Hi!... My name is Roberto Fresca, and I'm part of the MAME Dev Team since 1998. Here you can find some technical things about rare hardware. This site is intended to avoid duplicated efforts. This page is *always* in construction... :-)
--- 2014/06/03 17:18 ---
Finaly found the attributes RAM for each reel and worked it to get perfect graphics. So the video circuitry finally has 5 (FIVE) layers!... About the 1Ch-1Dh commands, I was right. They are just part of a RAMDAC initialization (thanks Angelo for the tip!), creating and dinamically modifying a 256-entries 6-6-6 RGB palette. Even when through attr RAM the tiles use bits 4-5-6-7 for colors, I can't find a proper way to map the color codes. Just displaced the color codes in the gfxdecode start to get the title girl ok, and most of the 1st gfx bank colors ok (it's a temporary hack till can understand the palette format/bank/access.
Here some snaps:
BTW, the game is working properly. The only issue is the lack of accurate colors.
--- 2014/05/24 23:33 ---
Here is another dark hardware manufactured by Sang Ho (bankrupt approx in 2010), silkscreened 'SANGHO PM-003 (VER-B2)'. It's a Z80 based system, with 1x KC89C72 (AY-3-8910 compatible), a couple of Altera CPLD's (one identified as EPM7032LC44-15), 1x unknown custom IC marked 0453LK003, and 1x (unused) OKI 6295 for ADPCM samples system.
I spent some hours on it, and got some preliminary response...
Seems that the system has at least two video RAMs that should render to different tilemaps, and finally merge for the final output. For the next step, I will try to find and hook the second one...
**** EDIT #1 ****
Well... Just identify the whole second video RAM, hooked it to a new tilemap, and cooked all together... ;-). Also adjusted the 1st graphics bank and tile codes accordingly.
Still preliminary, but at least now we know for sure the game involved *IS* 'Star 100'. Here some snaps:
The garbage added to the first two snapshots is suppossed to be girls posing, and respect the last one, obviously must be the reels symbols. I'll try to do an attempt to fix some things...
**** EDIT #2 ****
The girls appeared (also the reel symbols), but obviously with 8 pixel lines instead of 32. The second tilemap is not accurate, since is supporting 8x8 pixels tiles, when the second graphics bank (that contains the girls and reels symbols) has tiles with a size of 8x32 pixels. This is why you can see the tiles lines 'clipped'.
Even when this is an improvement, I'm forced to break the second tilemap in different pieces, giving 8x32 pixels support. I'm strongly think that this video circuitry starts to looks close to the Golden Star video hardware...
**** EDIT #3 ****
Tried setting an accurate second tilemap for these 8x32 tiles. Still misaligned, but it's a start. Palette seems to be RAM based, placed at 0xF300, but I couldn't find an accurate way to decode it, yet.
Also the program sets a sort of initialization through commands using ports 1Ch & 1Dh, writting the index or address through port 1Ch, and then writting 3 consecutive values to port 1Dh. The range starts with index 0 and ends with 255, what made me think that maybe the program is writting a 256 colors palette based on RGB values... Need more investigation.
--- 2014/05/19 18:58 ---
More work on Bordun butterfly games. This is a new game called Gallag / Petalouda (greek for butterfly). The game has both titles (selectable through a DIP switch). Also seems to have Butterfly x50 bonus. About the program, it has the capabilities to arrange the correct graphics based on the hardware type (also selectable through different DIP switches). Here some snaps:
--- 2014/05/12 15:44 ---
Just doing some work on Bordun games. Finally got Leader working, and a new addition called Neraidoula (Fairy Butterfly). Seems that all these games are working in at least 3 different hardwares (Rolla, Sky, and Cleco). Not clear which is which here, though... Here some snaps:
Leader (version Z 2E, Greece):
And now... Neraidoula (Fairy Butterfly):
--- 2014/05/12 10:58 ---
Oh... Yes. After long years running robertofresca.com.ar, I was forced to migrate to a global one. The reasons: The government transferred nic.ar to a militant group that seems that has no idea about right direction. They started to charge the equivalent of USD 58.- to allow you to get a '.ar' domain for one year!... Yes. FIFTY-EIGHT DOLLARS just for allow people to use their own domains for ONE YEAR, against the USD 8-12 usually charged worldwide. After seen some oddities, also seems that they will decide if you deserve the domain when is next to be renewed. I don't want them scamming and biting my legs like dogs.
So, moved to...
...that is more global, cheaper and sane. Update your bookmarks...
--- 2014/05/10 12:33 ---
This is a PacMan style game, manufactured by a spanish company called Petaco S.A.. The year is unknown, but seems from early 80's. Being a well known title for spanish people, the game was one of "most wanted" in the emulation scene... Till now.
