0.122u4 [Samuele Zannoli, R. Belmont, ElSemi, David Haywood Angelo Salese]
0.114u1 [R. Belmont, Guru]

NOTES :
- Sega NAOMI is Dreamcast based Arcade hardware.
- NAMCO NAOMI games require a Namco specific BIOS
- Several early NAOMI games are running on an earlier revision mainboard (HOTD2 etc.) which appears to have an earlier revision of the graphic chip. Attempting to run these games on the later board results in graphical glitches and/or other problems.
- NAOMI 2 is backwards compatible with NAOMI 1
- The later revision games (released after GD-ROM had been discontinued) require the 'h' revision bios, which appears to have additional hardware checks.

WIP :
- 0.136u4: NAOMI updates [R. Belmont, David Haywood, Andreas Naive, Deunan Knute, Guru, ANY]: Fixed ROM mirroring in gram2000 and friends. Preliminary hookup of live decryption. Decryption + decompression is not yet supported so some games still load trojaned data instead. Deunan Knute and R. Belmont fixed decrypt method 2, added many more game keys. David Haywood converted GD-ROM games to use real PIC dumps.
- 25th February 2010: R. Belmont - Andreas Naive, with help from Deunan Knute, Guru and ANY, has come up with a working on-the-fly decryption for 2 of the 4 kinds of NAOMI cartridge decryption. This means trojaned protection data will soon no longer be necessary to play NAOMI games in emulators like MAME, Makaron, and DEMUL so your ROMs will shrink a bit. I'm doing the hookup in MAME and have some games working already, with the rest coming.
- 0.136u3: Implemented Actel FPGA ID, used by some NAOMI games for accessing a special ROM check mode [Deunan Knute, Angelo Salese].
- 0.135: Alex added later version of NAOMI analog I/O board program.
- 0.134u4: NAOMI updates [R. Belmont]: Hackish support for 8888 tiles (Shootout Pool displays something). Increased number of JVS analog channels for driving games. Added revision G export BIOS. Deunan Knute and ANY added protection data for 18 Wheeler, Crazy Taxi, Jambo! Safari and Virtua Tennis 2. Gerald (COY) added 18 Wheeler JVS I/O and motor controller dumps.
- 0.134u1: Andreas Naive further simplifications to Atomiswave's keys. Guru updated Atomiswave hardware documentation.
- 0.134: R. Belmont enforced read-only status of PowerVR ID registers in NAOMI/AW.
- 0.133u5: Major NAOMI/Atomiswave update [R. Belmont]: Emulated Atomiswave coin slots and standard Dreamcast controller. Non-lightgun games without other problems are now playable. NAOMI/Atomiswave video hardware update [David Haywood]: Basically the region array is processed, and the data from the accumulation buffer where we render is now copied into the framebuffer depending on the tiles specified in the tile list. This isn't full 'tile-based rendering' as that would require more work on the preprocessing side, and writing out the data pointers and data in real formats, however it's a step in the right direction. Currently only 565 formats are supported for both framebuffer write and framebuffer read. The accumulation buffer is alwys 8888, data is converted down to the specified format. The real accumulation buffer is 32x32 and filled per tile, ours is still full screen. The video update now just displays the framebuffer using the framebuffer read parameters specified (again only 565 is currently supported) This allows the Atomiswave bios screens and the rendered output to co-exist.
- 2nd September 2009: R. Belmont - Guru and I are also working on protection data dumps for other previously dumped NAOMI carts including Heavy Metal Geomatrix. We believe all the data comes from the ROM contents. There are 3 forms of protection on NAOMI carts: 1) DMA with a special flag. This apparently does an address line scramble although I don't know if anyone's proven that yet. Not used often, but e.g. Death Crimson OX uses it for nearly the entire contents of the cartridge. 2) PIO with direct decryption. Here you give the cartridge's ASIC an address in ROM to read and a 16-bit decryption key and start reading data - it gets decrypted on the fly by the ASIC. 3) PIO upload/download. The ASIC has 64k of internal RAM to buffer encrypted data, so games can also upload data to that RAM, set the 16-bit decryption key, and read back decrypted data. Andreas N. has been working on methods 2 and 3 with some success, but he hasn't fully cracked it yet so we still have to trojan decrypted data out of the real system.
