I can, (and nearly have) but I need to finish what I'm working on at the moment.
Story of my life
Thanks in advance Doug!
I'm still working on the I2C mouse/keyboard stuff. Things are going Ok, and I've hacked the GraphicsDemo to make things easier to test mouse input.
At some point though, I'd like to get a TV GUI program running that uses mouse and keyboard. The VGA demos are very nice, but I don't have the luxury of using VGA with one of my key hardware configurations.
Yeah, and I'm dog slow right now too. Day job is rocking, which is never a bad thing.
The scan-line buffer color capability in Potatotext is already there. The work left to go is to fit the mouse writes in to that complex timing. If they won't fit, that means another COG, but one that might be able to do other stuff too. Don't know.
One other thing to do is build the TV COG as PAL. Started on that, got my Smile handed to me, and intend to toss out a call for some help when it makes sense to continue that. I have NTSC timings and where the "slop" can be down cold. That's not true for PAL, and PAL costs more compute in the COG.
Ideally, that would be a combined TV COG, though I'm not sure whether or not that really makes sense.
@Toby, I have the choice of a TV DAC and SDCARD or a VGA DAC without SDCARD ... there are no pins to spare I could probably do a monochrome VGA if the SDCARD pins were different. I would have to reroute the pins and rewrite the h/v sync parts of the driver for monochrome VGA for compatability, and I'm not sure it's worth the effort.
When I was messing around, with dreams of SDRAMs, I pushed the VGA (forground colour, backgtound colour, H-syncs and V-syncs) up onto P24-P27 and had the SD card on P20-P23. This was with the KBD mplx'd onto the P28-P29 pair.
When I was messing around, with dreams of SDRAMs, I pushed the VGA (forground colour, backgtound colour, H-syncs and V-syncs) up onto P24-P27 and had the SD card on P20-P23. This was with the KBD mplx'd onto the P28-P29 pair.
I'm doing something similar except the H/V pins are routed for P16/17 now ... a few short cuts/jumps would fix that though.
Nick are building some boards Oct 1 for the Gadget Gangster Platform. You might like one. The minimum configuration has not been decided, but is likely to have 32MB SDRAM installed.
If I were to spend money on (any) forms of enjoyment, SHE would be just so disapproving ...
I will keep an eye on the boards as it is one of the things I thought of trying, but will never get the end result. I used to have much more spare time and even then it was probably a forlorn hope
Current project is working through potatotext 2, and doing some work with OBC on the Sprite / Tile Driver. TV portion needs cleanup. Rushed it!
The older potatotext was designed for a mouse, but it's very memory hungry, and not really something I feel makes sense to build on. This one is lean, and does color, but it's just two color, so a mouse cannot be a unique color without using another COG, just for the mouse to overlay. If that is worth it, it can be added. If not, the other way to do a mouse is to dedicate 4 characters to be used as a buffer, and draw the mouse in, using the two colors of those characters. That shouldn't take another COG, and can be added to the current TV driver, which is only 2 color.
IMHO, what needs to happen is a driver that does 4 color characters. I've more or less avoided it, because the Parallax tile drivers are 4 color, and the extra font size with a limited 256 character TV driver seemed inefficient to not see use. For something like the Chess game shown, my 8x8 style drivers are not a very good fit. Larger tiles and more tiles available make the most sense, and that's the Parallax driver. I suppose a mouse could be added to that, like the VGA driver did, at the cost of a COG, done overlay style. Would have to be a white mouse, because of how the video signals work, and the complexity of color overlay on TV. VGA has some real advantages there.
A mouse can be done on the existing drivers, if somebody wants to define a mouse pointer, and then combine it with whatever characters are underneath. That requires 4 characters, and a small SPIN routine to run during the screen blank. Again, that method would be limited to the colors allocated on screen, no independent color for the mouse.
I really can't look at that for a bit. Need to finish what I'm on, or it basically won't get finished. One final option for a mouse is to simply invert the screen colors where the mouse is positioned. No pointer needed for that. This was done on PC DOS computers to good effect. Worth thinking about.
jazzed
Posts: 11,803
Comments
But, I did do a "magic cursor" for my Chess program that has a version based off of TV_Text...
It is kinda strange that there's not much mouse support for TV mode...
Thanks Rayman. All I found was a VGA variant. Interesting mouse driver though.
Seems like TV should have at least one demo similar to the VGA high-res demo. Maybe Potatohead can be lured into making one.
Thanks in advance Doug!
I'm still working on the I2C mouse/keyboard stuff. Things are going Ok, and I've hacked the GraphicsDemo to make things easier to test mouse input.
At some point though, I'd like to get a TV GUI program running that uses mouse and keyboard. The VGA demos are very nice, but I don't have the luxury of using VGA with one of my key hardware configurations.
Cheers.
The scan-line buffer color capability in Potatotext is already there. The work left to go is to fit the mouse writes in to that complex timing. If they won't fit, that means another COG, but one that might be able to do other stuff too. Don't know.
One other thing to do is build the TV COG as PAL. Started on that, got my Smile handed to me, and intend to toss out a call for some help when it makes sense to continue that. I have NTSC timings and where the "slop" can be down cold. That's not true for PAL, and PAL costs more compute in the COG.
Ideally, that would be a combined TV COG, though I'm not sure whether or not that really makes sense.
@Toby, I have the choice of a TV DAC and SDCARD or a VGA DAC without SDCARD ... there are no pins to spare
Nick are building some boards Oct 1 for the Gadget Gangster Platform. You might like one. The minimum configuration has not been decided, but is likely to have 32MB SDRAM installed.
If I were to spend money on (any) forms of enjoyment, SHE would be just so disapproving ...
I will keep an eye on the boards as it is one of the things I thought of trying, but will never get the end result. I used to have much more spare time and even then it was probably a forlorn hope
Current project is working through potatotext 2, and doing some work with OBC on the Sprite / Tile Driver. TV portion needs cleanup. Rushed it!
The older potatotext was designed for a mouse, but it's very memory hungry, and not really something I feel makes sense to build on. This one is lean, and does color, but it's just two color, so a mouse cannot be a unique color without using another COG, just for the mouse to overlay. If that is worth it, it can be added. If not, the other way to do a mouse is to dedicate 4 characters to be used as a buffer, and draw the mouse in, using the two colors of those characters. That shouldn't take another COG, and can be added to the current TV driver, which is only 2 color.
IMHO, what needs to happen is a driver that does 4 color characters. I've more or less avoided it, because the Parallax tile drivers are 4 color, and the extra font size with a limited 256 character TV driver seemed inefficient to not see use. For something like the Chess game shown, my 8x8 style drivers are not a very good fit. Larger tiles and more tiles available make the most sense, and that's the Parallax driver. I suppose a mouse could be added to that, like the VGA driver did, at the cost of a COG, done overlay style. Would have to be a white mouse, because of how the video signals work, and the complexity of color overlay on TV. VGA has some real advantages there.
A mouse can be done on the existing drivers, if somebody wants to define a mouse pointer, and then combine it with whatever characters are underneath. That requires 4 characters, and a small SPIN routine to run during the screen blank. Again, that method would be limited to the colors allocated on screen, no independent color for the mouse.
I really can't look at that for a bit. Need to finish what I'm on, or it basically won't get finished. One final option for a mouse is to simply invert the screen colors where the mouse is positioned. No pointer needed for that. This was done on PC DOS computers to good effect. Worth thinking about.