Earlier this month we saw the first sign of life of TwBurn's new CD-i development software running on CD-i Emulator. Meanwhile, CD-i member Stovent is busy developing his new FOSS (Free and Open -Source Software) CD-i Emulator 'CeDImu': My plans are for now to be able to boot a bios rom, and then to boot a
game. I'm not planning in emulating every single CD-i model that
existed. I started it because I wanted to try emulation, but I didn't
wanted to do another NES emulator, I wanted to be original. I want it to be simple to use (unlike MAME), but with a bunch of
settings for advanced users. And of course free and open source (unlike
CD-i Emulator). And it is designed to be embedded in other software if
necessary."
The recent article by cdifan about CD-i Emulation gave CD-i member Stovent new ideas on how to find the missing information he needs for CeDImu. For example, he was looking if the BIOS sometimes exits supervisor mode and goes in user mode for executing instructions during the boot. He noticed strange behaviour in the core, and the supervisor mode is one of it. "It returns to user mode just before starting the "sysgo" module, which is quite early in the boot process." - CD-i member cdifan is always the helping hand. He still had an issue that it loops, waiting for an interrupt. Which chip is supposed to send the interrupt that wakes it up? According to cdifan, adding OS-9 call tracing would be a logic next step. Stovent: "I have it nearly ready, it is counting but it's not sending interrupts yet. The timer interrupts occurs every 10ms. This is what the OS-9 manual says in 'F$Sleep' and this is what I'm seeing in the timer register, but now the processor is stuck in the video module, looping indefinitely. It checks the PA bit, but it is always 0 because it is in 'non-interlace mode', but if I hack it to always be '1', I got [the picture above] !"
[Thanks, Stovent, cdifan - Join the developments in our CD-i server (find the invitation here) ]