- 1st September 2009: R. Belmont - I hooked up the controls for Atomiswave games using stick/button controllers in MAME so now you can (slowly) play the games. Guru dumped the NAOMI cartridge Project Justice. We're still working on the protection data but we've got enough for some nice new screens below. This is completely independent of Alex's dump which he still has not released, but if he'd like to release right now I'd certainly appreciate the time savings on the protection. And yes, Guru is now capable of dumping the protection data - he's got a working NAOMI USB dumper and 2 of the MAME Italia crew have the parts for theirs on order.
- 31st August 2009: R. Belmont - In reference to recent "dumper drama", I stated that dumpers who want to be #1 on my list should flood me with dumps. The Guru is taking me up on that offer. Here's some screens of new dumps running in MAME. More games will be added throughout this week, so keep checking back (NOTE: Neo-Geo Battle Coliseum was also dumped but couldn't be completed yet due to weird TSOP-70 mask ROMs. It doesn't boot due to the missing data). Update: King of Fighters Neowave has been dumped but doesn't run in MAME yet. DEMUL is looking good though!
- 0.133u3: NAOMI update [R. Belmont, Deunan Knute, Alex, Mr. Mudkips, Frank Bukor]: Use new dump for ic31/32, ROM board test passes now. Added protection hookup for Marvel vs. Capcom 2, game is playable w/sound.
- 19th August 2009: R. Belmont - I'm making good progress on a dynamic recompiler for the SH-3 and SH-4 CPUs (NAOMI/Atomiswave/Dreamcast and friends, and eventually some of the later Cave games like Ibara). It helps that these processors are back compatible to the SH-2 which I previously wrote a DRC for, so that previous work serves as a well-tested base on which to add support for the newer features of the newer processors. With only 1 new instruction added the new DRC already can run the entire Atomiswave BIOS and start executing the actual cartridge (that BIOS is very minimal - it's roughly 4k of total code). No, I don't know how much the speed will improve, and there's plenty of work to go yet.
- 0.133u2: Atomiswave updates [R. Belmont, Cah4e3]: Hooked up flash ROM at 0. Use proper AW memory map. Emulate AW ROM board (different from NAOMI). Wind fixed Atomiswave COIN ERROR.
- 6th August 2009: R. Belmont - Thanks to the awesome help of Cah4e3 games on the Atomiswave system are booting in MAME. Here's some screens from Sports Shooting USA. Update 7th August: Fixed the Atomiswave COIN ERROR thanks to Wind from the DEMUL project, and here's a few more screens.
- 0.133u1: NAOMI/AW updates [Deunan Knute, CaH4e3, R. Belmont]: DMA protection emulation for 'Quiz Ah Megamisama', fixes in-game graphics. Correct decryption key for Extreme Hunting. Added protection emulation for Cosmic Smash. Added protection emulation for Capcom vs. SNK (cartridge ver). Andreas Naive simplified Atomiswave's decryption routine & keys.
- 21st July 2009: R. Belmont - The talented and friendly Deunan Knute from the excellent Dreamcast/NAOMI emulator Makaron has been doing excellent work understanding the protection used on NAOMI cartridges (it's not present on GD-ROMs). You may have noticed MAME 0.133 had protection emulation for DOA2 and DOA2 Millenium, but other faults in MAME's emulation are still making them unable to run. Better news comes in the form of qmegamis, aka Ah! My Goddess! Quiz. Here's how it looked in MAME 0.133 without protection emulation. Big thanks to Deunan and do try his excellent emulator while you wait for the speed to improve in MAME.
