you start noticing it - less than that, is still narrow.

I do the same on the bass. It doesn't weaken it at all. Usually I use a mid/side EQ to shave of the LF from the sides.

A good pair of monitors and audio interface are important too I guess, to achieve that kind of sound.

Thanks for your input guys. I forogot to mention that M/S processing is also what I've used. Anyway from your replies it seems that "channel offseting" is the most popular technique. I will try it on a bass tracks as suggested.

Big thanks go to Syncretia. Especially for the last tip (mono content cut). So simple that I can't believe that I haven't figured it out by myself

Once again thanks guys for your opinions. Top stuff

What's the difference between L/R delay and widening?

I got this article link in a thread about widening:
http://www.soundonsound.com/sos/nov10/articles/stereoprocess ing.htm

It's probably the Mid/Side widening stuff you are asking about.

As I understand it, stereo width comes from having a difference between the left and right channels. To make a part wider, you need to make its left and right signal differ from one another. You can do this with delay, EQ, distortion, compression, any kind of processing really.

I see. So channel delay offsetting is a kind of stereo widening, but so is detuning a channel, and other methods.

Yes, the difference being that when you delay one side of a stereo instrument, it seemingly moves backward in the stereo field. With different panning widths you can get some subtle placements that have more front to back depth than conventional panning. Just watch out when you get under 8ms, or comb filtering can be a problem in mono for anything both tracks share identically.

Adding a few db of upper mid parametric eq to the delayed side pops it forward again when parts hit notes the eq favors, which gives nice side to side animation when you pan it wide.

The longer the delay, the further "forward" the undelayed side comes. I wouldn't use it on whole mixes though. It would look fine on a scope and meters, but you'd perceive it as quieter on one side.

And always check any mid-side processing in mono too, some weird phasing can happen at extreme settings.

If I haven't made my point, check all your widening in mono!

One last trick is to set up a tight ambient reverb (half sec. or so) with true stereo input/output on a stereo aux send/return and pan the instrument to one side, and send it at unity gain to only the opposite side of the verb.
It works great to carve out space for rhythmic parts, and no phase trouble!

Using both, you can make things deep and wide, in subtle or extreme ways. Just remember to arrange so you can hear all this foo!

KVR/eSoundz: Xenobt

fwiw I mix in mono with only one speaker...if I get it sounding good there, listening in stereo just sounds that much better (after some tweaks)

padillac, are any of your tweaks to remove lower mids in mono? When I get my stereo eq where I like it, in mono I get a bump from 200-350 Hz, especially on things that do the mono summing for you like tvs. But then the stereo mix sounds harsher than I'd like after I scoop it out.

I know it's off topic, but perhaps related. It seems that instruments that share both channels like kick, bass and vocals are the main offenders.

Any thoughts, KVR? Averaging between the two eqs just seems so old-fashioned!

KVR/eSoundz: Xenobt

The answer:



But...

There is one more element. As mentioned earlier, you can actually effect the "sides" which is the difference between the two channels. I mentioned that I like to compress this signal so that it is flatter and louder. But you can also throw any effect you like on there (e.g. saturation, chorus, or anything else). This can create an interesting effect.

Also, another good thing is to make move between less and more width. It's one thing to just have width, it's another to make that move. One good trick is to LFO the Width. I created a MaxForLive LFO called Wofo:

http://www.maxforlive.com/library/device.php?id=891

I sometimes LFO the send level for the stereo widener. This gives the effect that the sound is moving closer and further away. If your DAW doesn't allow you to do things like this, you can just draw an automation envelope.

Yet another trick to create width on a mono signal is to simply split the channel in to left and right, and then apply different effects/EQ to either side. It's dead simple. Then you can apply more widening to that stereo signal.

I read about the difference between Left/Right and Mid/Side, and for some reason I still don't get it. How is that stereo? Side is the difference between left and right? So then how do you pan to one side or the other?

I saw this explanation on another site:



This seems to make sense, as far as file compression goes. I fail to see any other benefit, through. I guess I just don't get it.

I have a couple other questions about stereo sound: Does it make sense to put chorus on a send? Would you just put a 100% wet signal on a channel and send other instruments to it? You don't normally pan multiple channels to the same place in the stereo field, but does it make sense to give multiple channels the same amount of separation? If you pan the input channel, would it return in the right place on the stereo field? Or is chorus better on a per-bus basis?

Another question: I learned on another thread that most reverb before the 80s was either mono or pseudo stereo, and I wonder how that was set up. In my DAW, I can put two identical mono spring reverbs on two channels. Pan one 100% L and set the routing channel to 1 (or left). Pan the other channel 100% R and assign the routing channel to 2 (or right). I delay the second channel by a few milliseconds. I create a third channel with no fx on it but it sends 100% L to the first channel and 100% R to the second.

Now, I can take any channel, panned anywhere on the stereo field, send it to that third channel, and it will return a stereo sounding reverb, coming from the same place in the stereo field. In this way, two mono verbs can become a 'true' stereo send. This is how I've been stereo-fying reverb but I wonder if this is how they actually did it in the 60s and 70s.

I'm probably asking more about the 70s, though. A lot of 60s stereo effects were actually pretty hamfisted, and pretty easy to figure out (like, all the drums and their reverb were on one channel completely). It would be interesting to learn more about stereo imaging from a 'history of pop production' perspective.