Naomi support status

[barbudreadmon]

Why fixing naomi support would be a good thing

I thought i would open a new issue about this, because the old ones don't cover everything.

First, for the remainder, there are quite a lot of games that were never ported from naomi to dreamcast or any other console, you can find a partial list there : http://www.thedreamcastjunkyard.co.uk/2016/01/guest-article-expanding-dreamcast.html, the list only covers until 2007, while there were other games later.

Second, yes, there is mame, but reicast would be far better for this, especially with @flyinghead recent work on OIT.

I wouldn't mind patreonizing this, but at the moment i'm kinda worried about the fact there are many forks with different features, all this good stuff should make its way here, and i don't know why it doesn't (mainly thinking about OIT here, to be honest i'm mainly using flyinghead's fork at the moment)

I was able to start the naomi mode a few days ago (it seems like no one know how it works, so i searched and tried until i found out).

How to start a naomi game

On linux, to start a naomi game, besides building with make NAOMI=1, you need the naomi_boot.bin file in the data directory, this file is known as epr-2157XY.ic27, where X is a number related to the region (6 for japan, 7 for usa, 8 for europe), and Y is the supported features. Depending on the game, it will accept certain bios, the bios with the best compatibility is supposed to be epr-21576h.ic27. Then you need a converted game rom + some kind of loader. The loader file is like a cue file with this kind of content :
Guilty Gear XX Accent Core
"GuiltyGearXXAccentCore_v6.bin", 0x0000000, 0x0F000000
Where GuiltyGearXXAccentCore_v6.bin is the path to the rom and 0x0F000000 is the hexadecimal size of the rom.
At the very least it's how things were with nullDC, personally i tried loading the rom file directly and didn't see any difference, so i'm not sure it's still working like this. But it should, right ?

Current issue

When i try launching a game that way, i get the naomi screen with an error saying "error 3 board malfunctioning", i don't know what it means, here is the output :