- 0.133: NAOMI update [Deunan Knute, R. Belmont]: Implemented cartridge protection chip mechanism. Added protection data for doa2 and doa2m. Added real dumped cartridge X76F100 data for doa2/doa2m.
- 0.132u3: Gerald (COY) and Dumping Union added latest US BIOS (EPR-21577G, version 1.70) to the NAOMI driver.
- 1st June 2009: CaH4e3 - NAOMI 2
WIP : Anyone, we need some pictures or movies from real hardware to compare some parts of image. Is it possible to make some capture from video output? Any of dumped just now, vf4tuned, initd, vs3, etc...
- 0.132u1: Guru updated README's for ATOMISWAVE hardware.
- 0.131u3: David Haywood splitted NAOMI video ram access into 32-bit / 64-bit areas pending more information on how they actually mirror (it's not a straight mirror).
- 0.131u2: MooglyGuy and Aaron Giles added bilinear filtering to NAOMI video emulation. Textures are still a bit crawly since we're always using the lowest MIP level. Olivier Galibert added all 64 NAOMI blending modes, and made it anal-retentive where it comes to position precision. Angelo Salese improved the debugging on the NAOMI Wave DMA handler. David Haywood added profile markers to NAOMI, to see what % of CPU time things are using, fixed a NAOMI bug in computedilated, was causing bad portraits in Shikigami No Shiro II, short-circuit the NAOMI rendering in non-textured cases, prevents a lot of bad graphics, applied some of the NAOMI exclusive texture mode rules and hacked the NAOMI YUV interrupt to fire when the base register is written (currently looking at implementing properly, but this prevents KuruCham and SS2005 from hanging).
- 9th May 2009: MooglyGuy - I've been eagerly watching all of the NAOMI development that's been going on in MAME behind the scenes, and I couldn't help but stick my own grubby mitts into the fray. It was pretty easy to get bilinear filtering going, and it definitely helps Spikers Battle's attract mode (compare with the shots on Haze's blog).
- 9th May 2009: Angelo Salese - Fixed some maple bugs in the NAOMI driver (especially one that was trashing memory with invalid commands), now Moeru Casinyo, Street Fighter Zero 3 Upper, Shinkigami no Shiro 2*, Trizeal and Under Defeat boots fine and are playable.
- 9th May 2009: David Haywood - NAOMI / Dreamcast can store textures in various formats in RAM. Like many 3D Systems there is a format which is optimized for drawing with the graphics hardware. This is called a 'Twiddled' texture, and has the data arranged in a way which means the hardware can process it more efficiently. The Playstation does the same with it's 'Twizzled' textures. Kale recent came up with a fix that got Shikigami No Shiro II booting (the sequel to the recently emulated G-Net title), however, there was a problem with it. The character portraits on the select screen were corrupt. This wasn't the first time we'd seen such corruption, Lupin The Third - The Typing suffered from a similar problem. I decided to take a look at the problem. After a lot of searching, and false leads, it actually turned out to be something really simple. The lookup array we use for decoding Twiddled textures wasn't being fully populated by the code, and driver was attempting to use an uninitialized entry, and things were generally appearing rather broken. A very simple fix gave fixed portraits in Shikigami and fixed HUD graphics in Lupin.
- 0.131u1: Olivier Galibert and David Haywood hacked up NAOMI 3D renderer. Still quite preliminary but allows most games to show graphics. Angelo Salese splitted NAOMI / Atomiswave memory maps, improved sorting of the driver and modified Mahjong games in NAOMI to be loaded with a specific configuration, will add the proper JVS hook-up later.
