I don't know the physics behind it all, but you'll get frequency smearing at the 'split' points. I think it's because there is always at least a little bit of crossover in frequencies at that split point, and the delayed signal vs. the undelayed signal creates phasing. The more splits you havethe more you will introduce phase distortion into the final signal.

However, why not give'er and see how she goes? Worst case scenario is that you don't end up liking the sound of it. Best case, you've found a cool new approach!

Multi-band wideners aren't uncommon, and some people swear by the sound of exciters that use a similar technique. But there IS a certain amount of smear/distortion that's inevitable.

Greg

I did a similar technique on a track that was recorded live mono.

I think I only split it into 2 bands and delayed one side slightly from the other. so basically the lower freq were mono and the upper were psuedo delayed stereo.

The effect was actually quite convincing, the cymbals actually sounded like they were in a room recorded stereo, a little exagerated though.

As for using it on a bus mix (which is I assume what your doing) I would think less is more and of course anything recorded in proper stereo would be better left alone but who cares, if it sounds good do it.

On a track level it could be a useful artistic tool.

That makes sense, I wonder if it dithers those points during the mixdown? Here's how I did it, it seems to sound OK but I haven't done any real double-blind listening tests or anything... 8-)

I can't hear the smearing. Anyhoo...

Nah, as usual I wasn't really paying attention-- I was thinking that you were progressively applying stereo widening to the frequencies, which is accomplished with a slight delay of frequency bands. The delays cause phase smearing at the crossover points.

Yours won't have the same phasing effects. There may indeed be some sort of artifacts, but not the kind I was thinking of.

Greg

Ahhh OK, I see. 8-)

I think I figured out the "problem" with this. At least from a musical standpoint anyhow. It arbitrarily redistributes portions of individual parts to new pan settings. So it actually destroys the stereo field by screwing up the original stereo positions of the parts. For example if we had a lead vocal dead center and a harmony vocal to the left (with it's reverb return panned behind it). When we apply this technique, the high end from both parts is panned, same for the low end, etc. So instead of being clearly "center and left" when we're done, both parts are now shifted across the stereo field according to their frequency data. Not good.

Applied subtly it may help some narrow mixes or a mix converted from mono, but it's certainly not good to do to a proper mix in hindsight. Oops. :-?

Kudos for the idea - that's nuts man. I'm just happy if my mixes sound better than mud. But I guess that's what separates the pros/veterans from the hobbiests - you guys actually think of stuff like this.

:wink:

I'm just a beginner, I was just messing around with this for fun... 8-)

This may be a bit old school, but what does it sound like if you mono the mix. Does anybody in the music world check their mix in mono anymore?

Isn't it simpler that way?

I usually check mixes in mono as I go along.

Yeah, I was gonna say that doing that would kind be unkind to the mix artistically. I'm learning more and more that pan really is a very musical consideration. It's the chance where, as a producer you get to show your own sense of musical expression. It's taken me a long time to realise that and that it's not just for "the effect".
Having said that I have found that just as a graphic EQ across your whole mix helps to unify and give it all a feel, setting that EQ differently for left and right gives it some of that stereo tonal character you were going for. Obviously the differences between left and right should shouldn't be too great unless you're being creative and you're making it work with your mixing or the circumstances of the song. Of course, like most other considerations in the mix one setting probably won't be suitable for all parts of a song.

Adam

FWIW that's not exactly the same Giovanni... You're splitting the bands "either -- or" whereas in the technique above the split is "equal" across the spectrum in the case of mono, and as equal as the mix for split stereo. Also your settings (which is neat) split the data according to a curve whereas a frequency band splitter splits it in a linear manner I think...

Coffee, this works fairly well on mono converting to stereo. Try it out sometime, it only takes a minute and it really does open up the mix a bit without adding any delay. (unless you add delay to any of the split tracks) It's definitely not something I'd do to a proper stereo mix for sure. 8-)