diff options
Diffstat (limited to 'src/mame/video/m62.c')
-rw-r--r-- | src/mame/video/m62.c | 30 |
1 files changed, 15 insertions, 15 deletions
diff --git a/src/mame/video/m62.c b/src/mame/video/m62.c index bf073bd1521..38d686c1cac 100644 --- a/src/mame/video/m62.c +++ b/src/mame/video/m62.c @@ -56,31 +56,31 @@ static INT32 spelunkr_palbank; The schematics also exhibit one pulldown for the sprite color guns. Since only either the sprite or tile gun is active, i.e. the other is in tri-state, this pulldown resistor also applies to the tile color guns. - + Which of the sprite/tilemap color guns is in tri-state mode is determined - by a pal. There is no good dump of this pal. If both sprite and tilemap - should be active (i.e. not in tristate) at a time, this would due to the + by a pal. There is no good dump of this pal. If both sprite and tilemap + should be active (i.e. not in tristate) at a time, this would due to the pulldown resistor have slightly darker colors as a consequence. - - This can explain the spelunkr bug m62_0116u4gre. - + + This can explain the spelunkr bug m62_0116u4gre. + The kidnike bug kidniki0104u3gre even looks worse and may imho only explained - by subpixel effects, i.e. delays when switching the proms into tristate. - + by subpixel effects, i.e. delays when switching the proms into tristate. + Since there are no dumps of the "priority" pal, the above is - speculation. - + speculation. + Priority PAL - + The PAL is at location 4F on the "T" board. The PAL's input are 3 bits from tilemap 0, 3 bits from tilemap 1 and A11-A14: - CB0A => 7 20 GND + CB0A => 7 20 GND CB1A => 8? 14 ==> PROM CS CB2A => 9 P 19 ==> Tilemap select 0/1 / to connector - C14A => 12 A 11 <== Sprite priority out / to connector - C13A => 4 L 18? ==> Sprite priority in / to connector - C12A => 5 10 GND + C14A => 12 A 11 <== Sprite priority out / to connector + C13A => 4 L 18? ==> Sprite priority in / to connector + C12A => 5 10 GND C11A => 6 CB0D => 1 CB1D => 2 |