Config dir is: /home/udoo/.reicast/
Data dir is: /home/udoo/.reicast/
Personality: 00000000
Updated personality: 00000000
Linux paging: 00001000 00001000 00000FFF
MAP 00800000 w/ 50331648
MAP 20000000 w/ 50331648
MAP 04000000 w/ 33554432
MAP 06000000 w/ 33554432
MAP 0C000000 w/ 0
A8
vmem reserve: base: A94D0000, aram: c94d0000, vram: AD4D0000, ram: B54D0000
Resetting mem
Mem alloc successful!wtf /media/DATA/Emulation/roms/dreamcast/Akatsuki Bk Ausf Achse.lst is supposed to do ?
libwebsockets test server - (C) Copyright 2010-2013 Andy Green andy@warmcat.com - licensed under LGPL2.1
Initial logging level 7
Library version: 1.3 c11b847
IPV6 not compiled in
libev support not compiled in
Malformed entry on config - ignoring @ 49(ElfFile)
Malformed entry on config - ignoring @ 52(ta.HashCheckFile)
Malformed entry on config - ignoring @ 53(ta.HashLogFile)
static allocation: 4488 + (16 x 1024 fds) = 20872 bytes
canonical_hostname = udoo
per-conn mem: 184 + 1554 headers + protocol rx buf
Listening on port 5678
Found 6 matching FB configs.
Chosen visual ID = 0x21
Loaded /home/udoo/.reicast/data/naomi_boot.bin as bootrom
Loaded /home/udoo/.reicast/data/naomi_nvmem.bin as nvram
Using Recompiler
Auto-selected audio backend "alsa" (Advanced Linux Sound Architecture).
Initializing audio backend "alsa" (Advanced Linux Sound Architecture)...
ALSA: period size set to 1024
ALSA: buffer size set to 5120
recSh4 Init
Freeing fpcb
CodeCache addr: 0x564d5eb49000 | from: 0x564d5eb49000 | addr here: 0x564d5eb47f60
nullDC-Naomi rom loader v1.2
EEPROM file : null.eeprom
GL context version: 4.5
Sh4 Reset
reios: IP.bin is ''
reios: Hardware ID is:
reios: Maker ID is:
reios: Device info is:
reios: Area symbols is:
reios: Peripherals is:
reios: Product number is:
reios: Product version is:
reios: Release date is:
reios: Boot filename is:
reios: Software company is:
reios: Software name is:
reios: loading ip.bin from FAD: 8323072
reios: bootfile is ''
OSD: 6
Error opening /home/udoo/.reicast/data/buttons.png
Error opening /home/udoo/.reicast/pixmaps/font.png
Error opening /home/udoo/.reicast/font.png
Failed to locate bootfile.
evdev: Trying to open device at '/dev/input/event0'
evdev: Found 'Power Button' at '/dev/input/event0'
evdev: reading mapping file: 'controller_generic.cfg'
btn_a = BTN_SOUTH (304)
btn_b = BTN_EAST (305)
btn_c = BTN_C (306)
btn_d = BTN_THUMBL (317)
btn_x = BTN_NORTH (307)
btn_y = BTN_WEST (308)
btn_z = BTN_Z (309)
btn_start = BTN_START (315)
btn_escape = BTN_SELECT (314)
btn_dpad1_left = BTN_DPAD_UP (544)
btn_dpad1_right = BTN_DPAD_DOWN (545)
btn_dpad1_up = BTN_DPAD_LEFT (546)
btn_dpad1_down = BTN_DPAD_RIGHT (547)
btn_trigger_left = BTN_TL (310)
btn_trigger_right = BTN_TR (311)
axis_dpad1_x = ABS_HAT0X (16)
axis_dpad1_y = ABS_HAT0Y (17)
axis_dpad2_x = ABS_HAT1X (18)
axis_x = ABS_X (0)
axis_y = ABS_Y (1)
axis_trigger_left = ABS_Z (2)
axis_trigger_right = ABS_RZ (5)
evdev: Using 'Generic Controller' mapping
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev: Controller 2 disabled by config.
evdev: Controller 3 disabled by config.
evdev: Controller 4 disabled by config.
Legacy Joystick input disabled by config.
cntx // fpcb offset: -135266304 // pc offset: -184 // pc A0000000
Get MemPtr unsupported area : addr=0xA0000000
Get MemPtr unsupported area : addr=0xA0000018
Sh4: i-cache invalidation A0000016
Freeing fpcb
recSh4:Dynarec Cache clear at A0000016
Get MemPtr unsupported area : addr=0x80000080
sh4io: Const write ignored @@ 00000028 <- 0000A400
Get MemPtr unsupported area : addr=0x800000A8
Get MemPtr unsupported area : addr=0x800000BE
Get MemPtr unsupported area : addr=0x800000C6
Freeing fpcb
recSh4:Dynarec Cache clear at 8C0000DE
Get MemPtr unsupported area : addr=0xA0000420
Get MemPtr unsupported area : addr=0xA0000440
sh4io: Const write ignored @@ 00000028 <- 00000000
Get MemPtr unsupported area : addr=0xA00004BC
sh4io: Const write ignored @@ 00000028 <- 00000000
Get MemPtr unsupported area : addr=0xA0000506
Get MemPtr unsupported area : addr=0xA000057C
Get MemPtr unsupported area : addr=0xA0000586
Get MemPtr unsupported area : addr=0xA000058E
Get MemPtr unsupported area : addr=0xA00005A6
Get MemPtr unsupported area : addr=0xA00005AC
Get MemPtr unsupported area : addr=0xA00005B8
Get MemPtr unsupported area : addr=0xA00005CE
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? WriteMem: 5F7074 <= 0, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? ReadMem: 5F7074, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? WriteMem: 5F7074 <= 1, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? ReadMem: 5F7074, 4
naomi?WTF? ReadMem: 5F7034, 2
sh4io: Invalid read access @@ 00000030
naomi?WTF? WriteMem: 5F7030 <= 0, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 8, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4
naomi?WTF? WriteMem: 5F7030 <= 28, 4

My guess is that the issue is somehow related to the Failed to locate bootfile. message, which seems like some kind of "i can't load the rom" error, except i'm doing it the right way (at least, that's how it worked with nullDC), and my roms are ok.

Any help would be really appreciated.

---

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[barbudreadmon]

Some more informations.

I tried nullDC-naomi with the epr-21576h.ic27 bios, i got a black screen, so i guess it would make sense if this bios doesn't work in reicast-naomi either.