Thanks to Punkyas (from AUMAP) that owns the whole cab, and Ricky2001 (from AUMAP) that kindly dumped the game and sent some pictures and info, we have a new legend preserved.
The game has a complex banked video RAM, that transfers 8 bits stripes from bitmapped graphics to the screen. I team'ed with Angelo Salese and got good results... See some snaps (click for full size):
The game has some oddities, like a missing coin input (25 pesetas) under some dark circuitry, current coin circuitry latched with NMI, a blank bipolar PROM that is supposed to store the color info (temporarily faked), etc... But is working properly. See a little video of the game runnung (switch to fullscreen for better results)...
And here, a couple of pics of the cab and control panel details...
--- 2014/04/3 23:58 ---
Just some work on Double Crown. Added MAX693 watchdog reset, coin in/out counters, fixed some inputs, and finally added button-lamps support. As always, I tried to do the most realistic layout allowed by our limited internal artwork system. Here some shots...
In other news, David Haywood hooked a second VRAM on Cuatro en Linea. Now we can see the 'interlaced' graphics. Respect to the rest, not much more improvements... Seems that *every* (data/address/switches/coins/etc) is passing through the big FPGA (ES2 9046). Also each time I try to disassemble the (bad called) audio CPU program, I end the attempt totally stressed. A lot of non-sense routines, checksuming empty offsets segments among other obfuscated things. The good thing is that thanks to Rockman from AUMAP, we're documenting the custom IC's and their peripheral circuitry.
Well... Here the (even/odd) fixed interleaved graphics.
--- 2014/03/27 16:20 ---
Thanks to the AUMAP people (who contributed with a lot of spanish sets before), we have a new game dumped... This is 'Cuatro en Linea', and it's running in a very rare and complex platform with 2x Z80 (one for audio), 1x AY-3-8910 for sound, 1x UM487F (HCGA controller), and 3 unknown custom ICs...
These custom ICs are:
1x ES2 CM3080 (unknown DIP-18 IC)
1x ES2 9046 (unknown PLCC-84 IC)
1x 8952 CM 32 (unknown DIP-40 IC)
We're trying to reverse them. If someone knows something about, please contact me through the contact page, or just post on Mameworld Forums.
Graphics are using a bitmap mode that seems way complex. Some days ago, David Haywood sent a preliminary driver, and I joined to see if we can improve the emulation. The system is using a HCGA controller (UM487F) that has a UM6845R embedded. So I reversed the code to see how this monster is set. There is not much, but we passed from this...
'Und seines Bellens lauter Schall... Beweist nur, daß wir reiten'. :-)
**** EDIT ****
Playing with the UM487F status register, got an overflow, that allow me to know our hero... (or our H.E.R.O.?) ;-)
More to come...
--- 2014/03/18 15:50 ---
Nice hardware mess... This is a 8-bit game running in an hybrid 16-bit platform (i8086 based) with some 8-bit devices. All in this hardware seems way complex.
Just team'ed with Angelo Salese and got some improvements. At least the game starts to show something. Unfortunately, the complexity of interrupts device + driver + timer devices intertwining with the CPU prevents us to get more improvements for now. Here a couple of snaps...
After a reset...
--- 2014/03/14 22:30 ---
Here is another rare hardware manufactured by 'Amusco' in earlies 80's. The PCB has an empty socket for the missing CPU. Since there are 1x AMD D8284A (Clock Generator and Driver for 8086/8088 Processors) and 1x NEC D8259AC (Programmable Interrupt Controller), we can assume that the missing CPU is a 8085/8086/8088 type.
The game is 'American Music Poker', and also has a MOS 6845 as CRT Controller, two P8255 (Programmable Peripheral Interface) for I/O, and a SN76489 (Digital Complex Sound Generator) obviously for sound.
I started a new driver for this rarity. Graphics are properly decoded (strange 8*10 tiles, also x-flipped). Now need to guess the correct CPU. For now, I can't find what fits there... The code lacks of sense. See some decoded tiles...
--- 2014/03/12 18:30 ---
I was digging into the Golden Star hardware and found a lot of things... Some years ago when I was starting to work in this hardware just suspected that Golden Star is a hack from another game. Now, after a lot of reverse engineering, I can say for sure that the IGS crew heavily hacked another game to bring Golden Star to the life.
All is pointing that the big hack was made to lower the hardware costs, since the PCB lacks of the 3x PPI 8255 devices originally meant for I/O. All inputs are driven through PPI programming mode, and also there are some init sequences inside the code. Briefly, the game code handles the inputs like they are tied to physical PPI 8255's when they aren't present. Also outputs have a lot of leftovers from another hardware, like the lamps support. All lamps are driven in a strange way through port F900h, but whilst some lamps are turning on (like stop lamps) the code try to write to the inexistent port F850h (from Cherry Bonus III hardware) trying to lite them through another format (like Cherry Bonus III hardware does).
