I appreciate your obvious experience in this area, however synthesisers with more features, and more complicated user interfaces than the OP-1 have existed longer than I've been alive. I know this is not from a small company, but take for instance Yamaha's SY77 from 1989. It's built around a 16-bit Hitachi H8/500 processor, with 1MB of SRAM. It's CPU, synth engine, and peripherals are all glued together with 7400-series logic chips. Its AFM/AWM engine, sequencer, and the UI that controls them hold up well by today's standards. Compared to whatever ARM SoC the OP-1 is built around, the EE involved in designing the SY77 is much more complex. There is no comparison between the tooling available to Yamaha in the 80s, and the open-source tooling/libraries/resources available for embedded development in 2022. I understand economies of scale, however can't we expect more from a company like Teenage Engineering for the huge prices for their gear?
My TG77 is 33 years old. I replaced some aging capacitors because they were making the main output a bit noisy, however aside from that it's worked very well.
Comparing TE with Yamaha is a pretty massive difference in size of engineering teams and overall engineering knowledge. In the 80s the products, while definitely complex, had some advantages: back then you could get very simple and precise parts from many manufacturers because you had more small chip foundries available. seriously, try and find options for some of the bread and butter chips and you go crazy. The industry has consolidated most functionality into complex SoC parts which increase the bugs. The OP-1 is all run on an ADI Blackfin and even ADI was quite impressed by how much they got running. The graphics, the peripherals, the audio, everything. I was quite impressed!
You’d think that the tools got better but… honestly? Not really. The complexity of the chips has grown dramatically faster than the ability of the tools to simplify development. I can give tons and tons of examples of this in this particular industry, even with that exact chip. You pay for cheap chips in development and engineering hours
Battery or mobile products are no joke. Small products are no joke. Not only do you have to optimize around cost/performance but also around power considerations. Each major variable you add to development isn’t additive, it isn’t even multiplicative. It’s very difficult to compare apples to oranges here
Then you have overall price… the TG77 would have been roughly 2700$ adjusting for inflation (from 300,000 yen in 1989)
What comparable products do we have in that price range? Those are better comparisons
Just some thoughts. There’s no one right answer to this but if you think you can make a better product for the cost there might just be a business opportunity there!
> You’d think that the tools got better but… honestly? Not really...
Really? Aren't debugging interfaces light years ahead of where they were in the 80s? Yamaha's entire DX/SY line predate JTAG. There was no GDB back when Yamaha made the DX7. Parallel interfaces were probably theoretically simpler to debug than modern protocols like i2c and whatnot are today, I admit. And sure, the DX7's 16kb ROM is much smaller in scope than what is in the OP-1.
All these synthesisers perform the actual tone generation on proprietary LSI chips. You'll have a hard time convincing me that ADI's tooling is more difficult to use than whatever EDI tooling was available to Yamaha in 1982. The stakes aren't quite as high in modern software either. If they made an unforeseen design error in the LSI chip, that's game over.
> Battery or mobile products are no joke...
I don't doubt that. No doubt a battery requires a lot of complex design considerations. I'm sure balancing charge, size, and cost is difficult when it comes to batteries. How does this complexity compare to a synth like the SY77 using mains power? It steps that down to 5V, and 12V rails. I can't find the OP-1 schematics right now, I'm guessing it uses a single 5V rail from USB in.
> Then you have overall price...
This is true. I'm more talking about the complexity of its design. I picked the SY77 because it's closer in terms of the amount of features.
> There’s no one right answer to this but...
I don't doubt for a second that Teenage Engineering have some very bright people working there, and that they do great engineering.
My TG77 is 33 years old. I replaced some aging capacitors because they were making the main output a bit noisy, however aside from that it's worked very well.