Then I tried nullDC with the epr-21576d.ic27 bios, this one allowed me to launch some games in nullDC, the same combination of bios/roms doesn't work in reicast, but give me a different error message : "Error 01 This game is not acceptable by main board". This same message appeared from C version to G version of the bios.
I also got fewer error in the output :

Config dir is: /home/udoo/.reicast/
Data dir is: /home/udoo/.reicast/
Personality: 00000000
Updated personality: 00000000
Linux paging: 00001000 00001000 00000FFF
MAP 00800000 w/ 50331648
MAP 20000000 w/ 50331648
MAP 04000000 w/ 33554432
MAP 06000000 w/ 33554432
MAP 0C000000 w/ 0
A8
vmem reserve: base: CDB70000, aram: edb70000, vram: D1B70000, ram: D9B70000
Resetting mem
Mem alloc successful!wtf /media/DATA/Emulation/roms/dreamcast/Guilty Gear XX Accent Core.lst is supposed to do ?
libwebsockets test server - (C) Copyright 2010-2013 Andy Green andy@warmcat.com - licensed under LGPL2.1
Initial logging level 7
Library version: 1.3 c11b847
IPV6 not compiled in
libev support not compiled in
static allocation: 4488 + (16 x 1024 fds) = 20872 bytes
Malformed entry on config - ignoring @ 49(ElfFile)
Malformed entry on config - ignoring @ 52(ta.HashCheckFile)
Malformed entry on config - ignoring @ 53(ta.HashLogFile)
canonical_hostname = udoo
per-conn mem: 184 + 1554 headers + protocol rx buf
Listening on port 5678
Found 6 matching FB configs.
Chosen visual ID = 0x21
Loaded /home/udoo/.reicast/data/naomi_boot.bin as bootrom
Loaded /home/udoo/.reicast/data/naomi_nvmem.bin as nvram
Using Recompiler
Auto-selected audio backend "alsa" (Advanced Linux Sound Architecture).
Initializing audio backend "alsa" (Advanced Linux Sound Architecture)...
ALSA: period size set to 1024
ALSA: buffer size set to 5120
recSh4 Init
Freeing fpcb
CodeCache addr: 0x55e6b4018000 | from: 0x55e6b4018000 | addr here: 0x55e6b4016f60
nullDC-Naomi rom loader v1.2
GL context version: 4.5
OSD: 6
Error opening /home/udoo/.reicast/data/buttons.png
Error opening /home/udoo/.reicast/pixmaps/font.png
Error opening /home/udoo/.reicast/font.png
EEPROM file : null.eeprom
Sh4 Reset
reios: IP.bin is ''
reios: Hardware ID is:
reios: Maker ID is:
reios: Device info is:
reios: Area symbols is:
reios: Peripherals is:
reios: Product number is:
reios: Product version is:
reios: Release date is:
reios: Boot filename is:
reios: Software company is:
reios: Software name is:
reios: loading ip.bin from FAD: 8323072
reios: bootfile is ''
Failed to locate bootfile.
evdev: Trying to open device at '/dev/input/event0'
evdev: Found 'Power Button' at '/dev/input/event0'
evdev: reading mapping file: 'controller_generic.cfg'
btn_a = BTN_SOUTH (304)
btn_b = BTN_EAST (305)
btn_c = BTN_C (306)
btn_d = BTN_THUMBL (317)
btn_x = BTN_NORTH (307)
btn_y = BTN_WEST (308)
btn_z = BTN_Z (309)
btn_start = BTN_START (315)
btn_escape = BTN_SELECT (314)
btn_dpad1_left = BTN_DPAD_UP (544)
btn_dpad1_right = BTN_DPAD_DOWN (545)
btn_dpad1_up = BTN_DPAD_LEFT (546)
btn_dpad1_down = BTN_DPAD_RIGHT (547)
btn_trigger_left = BTN_TL (310)
btn_trigger_right = BTN_TR (311)
axis_dpad1_x = ABS_HAT0X (16)
axis_dpad1_y = ABS_HAT0Y (17)
axis_dpad2_x = ABS_HAT1X (18)
axis_x = ABS_X (0)
axis_y = ABS_Y (1)
axis_trigger_left = ABS_Z (2)
axis_trigger_right = ABS_RZ (5)
evdev: Using 'Generic Controller' mapping
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev ioctl: Invalid argument
evdev: Controller 2 disabled by config.
evdev: Controller 3 disabled by config.
evdev: Controller 4 disabled by config.
Legacy Joystick input disabled by config.
cntx // fpcb offset: -135266304 // pc offset: -184 // pc A0000000
Get MemPtr unsupported area : addr=0xA0000000
Get MemPtr unsupported area : addr=0xA0000018
Sh4: i-cache invalidation A0000016
Freeing fpcb
recSh4:Dynarec Cache clear at A0000016
Get MemPtr unsupported area : addr=0x80000080
sh4io: Const write ignored @@ 00000028 <- 0000A400
Get MemPtr unsupported area : addr=0x800000A8
Get MemPtr unsupported area : addr=0x800000BE
Get MemPtr unsupported area : addr=0x800000C6
Freeing fpcb
recSh4:Dynarec Cache clear at 8C0000DE
Get MemPtr unsupported area : addr=0xA0000420
Get MemPtr unsupported area : addr=0xA0000440
sh4io: Const write ignored @@ 00000028 <- 00000000
Get MemPtr unsupported area : addr=0xA00004BC
sh4io: Const write ignored @@ 00000028 <- 00000000
Get MemPtr unsupported area : addr=0xA0000506
Get MemPtr unsupported area : addr=0xA000057C
Get MemPtr unsupported area : addr=0xA0000586
Get MemPtr unsupported area : addr=0xA000058E
Get MemPtr unsupported area : addr=0xA00005A6
