jhs@MITRE-BEDFORD.ARPA (05/01/87)
I would appreciate a little tutorial from someone on the net who knows a lot about graphics, on the following subject: What is the "correct" way to translate from ordinary geometric dimensions into pixel counts, to get undistorted screen images of geometric figures which exist "in the computer's mind"? I am specifically interested in the highest-resolution mode (Graphics 8?) which displays 320 pixels horizontally by 192 vertically. I note that this is exactly in the ratio 5:3, and is somewhat less than the full screen size in both directions. I also know from expert broadcast television sources that the TV "raster" (the whole thing, which one normally does not see because it extends beyond the edges of the consumer TV screen) is in the ratio 4:3. I also know that the vertical scan is 262.5 lines high, so the 192-line active area the Atari uses is 192/262.5 or about 0.73 of the full raster height. I am not sure what fraction of the horizontal scan line is represented by the 320 pixels, or equivalently, how many pixels there would be from edge to edge of the raster if the same spacing were used. I would like to work with a monitor or TV which is set up correctly (4:3 for the whole raster) and get geometrically correct images from the Atari. Can I assume that one pixel spacing horizontally is the same distance as one pixel spacing vertically? If not, what correction factor should I use and how should I apply it? Any (factual) help will be greatly appreciated. -John Sangster / jhs@mitre-bedford.arpa