OMG I did not know this, no wonder they shot up in value a little while back!23 wrote: Formant sequencing on the FS did get fully resolved, however, it was third parties, and not Yamaha, that picked the ball up running. There's TONS of frames that hit up into a formant sequence, and ultimately, a lot of times it ultimately became as simple as sample something, have the sample anaylyzed, and then proceed to convert it into a formant sequence from there. It was this part of the process that Yamaha asssured to complete but never did.....ultimately, this part of the process did get completed.
In this regard, not only is a good amount of paramaters missing from the screenshots of an editor that were given here, but not a single screenshot in regard to the detailed workings of formant sequence creation were offered.
Again, if one were to fuse each unvoiced op with it's corresponding voiced op, thus making for an 8 OP FM piece, the amount of waveforms each OP is capable of heads way beyond the hundreds....in all honesty, I wouldn't shy away from saying thousands.
The closest thing I know to compare it to would be the Supernova's wave generation system. Where on the surface, you only had Pulse, Saw, and Dual Saw available, which was in and of itself fairly limiting. But then throw into account that Sine waves were actually there for example, but they required that the hardness paramater on a square wave be applied harshly. Pretty much, once not only the fact that each osc could independently serve as it's own master and slave osc, but factors like Sync Skew and Hardness were factored in as well, it quickly became evident that the amount of waves each Osc was capable of independently generating was actually FAR FAR FAR beyond the norm and not at all limited to the 3 waveform range it looked like on the surface.
This same type of outlook truly should be applied to the FS.
The trouble with the FS didn't come in capability. The thing was, and years later still is, pretty much unparalleled in regard to it's capabilities; HOWEVER, I'd almost go as far to say that it was not "A" programing nightmare, but that it was "THE" programming nightmare. Going from the Front Panel, fact of the matter was NOT every paramater was accessible (yes, even with all the menu diving) and at that, what was available was a nightmare to deal with. Further, the way things ultimately rolled out, there literally were two FULLY FEATURED editors that were developed for it. One for it's Fseq capabilities, and another for everything else. Both Editors, as I said, were fully featured, it wasn't as if one was a light handed editor....both were pretty darn indepth. Even if the two editors were to be combined, at the end of the day was you'd end up with even an editor that was so darn indepth/complex that it itself wouldn't exactly be immediately user friendly.
Ultimately, what did the FS1R result in? I really don't know what to compare it to as there simply hasn't really been anything commercially released that has it's abilities. I mean one really could squeeze some things out of it that would liken many to comparing the results to K2000/5000 additive like results, still other things screamed WaveStation Wave Sequence like results, still others more (Waldorf) Wavetable like results, other things threw out the notion of it being some weird sort of vocoder (I'm speaking in terms of result, as of course it had no external input), but then of course there was the more traditional FM like results that could be reached.
Realistic choir voices could be morphed in ways that screamed some sort of Additive process must be going on. Beats and rhythms could be rolled out that all remained in sync regardless of key played (unlike on the wavestation) screaming Wave sequence. The voice of some man clearly speaking into a hand held radio could be accomplished.....but is that a man? because how is that being morphed into a woman? Really all sorts of crazy and quite odd things. Number of things that would have one swearing samples were having to enter into the fray somewhere along the line (like I said, i.e. some guy talking). But of course analog type basses and leads, and crystalline bells also quite possible to accomplish.
TONS of capabilbilty, but like I said, all lying behind what may quite possibly be looked at as "The" programming nightmare.
I'll hit up a demo of the FS some time. Trust me, getting it to go not just a couple steps, but FAR beyond the capabilities of other FM based synths can quickly be shown.
Virsyn are doing some similar stuff with formant and complex additive wave sequencing probably inspired by the FS1R. I'm not sure exactly how they compare tho and I would probably favour a dedicated softsynth rather than a 1u rack and software editor but I'm still very intrigued (must resist the GAS ).
One other thing in the FM7’s corner apart from the ease of programming is the ability to stack as many voices as your CPU will allow resulting in some very ‘BIG’ FM tones.