Get MemPtr unsupported area : addr=0xA00005AC
Get MemPtr unsupported area : addr=0xA00005B8
Get MemPtr unsupported area : addr=0xA00005CE
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? WriteMem: 5F7074 <= 0, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? ReadMem: 5F7074, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? WriteMem: 5F7074 <= 1, 4
naomi?WTF? WriteMem: 5F7070 <= 7, 4
naomi?WTF? ReadMem: 5F7074, 4
DIV32S matched 1% @ 0xC033F1C
DIV32S matched 100% @ 0xC033F24
div32s 1/0/3
DIV32S matched 1% @ 0xC033FA6
DIV32S matched 1% @ 0xC033FB0
unknown MAPLE Frame
Unknown 0x86 : SubCommand 0x1 - State: Cmd 0x0 Mode : 0x0 Node : 0x0
Command : 0x86,Data : 4 bytes
00000001
new tactx
new tactx
DIV32S matched 1% @ 0xC033E66
DIV32S matched 100% @ 0xC033E6C
div32s 1/0/3
VREG = 00 ARMRST 01
VREG = 00 ARMRST 00
reicast-Naomi git/n - 14.16 (0.00) - 70.63 - V: 42.23 (85.00, VGA480p59.79) R: 0.50+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.24 - V: 59.94 (inf, VGA480p59.79) R: 0.00+0.00 VTX: 0.00 , MIPS: 0.00
Unknown 0x86 : SubCommand 0x13 - State: Cmd 0x14 Mode : 0x0 Node : 0x1
Command : 0x86,Data : 12 bytes
20070113 02210202 00000222
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
reicast-Naomi git/n - 10.10 (0.00) - 99.03 - V: 59.21 (1.49, VGA480p59.79) R: 39.31+0.50 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.16 - V: 59.89 (1.00, VGA480p59.79) R: 59.89+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.38 - V: 59.42 (1.00, VGA480p59.79) R: 59.42+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.16 - V: 59.89 (1.00, VGA480p59.79) R: 59.89+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.13 - V: 59.87 (1.08, VGA480p59.79) R: 55.38+0.00 VTX: 0.00 , MIPS: 0.00
unknown MAPLE Frame
Unknown 0x86 : SubCommand 0x1 - State: Cmd 0x0 Mode : 0x1 Node : 0x0
Command : 0x86,Data : 4 bytes
00000001
VREG = 00 ARMRST 01
VREG = 00 ARMRST 00
reicast-Naomi git/n - 10.13 (0.00) - 98.70 - V: 59.02 (59.50, VGA480p59.79) R: 0.99+0.00 VTX: 0.00 , MIPS: 0.00
Unknown 0x86 : SubCommand 0x13 - State: Cmd 0x14 Mode : 0x0 Node : 0x1
Command : 0x86,Data : 12 bytes
20070113 02210202 00000222
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F705C <= 4, 4
naomi?WTF? WriteMem: 5F7068 <= FFFFFFFF, 4
reicast-Naomi git/n - 9.96 (0.00) - 100.40 - V: 60.03 (4.65, VGA480p59.79) R: 12.90+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.13 - V: 59.87 (1.00, VGA480p59.79) R: 59.87+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.18 - V: 59.90 (1.01, VGA480p59.79) R: 59.41+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.37 - V: 59.41 (0.99, VGA480p59.79) R: 59.91+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.18 - V: 59.90 (1.00, VGA480p59.79) R: 59.90+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.21 - V: 59.92 (1.00, VGA480p59.79) R: 59.92+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.12 - V: 59.86 (1.01, VGA480p59.79) R: 59.37+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.44 - V: 59.46 (1.00, VGA480p59.79) R: 59.46+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.13 - V: 59.87 (0.99, VGA480p59.79) R: 60.37+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.20 - V: 59.92 (1.01, VGA480p59.79) R: 59.42+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.14 - V: 59.87 (0.99, VGA480p59.79) R: 60.37+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.42 - V: 59.45 (1.00, VGA480p59.79) R: 59.45+0.00 VTX: 0.00 , MIPS: 0.00
VREG = 00 ARMRST 01
VREG = 00 ARMRST 00
reicast-Naomi git/n - 9.99 (0.00) - 100.11 - V: 59.86 (1.03, VGA480p59.79) R: 57.88+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.24 - V: 59.93 (1.00, VGA480p59.79) R: 59.93+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.14 - V: 59.87 (1.00, VGA480p59.79) R: 59.87+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.41 - V: 59.44 (1.01, VGA480p59.79) R: 58.94+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.17 - V: 59.90 (0.99, VGA480p59.79) R: 60.39+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.19 - V: 59.91 (1.01, VGA480p59.79) R: 59.41+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.16 - V: 59.89 (0.99, VGA480p59.79) R: 60.39+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 10.06 (0.00) - 99.40 - V: 59.43 (1.00, VGA480p59.79) R: 59.43+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.99 (0.00) - 100.13 - V: 59.87 (1.01, VGA480p59.79) R: 59.37+0.00 VTX: 0.00 , MIPS: 0.00
reicast-Naomi git/n - 9.98 (0.00) - 100.21 - V: 59.92 (1.00, VGA480p59.79) R: 59.92+0.00 VTX: 0.00 , MIPS: 0.00

