Holy drama, Batman!
I pre-ordered my Tempest. It was disappointing and frustrating when I got it, and I shelved it for about a year. Maybe more? It was around the time the TR-8 was announced that I dusted off my Tempest for real and learned how to use it. It's incredible, it's still my desert island synth.
Shipping an unfinished product is never a good idea. Not having the resources to bring it up to speed post-launch is - well, let's face it, that's the world of tech. When you're taking on a complex, nuanced, niche project, the best you can do is estimate what effort it's going to take - but it's an estimation, not a prophecy.
This provided fuel for one of the most insufferably entitled communities I've ever had the displeasure of engaging with. I'm out there recording music and playing shows with something that, if the forums at the time were to be believed, was crippled beyond use. You still see that language today, in this very thread!
John, good on you for wringing out that last set of bugs & features. It's a bummer that it took such a coordinated campaign to make it happen. And the community feedback in the years after launch helped improve the Tempest in ways that went beyond the features in the original manual, even while failing for so long to deliver on so many other promised features.
But I never let that get in the way of using the damn thing. It was literally the most expensive piece of musical equipment I'd ever bought - might still be, now that I think of it - and it helped me find my voice, musically. It's too bad that so many people had the trouble they did, I don't mean to discount that.
At the same, speaking as a software developer who's worked on incredibly complicated, niche applications, I can't imagine how burned out Pym was after years of returning to the same project, receiving a nonstop barrage of overwrought criticism the entire time. I'm astonished he didn't quit.
If Sequential were to follow the Tempest up, I'm guessing the OS would be a rewrite from the ground up. It would be running on entirely different chips, and given the space limitations and the rigidity around additional features, not to mention the cryptic sysex, I wouldn't be surprised to learn that the code for the Tempest was a steps away from Assembly language, if that. Even if you were able to more directly port code to updated firmware, reusable code isn't like lego blocks, it's like an organ transplant.
Personally, what I'd want in a Tempest successor would likely go against what would make a more successful machine: give me more! Per-voice comp & distortion. Global verb & delay with individual per-voice sends! Polyphonic MIDI sequence tracks for outboard gear! And yes, the ability to load samples! But no... most people can't manage the Tempest as-is.
It's too bad that upon revisiting this community, there's still so much dick swinging and posturing. "Some of us have been around here for aeons and are therefore pretty jaded and cynical." As I said at the start of this post, I've been here from the start of the Tempest saga - if you're that jaded and cynical, that's on you.