Most clap VSTi’s I’ve seen have different claps on each note. What I do is the following which is Logical Editor centric:
- Create and edit a MIDI Part for one complete cycle or bar of your clap pattern - just a single clap right now.
- Edit the clap note’s velocity to put accents where needed.
- Duplicate the MIDI Part as needed for the entire song, then Glue all theses copies into a single big Part
- Edit this big Part and use the LE (or alt+drag) to duplicate your clap notes both above & below your starting line - use fairly large intervals like a 5th or octave when you copy the line. Do this a bunch of times to create your “crowd” size.
- At this point you’ll have a bunch of hands clapping your pattern but it will be very repetitive, always the same hand, velocity/accents, etc.
- Use the LE to randomize all the hand-clap note’s pitch value. Use the relative randomize so it moves the notes X amount from their current pitch. It’s best to make X equal to half the amount of the interval used when copying the line in step 4. That way you won’t end up with 2 notes landing on the same pitch at the same time.
- Next randomize the note positions using the LE, again using relative random to move them Y ticks before or after their current position. I recommend randomizing the position by a small amount multiple times rather than a larger amount just once. *
- Finally relative randomize all the velocities. The amount to randomize here will depend on what velocities the accented & unaccented notes used in step 2. Randomizing too much here can swamp out your accents.
I know this seems like a lot of steps but you can put a lot of it into Macros calling LE Presets in sequence. So once you do the setup work the actual execution is quick. Also a bunch of elements in this can be used on sounds other than claps.
*** Why randomize by multiple small steps instead of one big step**
If you look at a someone trying to hit a drum on a specific beat they of course don’t always hit it exactly at the intended time - instead they are off by small amounts (or for some players…). But the amount they are off is not a random amount. Usually they are very close to the beat, and sometimes they are a bit further, and every so often they’ll be even further off. If you made a graph of the distribution of this “offness” it would be a bell curve with more & more hits occurring the closer you get to the beat. If you randomize multiple times by a small amount the results will look like a bell curve where most of the time hits are close to the beat. If you randomize once by a large amount you’ll end up with the same amount (on average) of hits that are waaaayyy off as those close to the beat. This does not sound natural