On Thu, 2007-03-22 at 12:50 +1000, ram wrote: Briefly, you've got a pair of speakers that can spin. They are driven by separate motors. At slow speed they run at about 25-30 rpm, at high speed they run at about 360-400rpm (around 0.3-0.5Hz and around 6-7Hz modulation rate). The two motors are unsynced and take different amounts of time to spin up. The treble horns are quite toppy with a bit of a peak at 2KHz and nothing much below 500-750Hz (I *think - that's about how it sounds to me). There are a pair of them mounted on the rotator, but one is a dummy and is only there for balance. It's quite light, so it changes speed quite quickly. The bass rotor is slightly different - it's got a wooden drum with a kind of curved ramp inside and a slot in the side. The speaker looks down onto this. As the drum rotates, the ramp directs the sound out through the slot in the baffle. It's really heavy, and takes a long time to spin up. To model this, you'd want a couple of bandpass filters to model the crossover network (or maybe a couple of EQs, if they weren't computationally expensive), and two sweepable delay lines. You may find that you want a sweepable lowpass filter on the outputs too. Split the signal into the horn and rotor channels and feed them into the delay lines. As you sweep the delay up and down the pitch will change. For best results you want a bit of clean signal mixed in so you get some chorusing going on. The lowpass filter is for modelling the loss of top as the horn or baffle is facing away from you - it should get a little quieter at this point too. If you think carefully about it, you will see how to tap the mix output off the delay in two places (which you might be able to change). This will let you get a stereo output from it. I found I got best results with the "mikes" 90 degrees apart and panned hard left and right. Want to take it even further? Well, now you're getting into the realms of physical artifacts. You could do a impulse convolution of a real Leslie cab to get the tone right. You could add in motor noise (in quiet passages, you'll hear it). Finally, if you've got a "mike distance" parameter, then when it's really close you'll need to add some filtered white noise to get the "whufff... whufff... whufff..." of the windage from the horns going round ;-) In my organ plugin, I didn't bother with the crossover or the output filter. I just made a simple chorus/vibrato effect, and arranged it to that when you tell it to change speed it ramps up and down (exaggeratedly) slowly. This is probably the most distinctive thing about a Leslie simulation, and the only thing you'll really hear in a mix. Of course, it won't help you get the tone right for processing other things through your Leslie plugin (like the opening piano parts of "Echoes" by Pink Floyd, or the backing vocals in Kenny Rogers' "Just Dropped In"). HTH, Gordon (that's me done typing for the day...)

ram escribe: Sinusoids are harmonicaly boring by definition. "Squaring" them using distortion adds funny harmonics. Cordially, Ismael -- Ismael Valladolid Torres m. +34679156321 La media hostia j. ivalladt(a)gmail.com http://lamediahostia.blogspot.com/