RT-900 Update

Hey everyone,

I've been deeply absorbed in the RT-900 firmware port over the past few days and have made a ton of progress. It's still far from being usable, but I’ve been focusing on getting the core functionality in place first.

That said, this one's proving to be quite the challenge. The RT-900 is very different from the 890. Oddly enough, the 890 actually has a much better and faster MCU—it's based on an M4, while the 900 is running an M0. On top of that, the Radtel source code, while occasionally helpful, is extremely poorly written.

One of the unwritten rules in embedded development is: "Always use the least common interface." That means avoiding reliance on built-in MCU functions like SPI or I2C, because they tend to complicate future porting efforts. Beyond that, chip-specific functions often aren’t that helpful anyway—typically, once you kick them off, you just end up sitting in a loop waiting for them to complete. At that point, you might as well just bitbang the whole thing. Sure, you can claw back some performance by writing more complex event-driven or interrupt-based routines, but that also eats up valuable flash space—which, is already at a premium.

Dual Tachyon clearly understood this tradeoff, which is why all of that functionality was bitbanged from the start, even though the MCU supported it natively. So right now, a lot of my work involves translating from those chip-specific implementations to more generic, raw GPIO manipulation.

At the moment, I’m feeling a bit burned out, so I’m taking a day to recharge. Once I’m back at it, I’ll shift gears briefly to address some updates and fixes for the TD-H3 and RT-890 firmware releases—then it’s back into the RT-900 adventure.

Thanks for sticking with me!

Cheers,
Marc