|
Post by bjonte on Feb 19, 2017 5:45:11 GMT
Good work on the resolution limit! I looked around for CGA signal information but that was harder to find. Anyway, lots of cables are sold using only the monochrome output to connect to TVs so I think PAL and NTSC standard signals are still relevant.
|
|
|
Post by bjonte on Feb 18, 2017 9:54:23 GMT
I'm a few hours away from San Francisco, and I have an extra C128! When are you arriving, and how long are you staying? This is just too good to be true! I'm arriving next Friday, the 24th and leave the 3rd next month. I'm going on a business trip so I'll be busy most of the day Monday to Friday, but I'll not be working the weekend, so I could definitely meet up somewhere then. I'll send a PM.
|
|
|
Post by bjonte on Feb 18, 2017 5:49:10 GMT
Ah, I'm going to Europe in September, and I'm planning a side trip to Sweden. Perhaps I can bring you a NTSC C128 in my suitcase. That would have been amazing! I'm actually travelling to San Francisco next week so if you happen to know anyone in the area with a C128 I can bring it home in my suitcase.
|
|
|
Post by bjonte on Feb 17, 2017 11:26:52 GMT
Where in Sweden are you located? I live in Gothenburg and work in Stockholm.
|
|
|
Post by bjonte on Feb 16, 2017 12:08:41 GMT
I have found some info that explains the standard resolution values. According to www.batsocks.co.uk/readme/video_timing.htm, in a composite signal the blanking should begin 1.65 µs (PAL) before the hsync and be 12.05 µs long. With a 16 MHz pixel clock each pixel takes 0.0625 µs. So to get the left blanking character, take the hsync character position, subtract as many characters as needed to get roughly 1.65 µs less. The standard VDC values subtracted two 8 pixel characters, which is 1 µs. Then to get the right blanking character, add as many characters as needed to add 12.05 µs. The standard VDC values adds 25 8 pixel characters, which is 12.5 µs. For NTSC, blanking should begin 1.5 µs before hsync and be 10.7 µs long.
|
|
|
Post by bjonte on Feb 15, 2017 20:38:13 GMT
graphic mode or character mode? Character mode.
|
|
|
Post by bjonte on Feb 15, 2017 20:37:47 GMT
Do you mean that the frame of reference is the left side of the visible screen area?
|
|
|
Post by bjonte on Feb 15, 2017 16:34:53 GMT
I have been playing around with character height and it seems to be a physical lower limit to the height of characters that is depending on the number of columns. I haven't heard about this before.
Using 80 columns, the height cannot be shorter than 5 without graphical corruption on my C128. In interlace mode the smallest working character height is 8. Reducing the number of screen columns makes it possible to reduce the height a bit further without corruption.
|
|
|
Post by bjonte on Feb 15, 2017 15:17:32 GMT
I can't get the double pixel mode working. Mapping the C128 has a small BASIC program that is supposed to showcase it, but the monitor doesn't like it and will not sync.
10 WR=DEC("CDCC"):RR=DEC("CDDA") 20 SYS WR,63,0:SYS WR,40,1:SYS WR,55,2 30 SYS RR,,25:RREG A:SYS WR, (A OR 16),25 40 SYS WR,(8*16)+9,22 50 SYS WR,40,27 60 POKE 238,39
It looks wrong because it doesn't modify the horizontal sync width or blanking start and end. My own test that takes those into account as well doesn't work either. Has anyone of you managed to get the mode running?
|
|
|
Post by bjonte on Feb 15, 2017 15:10:09 GMT
has to be as big as the active screen area. You mean "inactive"? It only exists to the right of the display, sort of, when counting pixels from 0 to 1024/1016 at least. The kernal doesn't set the whole inactive area to the right of the screen either so it can't be right.
|
|