I disagree. It's true that Casio developed its own digital synthesis approach as it couldn't do FM. But this was true of Kawai (additive, hybrid single cycle, then PCM), Ensoniq (hybrid single cycle), Roland (single cycle + PCM), Korg (PCM), PPG (wavetables) etc. They all "worked around" the patent but none of these methods were anything like FM. PD in particular was designed to simulate low pass and resonant filters on traditional waves in a clever and cheap way. This really was a far cry from FM. It was inverse waveshaping with a window. It IS true that Casio eventually gave up and moved towards FM in later models (VZ).
I think we're saying the same thing. I wasn't suggesting any of these approaches offered some loophole that replicated FM, but that these manufacturers decided there's enough synthesis techniques that come close in sound that they opted to do so rather than pay Yamaha, as some other manufacturers did at the time.
Even if the underlying algorithms are nothing like FM if you get similar sounds (the gritty, busy waveforms) it made a lot of sense to market it and dilute the appeal of "FM" as THE digital synthesis method of the 1980s, as the DX7 nearly did.
I also wouldn't loop in wavetables with these approaches.
