3 ms·

Why I loved the Apple II (memory-mapped video)

• ggchappell 14y ago
  > That idea didn't make it to the Macintosh, though I wish it had. But they had something even cooler, Quickdraw. Those were the days. If by "that idea", you mean a memory-mapped display, then it certainly did make it to the original Macintosh -- as well as a number of subsequent models. The original Mac display was a little 9-inch CRT, 512 x 342 black-and-white-only pixels. One bit per pixel, so each scan line was 64 bytes. 342 lines made for a bit over 21K of display memory, out of 128K total. And that had to hold the OS & applications, too. (Those were the days....) OTOH, I believe these Macs had only a single memory-mapped display buffer, not the multiple buffers that the Apple II line had. And of course no Mac ever had a text display buffer. And, true, you weren't supposed to write to the display buffer yourself. If you were a Good Person, then you used QuickDraw and the Window Manager. But you could write to it; no memory protection back then. Another funky bit of info you didn't mention: the Apple II display buffers were not exactly 2-D arrays. The screen lines were not stored in order. I seem to remember that Woz mixed them up because it allowed for easier computation of the memory location of a character (?). (If this sounds implausible, remember that, back then, multiplication was hard.) And there were "holes" in the buffer that were not used for display. 1024 bytes for the text display buffer only had to hold 40 x 24 = 960 characters, leaving 64 bytes of free space scattered throughout the buffer. These extra locations were allowed to be used as scratch space by I/O devices: 8 bytes for each slot.