Sprinter/mame
2
0
Fork 0
mirror of https://github.com/holub/mame synced 2025-10-04 16:34:53 +03:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity

Some HW notes

Browse Source
This commit is contained in:
Angelo Salese 2013-02-27 01:52:15 +00:00
parent 773fa86b57
commit 5d9e1c46a0
2 changed files with 43 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
hash/saturn.xml
View File

@ -1164,7 +1164,7 @@
Super Real Mahjong P.V P's Club (Japan) (Genteiban) [T-16503G] | ??
Super Real Mahjong PVI (Japan) [T-16507G] | (srmp6)
Super Real Mahjong PVI P's Club (Japan) (Genteiban) [T-16508G] | (srmp6cg)
Super Real Mahjong P7 (Japan) [T-16509G] | (ssmp7)
Super Real Mahjong P7 (Japan) [T-16509G] | (srmp7)
Super Real Mahjong P7 P's Club (Japan) (Genteiban) [T-16510G] | (srmp7cg)
Super Robot Taisen F (Japan) [T-20610G] | (srobotf)
Super Robot Taisen F Kanketsuhen (Japan) [T-20612G] | (srobotfk)

42
src/emu/machine/saturn.c
View File

@ -1,3 +1,45 @@
/**************************************************************************************
Sega Saturn (c) 1994 Sega
List of things that needs to be implemented:
- There's definitely an ack mechanism in SCU irqs. This is almost surely done via
the ISM register (i.e. going 0->1 to the given bit acks it).
- There might be a delay to exactly when SCU irqs happens. This is due to the basic
fact that SCU runs at 14-ish MHz, so it needs some time before actually firing the
irq.
- Vblank-Out actually happens at the last screen line, not at 0.
- VDP2 V counter has a similar roll-back as MD correspondent register:
vpos line 0 == 0x1ff (Vblank-Out happens here)
vpos line 1 == 0
...
vpos line 241 == 0xf0 (Vblank-In happens here)
vpos line 246 == 0xf5
vpos line 247 == 0x1ef (rolls back here)
vpos line 263 == 0x1ff again
- HBlank bit seems to follow a normal logic instead.
- Timer 0 doesn't work if the TENB bit isn't enabled (documentation is a bit fussy
over this).
- Timer 0 fires at the HBlank-In signal, not before.
- VDP2 H Counter actually counts x2 in non Hi-Res mode.
- Timer 1 is definitely annoying. Starts from H-Blank signal and starts counting from
that position.
H counter value 0x282 (642) -> timer 1 fires at setting 1
H counter value 0x284 (644) -> 2
H counter value 0x2a0 (672) -> 0x10
H counter value 0x2c0 (704) -> 0x20
H counter value 0x300 (768) -> 0x40
H counter value 0x340 (832) -> 0x60
H counter value 0x352 (850) -> 0x69
H counter value 0x000 (0) -> 0x6a, V counter goes +1 here (max range?)
H counter value 0x02c (44) -> 0x80
H counter value 0x0ec (236) -> 0xe0
H counter value 0x12c (300) -> 0x100
- Timer 1 seems to count backwards compared to Timer 0 from setting 0x6b onward.
**************************************************************************************/
#include "emu.h"
#include "includes/stv.h"
#include "machine/scudsp.h"