Skip to main content
Retrocomputing

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

3
  • Did this methodology ended up in something concrete or was it just left as a proof of concept? It would be interesting to see a working example of this rendering strategy, as it would help to appreciate the decisions that you describe. In any case, many thanks for the detailed explanations, I really enjoyed reading this. Commented Mar 21, 2018 at 20:56
  • 1
    @introspec unfortunately I made a really dumb foundational assumption that 8-bit coordinates would do because I'd just clip before projection. I had not thought through precision. And going back to convert all of those parts to 16-bit with 2d span clipping was too discouraging. I've been thinking about revisiting it, possibly with a refined system for walking down the screen that eliminates the vertical sort by just putting a global limit on polygons and tagging a bit field for start and stop y locations, but have yet actually to act. I'm the worst. Commented Mar 21, 2018 at 23:06
  • 1
    @tommy re PowerVR "a process called tile-based deferred rendering, which required the programmer to track and bucket all geometry in a scene before submission." The programmer was never required to do this. In Series 1 (PCX1&PXC2) the API/Driver (plus HW during rendering phase) and on Series 2 (eg Dreamcast and Neon250 on PC) and all later generations, this was all done by hardware. Also it was only PCX1 that didn't have bilinear filtering (but it did have linear filtering between MIP map levels) Commented Dec 14, 2019 at 8:25