Hi Guys,
First, Superb job on the Preen algorithms! I didn’t get it at first, but after I studied them for awhile I was able to replicate ALL of the DX7 algorithms, ALL of the DX11 4-Operator algorithms, ALL of the RefaceDX algorithms, and ALL of the possible 3-Op algorithms. Way to go Xavier! And it only took 27. Feedback was the only difference. Since feedback mostly results in a sawtooth waveform, it’s not hard to replace (or you can load in a custom feedback waveform). Mostly I use feedback for brass instrument patches, so once my PreenFM2 arrives I’ll figure out how to best replicate those patches.
I’ve worked a lot on the Yamaha FS1R, so until my PreenFM2 arrives I have a few ideas.
1. The Yamaha FS1R and the Montage/MODX series add in six custom waveforms and call the results 8-Operator FM-X. Since there are six user waveform slots in the Preen, we can create all of the FM-X waveforms, ALL1, ALL2, ODD1, ODD2, RES1, and RES2, and load them in. The new waveforms add a lot of capability into FM, which I’ve gone over in the FM-X section of the FM Programming Guide on my website. RES1 and RES2 are especially useful. You can download the guide here if you’re interested:
http://javelinart.com/fm-synth-programming.html2. The FS1R also has Formants, which are really not that complicated to create. The formant operators pass the ALL1 waveform through a set of bandpass filters. We need probably six adjustable very narrow bandpass filters that work in parallel. The center frequency, bandwidth, and amplitude need to be adjustable on each filter segment. Yes, it would reduce polyphony, but the formant patches I’ve created on the FS1R are generally solo voices and don’t require much polyphony, just one or two notes at a time. I see custom filters people have added into the Preen code in the Forum, including a triple formant filter, so we could create a larger adjustable set of formant filters, can’t we? I have a really nice female voice patch on the FS1R (Christina Aguilera’s lower voice register) as an example of what this would add to the Preen:
http://javelinart.com/fs1r-human-voices.htmlAdjustable formants add more than human voices too, you can replicate realistic wind instruments, do more realistic strings, and create crazy synth patches based on more resonant, expressive instrument models.
3. Human voices take a lot of Portamento, called Glide? on the PreenFM2, right? We can do that.
If we make these changes we can claim the PreenFM2 can do 6-Op FM-X, and formants like the FS1R.
Essentially, a baby FS1R! Marketing? Are you listening?
Adding formants also opens up possibilities similar to the Technics WSA1 that even the FS1R can’t do. On the Technics synth you pair a Driver waveform with a Resonator. Figuratively, you pick an instrument like a Trumpet (the Driver) and pass the sound through a Violin body (the Resonator). You can mix and match any pairing of Drivers to Resonators from the hard-coded selections. On the Preen we could create our own Drivers (custom waveforms or FM patches) and pair it with our own custom Resonators using the formant filter set. The FS1R can’t do the first part, it locks you into the ALL1 waveform when you use formants, but the Preen could. We could put together a set of resonant filters (just settings) for each instrument type for the PreenFM2 to mate with our DX7 FM patches and create a baby Technics WSA1 while we’re at it. Pretty cool, eh?