- 6th May 2009: David Haywood - On the Dreamcast / NAOMI if a texture has the 'MipMap' flag set, then you have to take special care when addressing the textures. Basically MipMapping means that there are a bunch of textures of different resolutions stored in texture ram. Depending on how large the image is to appear on screen a different texture will be used. We don't emulate MipMapping yet, but that doesn't mean we can just ignore it. The way the textures are stored in memory means that the lowest resolution textures are stored first. Therefore if your texture should be 512×512, there are copies at 1×1, 2×2, 4×4, 8×8, 16×16, 32×32, 64×64, 128×128 and 256×256 stored in video ram before it. Depending on the texture mode (bpp, palettized, compressed etc.) and size of the texture, you have to apply the correct offset to get the full resolution texture. Previously we were just drawing the first one every time, which was the source of the issues seen in the Ikaruga and Border Down screenshots in the previous update. By selecting the correct texture things look much better (Ikaruga, Border Down, Radirgy, Karous, Choco Marker and Spiker's Battle).
- 5th May 2009: David Haywood - NAOMI progress has been rumbling on... Probably the most significant progress is that Olivier Galibert beefed up the test renderer a little bit, I helped find a few bugs in the new code, and combined they give us some much nicer (although FAR from perfect) images. I'll add images to this post as the night goes on. These things run at anything between 10-20% speed, so as you can imagine, making images and testing them is somewhat time consuming, but these are 100% genuine MAME shots (Guilty Gear XX Slash, Ikaruga, Border Down, Slashout...).
- 0.131: Added machine\naomi.c and includes\naomi.h. Miscellaneous NAOMI work and improvements. Some refactoring done, new games hooked up. Moved NAOMI DRIVER_INITs to specific machine/naomi.c & includes/naomi.h files, in preparation to add per-game JVS settings [Angelo Salese, David Haywood].
- 18th April 2009: CaH4e3 - Demul NAOMI2 Virtua Striker 3
WIP : First approach. Memory areas, interrupts and SQ transitions are correctly mapped, game start to send vertex data and textures to GPU chips. Some of them - standart TA polygons like sprites, so they was directly moved to TA pipeline. Second approach. Apart from standart TA data, TL packets was reversed and extracted some geometry for scene. Actually first guess for transformation matrix was wrong, so we see here only static geometry. Third approach. Real transformation matrix was found and now we can see moving objects and more geometry details. But something missing. Fourth approach. Ahhh... Found more geometry packets. Now scene look actually complete. What is next? Textures. Some research needed and we got packets with almost the same TA polygon control words in TL vertex data. After some work with TL packets to TA converter we've got both TL and TA geometry textured on screen. Just axis flipping is wrong. Not really big problem. Colors some kind of fixed now too. When hardware T&L will be implemented, colors will me more natural. Current state, after little bit of transformations we have here perspective correction. It is almost playable now.
- 17th April 2009: Guru - I have been trying to get Makaron working for some time and finally figured out the missing pieces and got it booted (multiple cryptic Microsoft errors caused by missing MSVC 2008 runtime libraries!), but couldn't get the ROM cart part of it working (since there doesn't seem to be any info around about exactly how to get the ROM carts working!!!). I would really appreciate it if someone could shed some light on that as it would be good to test the new dumps in Makaron to make sure they are good. So I decided to look at the GDROM based games. I picked GDL-0007A that I dumped some time ago. You should all know that is 'Capcom Vs SNK 2 Millionaire Fighting 2001' right? Well I thought so too. After hacking out track 3 from the CHD by using CHDMAN then BIN2ISO then ISOBUSTER (OMG come on guys this is 2009 now, can't we just support the official CHD and be done with all the hacking??), decrypting it (using a program that comes with Makaron) and then running it in Makaron it seems we have a new game! It seems if you run it with a Revision D Japanese BIOS (EPR-21576D) it comes up as 'Mark Of The Millennium' and with Revision G Japanese BIOS (EPR-21576G) it comes up as 'Millionaire Fighting'. It's funny that no one realised. The rest of the game is identical, of course. I wonder if any other games come up with alternative titles when run with a BIOS of the same region but different revision?
- 0.130u4: NAOMI updates: [Angelo Salese, David Haywood]: Added H-Blank IN interrupt hook-up. Generate a few more interrupts (ISP end of render, VIDEO end of render). Some minor improvements to the irq system. Added a bunch of Maple-DMA fixes, now moeru boots too.
