File talk:Ntsc timing.png: Difference between revisions
From NESdev Wiki
Jump to navigationJump to search
(Off by one) |
(Skipped tick on odd frames with rendering enabled) |
||
Line 4: | Line 4: | ||
*One of the comments (at the lower-right) states that the skipped cycle is actually at y=261,x=340 (and circuit analysis confirms that this is when it happens); I'm not sure why it's shown the way it is on the diagram. --[[User:Quietust|Quietust]] ([[User talk:Quietust|talk]]) 05:23, 11 December 2013 (MST) | *One of the comments (at the lower-right) states that the skipped cycle is actually at y=261,x=340 (and circuit analysis confirms that this is when it happens); I'm not sure why it's shown the way it is on the diagram. --[[User:Quietust|Quietust]] ([[User talk:Quietust|talk]]) 05:23, 11 December 2013 (MST) | ||
**Might it have to do with the long-standing off-by-one error in cycle counts? --[[User:Tepples|Tepples]] ([[User talk:Tepples|talk]]) 06:28, 11 December 2013 (MST) | **Might it have to do with the long-standing off-by-one error in cycle counts? --[[User:Tepples|Tepples]] ([[User talk:Tepples|talk]]) 06:28, 11 December 2013 (MST) | ||
**I think I drew it like that to avoid giving the impression that the final NT fetch on the pre-render line is skipped during odd frames. What really happens is that the final cycle of the fetch moves to (0,0), replacing the idle tick. Can't think of a non-convoluted way to draw it. --[[User:Ulfalizer|Ulfalizer]] ([[User talk:Ulfalizer|talk]]) |
Revision as of 06:56, 12 December 2013
the first green ppu cycle 0 on the visible scanline 0 seems to be misleading. according to blargg's odd/even frame test, this skipped cycle should be arranged at the last ppu cycle of the pre-render scanline, and the idle cycle presented in the image should be a read (if odd + rendering enabled). --110.206.2.212 01:34, 10 December 2013 (MST)
- One of the comments (at the lower-right) states that the skipped cycle is actually at y=261,x=340 (and circuit analysis confirms that this is when it happens); I'm not sure why it's shown the way it is on the diagram. --Quietust (talk) 05:23, 11 December 2013 (MST)
- Might it have to do with the long-standing off-by-one error in cycle counts? --Tepples (talk) 06:28, 11 December 2013 (MST)
- I think I drew it like that to avoid giving the impression that the final NT fetch on the pre-render line is skipped during odd frames. What really happens is that the final cycle of the fetch moves to (0,0), replacing the idle tick. Can't think of a non-convoluted way to draw it. --Ulfalizer (talk)