Another thing i noticed is there was no <insert_rom_name_here>.eeprom generated in the rom folder, unlike nullDC. The log mentions null.eeprom but i didn't see this file anywhere, also the eeprom files seem to be part of the romset in mame, so i guess they are important.

[barbudreadmon]

Good news ! I figured a fix.

Actually, it seems the .lst extension is omitted in https://github.com/reicast/reicast-emulator/blob/75189f0a783d89844e30caf4b69961230b3b8098/core/cfg/cl.cpp#L131-L151
Adding lst in the cdi/gdi/chd condition allowed me to launch a game successfully :).

[AbandonedCart]

If that is all it needs, that sounds harmless enough to fast track.

[skmp]

Lol people reverse engineering reicast xD. That’s dedication! Sorry for the (total) lack of documentation and testing, got lazy. If you have questions about how things work, why don’t you drop by #reicast in freenode? I idle there a lot. Good catch with the missing lst extension!

On Thu, 21 Jun 2018 at 03:05, TwistedUmbrella ***@***.***> wrote: If that is all it needs, that sounds harmless enough to fast track. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1188 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAYAMkgWrKlKhZiaoCQ5wS7d8oZgj1-Kks5t-vF2gaJpZM4UvGrG> .

-- ~skmp

[barbudreadmon]

If that is all it needs, that sounds harmless enough to fast track.

I didn't check further yet, there could be more issues, i'll let you know as soon as i get some time for this.

Lol people reverse engineering reicast xD. That’s dedication! Sorry for the (total) lack of documentation and testing, got lazy.

I'm working on an emulator myself (fbalpha, mainly the libretro version but i also help with the standalone version), so i'm kinda determined on doing as much as i can to fix this ^^. Not sure i'll have the skills to do it alone though, honestly i'm mainly good at tracking issues and finding small fixes.

[barbudreadmon]

Here comes the next batch of informations.

The japan H bios still doesn't work, so i decided to use the japan G bios since he is next in line for the best compatibility.