- 0.130u3: Angel Salese improved the Wave-DMA behaviour in NAOMI and added PVR-DMA support. ElSemi and Angelo Salese added direct mode JVS command, used by the I/O for the later NAOMI games. NAOMI video cleanups/reworking [David Haywood]: Many NAOMI tweaks and improvements, plus a number of new sets added but with no credit specified.
- 5th April 2009: DEmul 0.5.3 WIP - Changelist: Vertical shooters screen rotate. Screen filter shaders similar to Pete OGL2 plugins (find examples at GPU). SPU sync and buffer options. Added all NAOMI/NAOMI2 dumps (not tested at all). Autosave SRAM/EEPROM settings. FLASH file autocreation. Fixed sound in most arcade games. Added multiple ROM path support.
- 30th March 2009: Angelo Salese - I'm been busy with the Sega NAOMI HW stuff with Haze and ElSemi. Thanks to the latter, I've implemented a JVS unimplemented I/O command used by the later games of this hardware. Now Ikaruga (and others) can be coined up and are playable (although still in wire-frame)...
- 29th March 2009: DEmul 0.5.2 WIP - Changelist: Atomisware support. Optional multithread SPU & GPU support. Implemented render-to-texture technique (fixed REZ, RE:CV, CT, ILLBLEED). Smooth texture palettes. Fixed AICA DMA. Added new DMA for PVR2 (fixed Sonic Shuffle). Fixed errors in the GD-ROM DMA, fixed launch WINCE GD-ROM. Optional SPU DSP. Video framelimit. X64 binaries.
- 0.130u2: Some various NAOMI improvements and SH4 hacks to get more NAOMI games booting farther [Angelo Salese, David Haywood].
- 25th March 2009: David Haywood - After hacking in some basic MMU support to the SH4 core, a couple more things boot (Border Down and Ikaruga). The majority of the 3rd party NAOMI games are built with an SDK version which requires the MMU, so, this resulted in quite a few more showing some graphics. Until I implement some kind of texture mapping for the 3D stuff you'll have to enjoy glorious wireframe.
- 23rd March 2009: Guru - NAOMI 2 + Dimm board arrived from Japan.
- 22nd March 2009: David Haywood - I recently fixed a few NAOMI bugs, and now Capcom vs SNK actually gets past the character select screens, and ingame. It runs at about 20% speed on my 3Ghz C2D, and that's without any real 3D emulation, but it's pretty cool to see all the same (and you can actually 'play' if you don't mind most of the graphics being blocks of nothing). Both this and Guilty Gear look like they could be potential 'first base' emulation targets for MAME's NAOMI emulation. I'll see if I can get them there.
- 0.130u1: David Haywood added idle skipping and a number of new sets to the NAOMI driver.
- 0.129u6: Yasuhiro Ogawa fixed descriptions of the games running on NAOMI series hardware and NAOMI game list.
- 23rd January 2008: Guru - NAOMI emulation is getting better so there's more focus to get more of the games dumped. Particularly the ROM carts. However there's a couple of issues. Some carts use strange ROMs that are not dumpable and all of them will take serious time to dump the normal way (remove surface mount ROMs and read, then put back etc) due to the large amount of ROMs on the boards. After discussing this problem with ElSemi, he's come up with a solution which allows dumping the carts while they are plugged into a NAOMI unit using a custom PIC-based NAOMI JVS adapter linked to a PC via USB that he developed. Initial tests proved positive but more tests are required and a custom NAOMI BIOS needs to be created and extensively tested.
- 0.129: Aaron Giles fixed assertion in all NAOMI sets.
- 27th November 2008: Guru - Various junk NAOMI pieces arrived this week, incl. 2 main boards, 2 GDROM units, and 2 DIMM units, possibly for decapping of the custom chips or other scientific experiments.
