DSP Robotics Support

Has anyone experimented with stretched tuning, like that usually applied to acoustic pianos to compensate for the slight inharmonicity of the upper octaves, or is standard tuning sort of baked into the FS cake? In my digging around about Yamaha's FM instruments, I found that one of the features added to the DX7II was the option of stretch tuning and microtuning. I know microtuning can be done by simply crossfading the MIDI Fr input with a fixed number, but can it be stretched?

Stretching the high notes is easy: just multiply the frequency at the MIDI2poly module output by a suitable stretching function. Something like (1 + c*f*f) where c is some constant. However, it will probably not sound good unless you also stretch the harmonics accordingly. That can be done but not with the usual oscs.

I implemented piano stretch tuning in my Global panel. If you look to my Quilcom SECTION you’ll see what I did.
It was based on published material and is an approximation. Positive stretch occurs above middle C and negative below (higher pitches are raised and lower ones are lowered).

I like it because if you play octaves there is a small beating effect which I think makes the sound more organic.

Of course Martin is right and on a real piano it’s more to do with the interaction of harmonics. Apparently a “perfectly” tuned piano, where the fundamentals are exactly equally tempered, sounds out of tune! I confess I didn’t really understand that aspect when I read about it.

Cheers

Spogg

Yes, I'm very curious to know what Yamaha did for the stretch tuning on the DX7II, did they try to imitate a stretch tuned piano by stretching the modulators only and leaving the carriers (fundamentals) alone? - or do I have that backwards?

I'll have a look at what you did in Section - maybe I can incorporate it the complex-ilated version of Skematic?