So sad, that the classic from IGS (and header of the driver) is just a complex hack to drive most of the functions/tasks from the big CPLD present in the board. Well... Once reversed the port, figured out how the lamps are meant in the control panel. Also adjusted the inputs accordingly. Here the results...
Note the Ace present in the bonus screen above. It has the 'D' from Dyna logo... ;-)
--- 2014/03/10 15:00 ---
I found the output ports related to lamps. Cherry Bonus III (and clones) have two controls set. These could be switched in game (like Mahjong games). One control set looks normal, and the second one seems to be meant for non-stop reels. Ports F850h and F860h drive the lamps for both controls set respectively.
Now I added lamps support, and created a realistic control panel layout with 3D button-lamps. Here a couple of snaps...
See the details...
--- 2014/03/09 21:30 ---
More news... I got running a Royal Card set that is totaly in French instead of the normal Austrian language, dumped by f205v.
Here some snaps...
In other news, we got perfect graphics decrypt/decode for the Cherry Bonus III (the one with a big CPLD for the encryption). With this work, the game is now working properly and was officially promoted to working state. Here some snaps...
--- 2014/03/04 15:45 ---
More dumps from ANY... Sometimes I think he owns the Felix Magic Bag of tricks...
(thanks Anna for the cute Felix art. Seems ANY bringing me more
troubles work) ;-)
Two different sets of Italian Jolly Card (with different colors and different cards deck graphics), One Royal Card, and finally a new Cherry Bonus III highly encrypted. Here some snaps:
Here the set with also different colors, and diferent cards deck graphics...
Some commercial spam in the graphics... Not shown in the game.
Now a new set of Royal Card. Also with different text color..
And now, the Cherry Bonus III, with a big CPLD for the encryption...
And my best attempt to decrypt/decode the graphics. They are bitswapped, and need a special graphics decode. The game is working properly, so was tagged as 'imperfect graphics', at least for now...
********** EDIT 2013/03/05 **********
Just fixed the colors of Cherry Bonus III (set 3) from Dyna. The bipolar PROMs that contain the color info are 4-bit devices, and the color data is 8-bit. So I took the 4-bit data from each PROM, and combinated them to get the proper 8-bit values, that match other games from the saga. Also added the missing PROM for the second bank of graphics. See the differences:
Before (as you can see in 0.152):
After the changes:
--- 2014/03/02 15:39 ---
Well... I found that the game name is Super Star '97. Seems to be a classic asian gambling game. Here the PCB:
And after some hard work decoding the graphics and other things, got it partially running. Colors and some graphics are wrong due to the lack of data (some ROMs are under/bad dumped). Here some snaps...
See the wrong banking due to the lack of graphics...
Unfortunatelly the PCB need to be redumped.
***** EDIT 2014/03/03 22:50 *****
See what we're missing due to the lack of palette and extra graphics...(click for full-size!)
Now I'm more convinced that we need another PCB to redump these underdumped ROMs... :-)
--- 2014/03/02 03:45 ---
This is a rare dump that was laying in my disk for ages. I never spent some spare time to even take a subtle look. This is a chinese game from Bordun International. After some hard work I got both graphics banks decoded, and started to show something. Here a couple of snaps...
This is still preliminary, and the game needs a lot of work. BTW, if someone can translate the title and the settings, please contact me through the CONTACT link at the left panel. Thanks!...
--- 2014/03/01 15:15 ---
A new dump thanks to ANY. This time is a NEC V30 based hardware, with a OKI MSM6376 for 4-bit ADPCM sound samples. The game name is 'Record'. Is an Italian gambling game where you can choose the items to play with (balls, chips, cans and numbers) for legal purposes (poker games were forbidden), and also has some protections at boot that were analyzed and worked out...
With this one too, at very begining I got constants resets due to the protection. Now is fully working. Still remains to check some weird service inputs. Here you can see some snaps...
And here the settings...
--- 2014/02/26 13:50 ---
A lot of work on the MCU simulation. After exhaustive analysis, I found how is the MCU communication.
The program writes to the MCU different commands for different tasks, and get a response from it. Some commands have different parameters. Command 11h is for Idle, and is just one byte. Other commands take till 4-5 bytes long (command + data). In the case of reels controls / state, the command is 1Ah. Just writes to the MCU a 6-bits value (bits 0-1-2-3-4-5), and expects a 3-bit response (bits 0-1-4) from it to define a reel state/graphics to generate. Any different combination produces reset or corrupted graphics.
So, here my first attempt to simulate the MCU. Also you can see the button-lamps working... Here see some snaps:
And the Mistery/Gyro/Roulette/Wheel bonus... :-)
The game now seems to work more close to the real thing. I'll try to go further, and find the relation of the injected seeds with the MCU response.