- 0.128u4: Added machine\naomibd.c and includes\naomibd.h. NAOMI improvements [Samuele Zannoli]: This patch implements a "plug-in board" device for the NAOMI. It represents the rom boards for the various games and the dimm board for the gdrom. Now the dimm board part contains only a routine to load the "rom file" from the gdrom chd, but in the future it will be fully implemented with its own sh4 processor, security chip and so on. Extracted .rom files are no longer needed, rom file is extracted in code from the CHD.
- 0.128u3: NAOMI update [R. Belmont]: Unmapped words in the AICA address range must read as zero.
- 0.128u2: NAOMI / DC improvements [Samuele Zannoli]: cvs2gd stops looping forever. However with no 3D emulation nothing is shown on the screen. Implemented the NAOMI RTC and there are a few verbose messages about transfers not implemented yet.
- 0.127u5: NAOMI update [Samuele Zannoli]: Now show the contents of the framebuffer if the 3D accelerator is not used so that now the atomiswave logo and messages are shown (if you wait enough). Removed a maple bug (no more strange messages in cvs2gd). Improveed documentation of the communication registers with the dimm board. If DEBUG_VERTICES is defined as 1 the vertices sent to the tile accelerator are collected and then drawn connected by a segment to give a wireframe-like view of the scene. Brian Troha added newest NAOMI 2 BIOS.
- 0.127u3: David Haywood started adding NAOMI2 sets, also started changing NAOMI GDROM sets to load the CHDs. Also moved all NAOMI-GD keys to external.
- 0.127u1: SH4, NAOMI and Hikaru updates [Samuele Zannoli]: Fixed SH-4 core to support multiple SH-4s. Fixed LDCSR instruction. Fixed SH-4 I/O ports. Skeleton Hikaru driver with memory maps and both SH-4s. Fixed JVS/Maple translation so NAOMI 2 BIOS runs.
- 0.126u3: NAOMI updates [Samuele Zannoli]: Refactors many global and static variables in video\dc.c into a structure to hold the tile accelerator state. Textures should be better decoded with more pixel formats. Adds placeholders for some register probably used by the dimm board. Corrects a little error in a rom board register (very important, now at least the gdrom games try to start). AICA updates [Deunan Knute, ajax16384, R. Belmont]: Fixed ADPCM decoder to remove bad optimization and match YMZ280B. DC offset is fixed and output quality is higher. Improved handling of slot monitor and timers B and C. Simplified interpolation code.
- 30th July 2008: Guru - ElSemi designed a nice circuit to dump PIC and GDROMs. I've just completed building it and wiring it up. Mine is somewhat sexier than ElSemi's proto though (the first pic) with white letters and blue backlight. Update: The fun has ended. Around 20 seconds per PIC and 20 minutes per GDROM and the job was done. You might notice some additional 'random' numbers now added to the NAOMI Dump Status page and a LOT more light green.
- 20th July 2008: Guru - NAOMI I/O board (with Toshiba MCUs) arrived. Thanks to a couple of people that donated some postage bucks, another package arrived from Japan a few days ago.
- 0.126u1: Added machine\gdcrypt.c (DES decryption, used by GD-ROM based titles).
- 0.126: AndyGeezer added new Ferrari F355 USA BIOS to NAOMI.
- 4th July 2008: Guru - NAOMI DIMM board arrived a couple of days ago. I've just dumped the DIMM firmware (look closely, it's the TSOP56 flashROM located in the middle of the board), so that's another piece of the NAOMI puzzle sorted.
- 0.125u7: Brian Troha and Guru added additional NAOMI BIOSes.
- 0.124u1: Added a little magic to make the Atomiswave BIOS try to be more verbose, but our video emulation isn't good enough to cooperate [R. Belmont, ElSemi].
- 0.123u4: NAOMI update [R. Belmont]: Use defines instead of magic numbers for IRQ sources and added VBL-out.
