reason unison modelling

Hi ya'll. I'm currently in the process of modelling some of the propellerheads reason devices in reaktor. The reason unison device splits one audio signal into 4, 8, or 16 voices and detunes each voice by a few cents (in case you didn't know). So I was wondering if there is any way to detune an audio signal after it is generated from it's source- whether that source is an oscillator or a sampler.

I always thought that the built-in reaktor unison was kind of gay. It uses too much processing power, there is no wet/dry adjustment, and I just don't like the results it produces. The unison on a Nordlead 3 sounds killer and the reason unison device isn't half bad, but I don't know what the deal is with reaktor's built in unison. That **** just doesnt fly. So I would just like to know if it's possible to detune an audio signal post-sourcel and how to do it. Any responses are greatly appreciated of course. Thank you.

The Deacon

 

the easy way would be to make a chorus-like macro, with 4 seperate delaylines modulated via lfo. you could also use a couple of pitch-shifting macros, like one of the new core ones, and detune them slightly (only a few cents in either direction). that may be a bit more cpu-intesive, even though the pitch shifter is in effect just a delayline, theres alot more going on underneath than a basic chorus effect. maybe someone else around here has a good macro/instrument for you to try out.

the standard reaktor unison mode shouldnt really be much differant from any other synth's. maybe some of the other synths also "detune" other parameters besides oscillator pitch, like maybe pan and lfo rates. this is easy to do in reaktor, since there is a unison spread module (its in the misc category). just stick that after some parameter controls that you want to spread and see what happens (if you send the synth's output to a stereo amplifier module, place the unison spread after the pan control, before the input, that should help fatten up the sound by spreading out the voices across the stereo field). also, if you have reaktor 5, check out the unison mode in subharmonic or carbon 2, which really get a great sounding unison chorus sound, along with pan spread.

 

The reaktor unison-mode uses an extra voice all the way through the signal chain - including things like VCA's and envelopes and what-not that aren't affected by the detuning of the pitch (though they might be if, for example, the decay time of a vca is set to be modulated by the pitch, but that's generally not the case).

For what you want to do, it's a little tough to get "realistic" sounding detuned unison mode without doubling (or tripling, etc) the signal chain up the point at which you want to recombine them into a "phat" unison mode sound. You could do a simple pitch-shift of the sound at any point in the signal chain and then recombine the right afterwards, but they may give you a lot of grainy artifacts.

I don't know much about Reason, but I'd guess the Nord simply detunes the oscillators and combines the sounds right after the oscillators - as opposed to having the filters and everything run in unison mode as well as the oscillators (but, then again, I don't know much about the Nord either). To try this, adding an extra oscilalltor or two with a slight pitch or frequency offset for the unison-sound doesn't take up too much CPU - see the 7-saw mode in oscillator #1 of my Koyaanisqatsi ensemble for an example.

The other option - pitch shifting by a small amount and combining the pitch-shifted sound could be done with the grain delay module with delay set to 0. This may sound a little odd, but might be musically useful so you might want to give it a try and see what you think. Honestly though, it may just sound like a digital chorus effect. (note that it will be more CPU intensive than a couple of extra oscillators, but you can use it anyplace in the signal chain you like). Still a chorus-sounding effect like this between oscialltors and the filter - or between the filter and some post-filter distortion - might just get the sound you're looking for.

- CList