Sadly, the controls are all over the place, fixing this will be tricky, because unlike dreamcast where everything is unified, there are several kind of controls in naomi. So i guess we'll need something to determine which game is running, and setup different controls depending on the game. For now it would be interesting to fix the "standard" naomi controls, which were a 8-way joystick + 6 buttons + coin + start.

I did try a game (Guilty Gear XX Accent Core) using those standard controls, first thing i notice is that i'm controlling both players at the same time, second thing is this layout :
D-Pad 1 UP => start
D-Pad 1 BOTTOM => up
D-Pad 1 LEFT => bottom
D-Pad 1 RIGHT => left
A => nothing ?
B => coin
C => nothing ?
D => nothing ?
X => probably button 2 on the top row
Y => probably button 1 on the top row
Z => nothing ?
Left trigger => nothing ?
Right trigger => nothing ?

I could be wrong about some buttons doing nothing, as it seems this game didn't use button 2 on the bottom row. Also, D-Pad 2 and analog stick are unmapped on my controller (i guess i'll give it a try later).

I could try fixing this on my own by comparing with https://github.com/mamedev/mame/blob/master/src/mame/drivers/naomi.cpp if you give me some lead on where i should look at in the reicast repository.

[AbandonedCart]

https://github.com/reicast/reicast-emulator/tree/master/core/hw/naomi

[barbudreadmon]

Either i don't understand how it works, or the code handling inputs is somewhere else. What about https://github.com/reicast/reicast-emulator/blob/master/core/linux-dist/evdev.cpp ? Seems to me like we should add something in https://github.com/reicast/reicast-emulator/blob/1f088f211d273b86875b91b567bd08a6a9fce0fe/core/linux-dist/evdev.cpp#L161-L193

Perhaps things like
load_keycode(&mf, "naomi", "btn_1"),
load_keycode(&mf, "naomi", "btn_2"),
load_keycode(&mf, "naomi", "btn_3"),
load_keycode(&mf, "naomi", "btn_4"),
load_keycode(&mf, "naomi", "btn_5"),
load_keycode(&mf, "naomi", "btn_6"),
load_keycode(&mf, "naomi", "btn_start"),
load_keycode(&mf, "naomi", "btn_coin"),
load_keycode(&mf, "naomi", "btn_dpad1_left"),
load_keycode(&mf, "naomi", "btn_dpad1_right"),
load_keycode(&mf, "naomi", "btn_dpad1_up"),
load_keycode(&mf, "naomi", "btn_dpad1_down"),
?

Not sure how to hook those new keys into naomi though, neither how to stop having 1 controller playing both players.

[spielvan]

One question, this latest version of Reicast for Android, is already supporting the Sega Naomi games? A hug

[blackman91]

It seems the Reicast Retroarch cores already have naomi support:
https://www.reddit.com/r/emulation/comments/8tizhz/retroarch_reicast_is_running_naomiatomiswave_games/

Why is the standalone lagging behind in all those new features and bug fixes?

[AbandonedCart]

@blackman91 reicast requires that the author signs off on submissions. That means we wouldn't simply wander over to another project and begin ransacking their source and copying code. For "common knowledge" one-liners, it is easy enough to simply credit the author, but anything significant is best done with the original author's consent and involvement.

It may help to remember that only so many things can be accomplished at once. Judging by the number of issues where you left comments, I am sure you know how much is already being done. You always have the option to locate the necessary changes and get the author to submit them. If you opt for leaving it to the team, then it will take a little more time.

[skmp]

Plus, in general, I don't like to see half-done/hackfixes merged in reicast, while @twinaphex has a more lenient policy around that.

[inactive123]

I don't think the Naomi changes is something that @skmp wants, because what I did was refactor the NAOMI_BUILD compile-time ifdefs so that Naomi can be enabled/disabled at runtime. In order to do this, I had to increase some array sizes and I had to introduce some conditionals along the way. So I don't think that is where skmp wants to go with mainline Reicast. I understand and sympathize with this, which is why I didn't make a PR for it. If it IS indeed wanted, then I could of course always still attempt a PR for it.

The reason I did it like this is because we already have two Reicast cores (one with the normal GLES2 renderer, the other with the GL4+ renderer), and if we made Naomi another separate core, we would already be looking at 4 cores to maintain. If we have to later add MMU-specific cores to the mix, that becomes way too many cores to juggle between. So I was prepared to swallow a tiny little bit of performance in exchange for being able to just select between Naomi and Dreamcast at runtime.