Oerg866, on 29 April 2012 - 05:53 PM, said:
Nice work! Now all I have to do is get an SCD :D
You might be happy to know that a thingie that I'm currently working on gets rid of the need to burn CDs =P
Ravenfreak, on 30 April 2012 - 12:40 AM, said:
While at the moment I have no interest in homebrew stuff (let alone Mega CD homebrew stuff) this is awesome stuff you got here.
I can't wait to see more stuff from you. ^_^
Heh, thanks. I'm rather unpredictable so yeah (maybe there'll be a PCM driver up soon? WHO KNOWS? FIND OUT NEXT EPISODE!)
Bgvanbur, on 30 April 2012 - 09:16 AM, said:
Looks nice. And it is great to see that someone is making use of my tools (and remember suggestions are always welcomed).
theocas, on 29 April 2012 - 03:50 PM, said:
I haven't written much code for the ASIC and PCM since my ASIC converter tool currently doesn't do much currently, and I need to really write a PCM driver (even if it's just streaming PCM from the disc or a tracked format) =P
Sounds like fun work. I haven't touched the ASIC yet but would be cool to see what things you can come up in this space.
I have also been planning on making a PCM driver for the game I am working on. The plan for my driver is to have the sound effects in SCD memory and stream it using from RAM which allows me to use CDDA for background music. But still working on gameplay for now. But the PCM seems simpler than I thought after I disassembled the Cinepak code.
Yeah, the way I originally designed my driver to work (key word designed - not a single line of code was written) was that it had a table of songs and their sector addresses, and a word offset into another table that points to the sample banks as I called them. Sample banks, as well as song data would be loaded to Program RAM, and the driver would operate on that. I had a few coordination flags to mainly do looping (which I envision to be a bitch and three quarters) panning, volume, and sample changing. Any byte from $0-$CF would be interpreted as a note, while anything above is a flag.
I should actually just get off my lazy arse and write the damn driver. Of course, there's the lack of samples. Plus I have no earthly idea how I'm going to time this properly. There's the level 3 interrupt, but that happens ~7.3 ms which might be a little too quickly, but it could give me accurate sample timing. Although I can probably circumvent this with a 'null note flag' or a tempo header. Probably I'll also write a converter to convert from XM to this. I should probably learn how to properly use an XM tracker, though =P
But thanks for all the feedback, guys. I appreciate it.