Insane Contraption
Re: Insane Contraption
Cool, I look forward to seeing how this turns out!
Re: Insane Contraption
Watching this silently!
One thing I discovered lately to help development of such things as yours is get yourself an Atmel-ICE and use that JTAG port!! I seriously can't understand why I left that extraordinary on-chip debugging facility away for all those years. Man, it just works. Step-debugging, being able to inspect all registers and memory, apply random breakpoints at runtime and so on right on the real chip is a real thrill...and saves so much time. Of course the JTAG pins had to be on the video port so no wonder I never used it. But I made myself a special "dev" PCB with a switch to toggle between JTAG and 4 video pins. After that, my life changed forever. 
Btw, although i'm not posting on the forums, it doesn't mean i'm doing nothing.Seems like it is a lot more difficult to get myself developing the emulator than soldering stuff together (now that I have the equipment for it).
One thing I discovered lately to help development of such things as yours is get yourself an Atmel-ICE and use that JTAG port!! I seriously can't understand why I left that extraordinary on-chip debugging facility away for all those years. Man, it just works. Step-debugging, being able to inspect all registers and memory, apply random breakpoints at runtime and so on right on the real chip is a real thrill...and saves so much time. Of course the JTAG pins had to be on the video port so no wonder I never used it. But I made myself a special "dev" PCB with a switch to toggle between JTAG and 4 video pins. After that, my life changed forever. 
Re: Insane Contraption
Thank you 
I don't feel that much bottlenecked by debugging capabilities myself, I could easily get a functional NTSC signal out of it (since it is by PWM on this system, it is easy to verify that it would be cycle accurate, really, it is actually surprisingly easy to just throw together some random video mode experiment).
Why I got so deadlocked on the emulator is that it would be needed anyway, and it would be a lot easier to iterate on stuff with it. Put together some code, compile, run, see it failing miserably, fix, rinse, repeat. I also tended to just instrument the emulator with random patches if I really needed to know something, in ways it is not necessarily possible with JTAG (such as various timing details), so it is very easy to experiment with code that way, both the emulator and the code being just a "make" away.
My current job unfortunately is severely draining me, seriously. That I have no energy to work on this (despite how much I would love to) is about the smallest of the problems. Then, I would also really like to pick up an NXP i.MX RT devboard to see whether I could trick that thing into being a superb quality display driver for these consoles (including Uzebox). It is fast, I could really envision it being capable to sample the digital 8 bit port producing the display data at a sufficiently high rate, scaling it for a panel.
I don't feel that much bottlenecked by debugging capabilities myself, I could easily get a functional NTSC signal out of it (since it is by PWM on this system, it is easy to verify that it would be cycle accurate, really, it is actually surprisingly easy to just throw together some random video mode experiment).
Why I got so deadlocked on the emulator is that it would be needed anyway, and it would be a lot easier to iterate on stuff with it. Put together some code, compile, run, see it failing miserably, fix, rinse, repeat. I also tended to just instrument the emulator with random patches if I really needed to know something, in ways it is not necessarily possible with JTAG (such as various timing details), so it is very easy to experiment with code that way, both the emulator and the code being just a "make" away.
My current job unfortunately is severely draining me, seriously. That I have no energy to work on this (despite how much I would love to) is about the smallest of the problems. Then, I would also really like to pick up an NXP i.MX RT devboard to see whether I could trick that thing into being a superb quality display driver for these consoles (including Uzebox). It is fast, I could really envision it being capable to sample the digital 8 bit port producing the display data at a sufficiently high rate, scaling it for a panel.
Who is online
Users browsing this forum: No registered users and 3 guests