Timeline for Limiting factor on sprite sizes
Current License: CC BY-SA 3.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 12, 2019 at 15:08 | answer | added | Arne Christian Rosenfeldt | timeline score: 3 | |
| Dec 31, 2018 at 6:56 | answer | added | Leevke Waller | timeline score: 0 | |
| Nov 23, 2017 at 15:13 | answer | added | user | timeline score: 5 | |
| Nov 23, 2017 at 6:23 | answer | added | hotpaw2 | timeline score: 7 | |
| Nov 19, 2017 at 13:54 | vote | accept | rwallace | ||
| Nov 19, 2017 at 0:03 | comment | added | lvd | " doesn't seem like the bandwidth requirement should depend on N or W." -- actually it does. For example, bad lines on C64 with the addition of sprite fetches could leave for the CPU just 3-4 clocks available out of the maximum 63 clocks per line. | |
| Nov 19, 2017 at 0:00 | comment | added | lvd | "So that's machines spanning six years of time, multiple iterations of Moore's law" -- this is simply not true since for example VIC-II and amiga OCS chipset were both using 5um technology. Home computers were not at the top notch technology during 80ties. | |
| Nov 18, 2017 at 23:44 | answer | added | lvd | timeline score: 8 | |
| Nov 18, 2017 at 22:41 | answer | added | Raffzahn | timeline score: 5 | |
| Nov 18, 2017 at 22:35 | review | Close votes | |||
| Nov 23, 2017 at 3:01 | |||||
| Nov 18, 2017 at 22:03 | comment | added | tofro | What was really limiting sprite size was the availability of sufficiently accessible and complex custom chips - The original Amiga Inc was driven into foldup (takeover) trying to develop them, Commodore could build the VIC in MOS, the others had to revert to what the market had to offer - And that was not much. Home computing was still a niche market for chip vendors in the 80ies. | |
| Nov 18, 2017 at 21:22 | comment | added | rwallace | @Tommy This is certainly a valid viewpoint, and if I were designing a computer, I wouldn't incorporate sprites... but for the designs that did commit to them? That having been said, might the Amiga have skimped on resources devoted to them, because the designers felt they were no longer really necessary? | |
| Nov 18, 2017 at 21:15 | comment | added | Tommy | Yet systems with no sprites whatsoever were always at least as successful as those with: (Spectrum and Amstrad) v Commodore, ST v Amiga, then the PC. Even in the '80s a whole bunch of substantial and influential titles were 3d: Elite, Mercenary, Stunt Car Racer, Hard Drivin', Star Wars, etc. So my answer, if further research doesn't defeat this observation is: the world quickly outgrew them. Even on the Amiga they're often viewed as an anachronism; Atari was designing them out from the 7800. | |
| Nov 18, 2017 at 20:35 | comment | added | dirkt | Guess: What's really limiting sprites per scanline is the time required to process the transparency - in the worst case, all sprites are involved. And while transistor count and processing speed went up over time, frame buffer size also went up, so there's less time to process a single pixel. But to really find out, you'd have to look at a reverse-engineered graphics chip. | |
| Nov 18, 2017 at 20:25 | comment | added | rwallace | @tofro Oh, it was very much needed. For example, on the C64, which had unusually wide sprites, it was still very common to use the double width bit despite that doing bad things to resolution, which indicates just how much wider sprites were needed. I would expect more height to be easy to get. (The Amiga had no height limit.) Number of colors is related to bytes per sprite per scan line. | |
| Nov 18, 2017 at 20:16 | comment | added | tofro | Trying to apply Moore's law on something that is not generally usable or even needed (there's only so many sprites you can reasonably place in 256x192) is probably wrong. Not taking sprite height and # of colors into account probably as well. | |
| Nov 18, 2017 at 19:35 | history | asked | rwallace | CC BY-SA 3.0 |