- 0.123u2: Added and hooked up AICA (NAOMI sound chip) [ElSemi, Deunan Knute, kingshriek, R. Belmont]. NAOMI boot sounds now play. Since it's an SCSPx2 with ADPCM instead of FM it's already feature-complete except filter envelopes. A bug in the ARM7 core causes incorrect pan/level values to be computed in e.g. the NAOMI SOUND TEST menu. This is not a defect in the AICA. Samuele Zannoli added missing SH4 instructions and bugfixed the FPU. Improved NAOMI I/O board emulation and added working controls.
- 20th February 2008: R. Belmont - Samuele Zannoli is at it again. Here's a few SH-4 bugfixes and NAOMI I/O board improvements later... I've also further improved the NAOMI audio emulation. No new samples this time, sorry.
- 17th February 2008: R. Belmont - A couple of nice coincidences linked together today. First up, I submitted the AICA core to MAME so NAOMI now has sound. Secondly, MAME Italia dumped the NAOMI game Toy Fighter, which successfully gets into it's test mode (gameplay doesn't work). Test mode includes the sound test, and here's what it sounds like (http://rbelmont.mameworld.info/naomi.mp3), with the "load times" cut out to make things managable. The first thing you hear is the NAOMI boot chimes - they're supposed to not be very loud, so do not crank your volume when you hear them. After that is song #1 in Toy Fighter's sound test.
- 0.123u1: NAOMI/DC updates [Samuele Zannoli]: Added SH4 I/O ports. Connected the 93C46 of the NAOMI and the x76f100 of the rom board and filled them with dummy data to satisfy the BIOS. Implemented some of the JVS transfers that will be needed to use the controls. Implemented ROM board DMA. Set proper NAOMI ram sizes (32 MB main, 8 MB for AICA). Improved PVR-TA graphics emulation. Deunan Knute set proper ARM7 clock for NAOMI. Arzeno Fabrice added HOTD2-specific BIOSes. Changed ARM7 CPU2 clock speed to 2822400 Hz.
- 8th February 2008: R. Belmont - I'm working on emulation of the AICA (the Dreamcast/NAOMI/Hikaru sound chip) behind the scenes with kingshriek and Deunan Knute (the Makaron author). Since AICA is basically an SCSP with double the voices and a few feature tweaks the obvious basis was MAME's now reference-quality SCSP emulation. That's allowed us to get to a pretty good state in a short amount of time. The ADPCM sample type isn't quite right yet, but the basic functionality and effects DSP are working. You can take a listen to the current code playing a piece from Sakura Taisen 4 right here (http://rbelmont.mameworld.info/sakutai4.mp3). Note: This is via a DSF rip, it doesn't mean Dreamcast works in MESS.
- 3rd February 2008: Guru - I dumped all of the NAOMI GDROMs I have here so maybe we'll see some progress on that one day.
- 0.122u7: Brian Troha added all the known NAOMI BIOS dumps along with descriptions and information about region bytes & version numbers.
- 22nd January 2008: drkIIRaziel - NAOMI .. or not ?: It seems that the NAOMI 2 BIOS just wanted some DIMM emulation to get to the error screen. NOTE: NAOMI1 or NAOMI2 GDROMS are STILL FAR AWAY FROM RUNNING.Dont get too excited. I must apologise to the ppl with slow connections but meh. Fixed some misc stuff on input etc. Update: Got the roms mapped to ram & implemented the basic I/O needed to make BIOS happy. I was messing today with some old NAOMI emulation code and after fixing some stuff on the dynarec/PowerVR plugin i got the NAOMI BIOS running .Lets see how hard it will be to get it to boot a game now.
- 18th January 2008: Corrado Tomaselli dumped NAOMI BIOS.
- 0.122u4: Updated NAOMI driver, now boots and shows some test menus (press F2 to get into test, use 9 to navigate). Also hooked up ROM board PIO and skeleton for DMA [R. Belmont, Samuele Zannoli, ElSemi]. Changed region user2 to user1 and visible area to 640x480.
- 5th January 2008: R. Belmont - A little help from my friends: Some small progress. Big thanks to the hero of Chankast, the man they call ElSemi for some technical details. You'd think Sega would have given up on throwing random weirdness into their hardware by this point. You'd be wrong.
