23-10-2016, 06:52 PM
I have done a little more and got all the lines that I need done but there is a fair way to go yet.
Now that the lines are done I have started sorting the miss timings on some of the lines. These appear to happen when a interrupt coincides with an instruction that changes the program counter. Each time this happens the interrupt comes back one cycle early making the line one pixel short.
To correct this every time it happens an extra pixel needs to be inserted after the interrupt. This will be in the area of video after the circumference of the circle but needs to be before the video changes. On the right side of the circle this will be in the grey area after the circle and before the first vertical white line.
As the circle arcs around this area will be in a different place horizontally.
As I have said earlier I am using the two upper bits of the table to indicate which lines and which side or sides needs correcting. A byte will be passes to each line or block of lines to indicate where horizontally the pixel should be placed.
One thing I don't understand is the TV I am viewing on, the 3 MHz gratings are not visible and the 2.5 MHZ are just about but if one pixel out of 1204 is missing (I would have thought that would work out at 7.5MHz), it can be clearly seen. I had assumed it would not be visible. An area of the picture is shown below.
Frank
Now that the lines are done I have started sorting the miss timings on some of the lines. These appear to happen when a interrupt coincides with an instruction that changes the program counter. Each time this happens the interrupt comes back one cycle early making the line one pixel short.
To correct this every time it happens an extra pixel needs to be inserted after the interrupt. This will be in the area of video after the circumference of the circle but needs to be before the video changes. On the right side of the circle this will be in the grey area after the circle and before the first vertical white line.
As the circle arcs around this area will be in a different place horizontally.
As I have said earlier I am using the two upper bits of the table to indicate which lines and which side or sides needs correcting. A byte will be passes to each line or block of lines to indicate where horizontally the pixel should be placed.
One thing I don't understand is the TV I am viewing on, the 3 MHz gratings are not visible and the 2.5 MHZ are just about but if one pixel out of 1204 is missing (I would have thought that would work out at 7.5MHz), it can be clearly seen. I had assumed it would not be visible. An area of the picture is shown below.
Frank