- 4th January 2008: R. Belmont - World of wonders: It turns out that Samuele Zannoli, who's been fixing up the SH-4 CPU core, had a little something extra to go with it... (NAOMI BIOS screenshot). Don't get too excited, we're still a long way from playable games, and the speed is pretty bad.
- 0.122u3: R. Belmont implemented a few SysCtrl, Maple and PVR/HOLLY bits for NAOMI. The BIOS gets slightly further but is nowhere near starting up.
- 0.117u3: Added machine\dc.c, video\dc.c and includes\dc.h. R. Belmont added some more setup code for the NAOMI driver. Changed ARM7 CPU2 clock speed to 45MHz.
- 0.117u2: More SH-4 core improvements [Samuele Zannoli]: Fixes some problems in the disassembly and instruction execution from the previous patch. Adds interrupt management. Implements register bank switching. Adds RTC timer counters. Adds memory refresh timer. Adds three timers of the TMU. Adds a configuration structure that specifies the processor mode pin settings. Internal processor clock frequencies calculated from mode pins and cpu clock. Systems using the sh4 processor secify its configuration structure.
- 0.116u1: Added cpu\sh4\sh4regs.h. SH4 CPU core improvements [Samuele Zannoli]. Corrected opcode decoding and decodes all opcodes of sh4. Corrected disassembly and added support for all instruction of sh4 not in sh2. Now disassembles all fpu instructions. Changed cpu to little-endian (sh4 can be either big or little). Corrected which data is read and written from memory. Added registers fpul and dbr. Added (non-fpu) instructions of sh4 non present in sh2. Added four fpu instructions. Modified memory-mapped internal register management so that all internal registers can be actually used. Added file sh4regs.h with defines for all sh7750 internal registers. Implemented internal registers for memory refresh. Removed some internal registers of sh2 not present in sh4. Added preliminary support for internal cache used as ram.
- 0.114u1: Added naomi.c driver. Added skeleton driver for Dreamcast-based systems. R. Belmont added very preliminary SH4 CPU core.
- 30th March 2007: Guru - One other free Atomiswave cart with Japanese text (i.e. unknown ATM) just arrived, the next haul from Japan. I suspect that many more free Atomis carts will be coming off the arcade floor eventually because they don't seem to last long in Japan before people are tired of them and revenue drops to zero.
- 21st February 2006: R. Belmont - Suppose the Dreamcast had hung on and perhaps even knocked out the Xbox or Gamecube. What kinds of crazy hidden powers would developers be pulling out of it now? Look no further than the latest NAOMI games. Yeah, NAOMI has more RAM, but the ports released during the DC's lifetime didn't exactly show huge degredation. Anyway, one of the great untapped resources in the DC was it's AICA audio chip - in every respect it's roughly 2-3 times more powerful than the Saturn and Model 2/3's already formidable SCSP, but Dreamcast games tended to just stream their music and ignore the available synthesis power. On arcade platforms it's much harder to justify streaming so NAOMI games tend to let the AICA really come out and play. So what all this is leading up to is that Warashi has posted an OST MP3 from their gorgeous new NAOMI shmup Trigger Heart Execelica. You can have a listen right here (http://www.warashi.co.jp/exelica/exelica_stage_1.mp3) and think about what might have been. (Even if you don't like the actual musical style, which is distinctly 80s synthpoppy, just listen to the quality of the instruments).
- 11th August 2003: Guru - A Slash Out GDRom version arrived. I've been fairly busy over the past week so didn't get chance to update here. However, this stuff arrived just recently thanks to Mameworld contributions. Also picked up a Dreamcast Broadband Adapter. So hopefully there will be some NAOMI GD-ROM dumps coming soon.
- 5th July 2003: Guru - Just picked up a cheap Dreamcast bundle, which will be hopefully used to dump NAOMI GDROMs.

56 views for sgtetris ::: Popularity rate : -2176 (sadly negative) ::: previous page ::: top of the page