A friend of mine, Daniele, gave me this board from his personal collection for a quick check-up and maintenance. As all we know CAPCOM uses on these CP SYSTEM II (CPS2) boards an anti-piracy measure protecting the game. The game code is encrypted and it’s decrypted on-the-fly by the main CPU using a special decryption key stored on a battery powered RAM. So if the battery is running out of power the decryption key is lost and the board “dies” preventing the CPU running the correct program code and obviusly the board stops working. This is why the battery who powers the special RAM is better known as “Suicide Battery”. So before the board commits its “programmed” suicide, the battery should be replaced.

Like Mr.Ozzi (a friend of mine) said, speaking about his personal arcade boards collection: “a suicide battery is like a cucumber in the a*s, the more batteries, the more cucumbers you have!“.

Here is the CPS2 board, a Japanese SUPER STREET FIGHTER 2X :

The CPS2 system is a two parts sandwich, a bottom mainboard with a black case and a top game board with a coloured case, which holds the game EPROMS and the SUICIDE BATTERY too. This game board is made in green plastic because is a japanese CPS2 but you can find them in different colours for each different region :

Region Case Version Screen
Japan Green plastic White text
U.S.A. Blue plastic Red text
Euro/Etc./World Blue plastic Blue text
Asia Grey plastic Yellow text
Hispanic Orange plastic Green text
Brazil Orange plastic Magenta text
Oceania Blue plastic Orange text
Rental (any of above) Yellow plastic (Any of above)
(Any of above) Black metal “all-in-one” (Any of above)

This table is taken from CPS2 page on Wikipedia.

Once opened the game ROM board we can see its PCB and the battery, the red one on the bottom right side, on the following photo:

Measuring the battery reveals it’s not in so good condition to power the security RAM anymore, it’s around 3.48 Volts. It’s enough to keep the game working but a new one should be around 3.6V and this is a clear signal that the battery which served this board for over 10 years il slowly depleting its power. So we must replace it, here is the new battery ready to be soldered on the PCB:

We must proceed to remove the old one desoldering its leads on the pads in the back side of the PCB:

Now there will be the “hardest” part, once removed the old battery the special RAM will remain with no power and we must hurry replacing the battery, because the battery-powered RAM will loose its contents in minutes. Some guys says 10 minutes, other guys says over 1 hour, by the way i don’t want to discover it for sure …

So if you are going to replace it tremendously really slowly you must use a helper battery for the process, but if you are a PRO there should be enough time to take a shot too … LOL !

Once removed the old battery, the new one can be soldered in its place, but be careful to install it in the proper way, the positive (+) side of the battery must go on the left side, the negative (-) side instead goes close to the edge of the PCB, like the photo below:

Solder the battery on the underside of the PCB, cut the exceding length of the leads off and make ABSOLUTELY sure that the battery is soldered properly and in the right way.

Job done … now we must test the board …

Yeah! Repeat the test and power up the board again after few hour to make sure the new battery is doing it’s job …

… if the board has committed suicide is a clear sign that something has gone wrong, but thanks to the effors of Eduardo Cruz, Artemio Urbina and Ian Court few months ago they announced the successful reverse engineering of CPS2 security protection, allowing to DE-SUICIDE and restoration of any dead games without hardware modifications.

Here is the links to their site … follow them, it’s a very interesting reading:

Eduardo Cruz (30 April 2016). Arcade Hacker: Important Capcom CPS2 Announcement.

Eduardo Cruz (13 September 2016). Capcom CPS2 Security Programming Guide.

… another case closed!


A friend of mine, Enrico, bought this boards it was sold as working but it suffers some kind of colors issue. Wrong colors are displayed on the screen, the board graphics seem to be too green and too blue. Here’s the PCB:

… and this is a video shot of the issue:

This is due to a lack of the red component in the video signal, it’s a quite common issue on mvs boards. Usually missing or incorrect colors can be caused by a damged part in the video DAC circuit. The digital part of the video output section shouldn’t be guilty this time. The parts involved causing the issue should be spotted between a pair of 74LS273 latches, a resistor network and a 74LS05 hex buffer-inverter IC.

The usual visual inspection reveals damage in a resistor in the video DAC circuit, it’s a common issue seeing this resistors failing, it’s due to Ohm’s law: the lower the resistance the higher is the current flowing through the circuit, so the resistors with lowest values are the ones which fails first:

The 150ohm resistor is burned and must be replaced, but this may be the consequence of the culprit and not the damage itself. So take a look at the video DAC circuit taken from the schematics at NeoGeoDev.Com site:

On the following picture you can also look how the RGB colors are mangled by the board:

Measuring the resistor value reports a too low value around 24ohm instead of the 150ohm needed by the circuit, but this damage should be caused by something else, so it needs further investigations.

Once replaced the resistor …

… diplaying color bars and measured the output signal intensity of the RED component with a scope the resultulting measurements reveals an unsuitable peek-to-peek value compared to other color (RGB) signals, the BLUE signal is in yellow and the RED signal is in green (LOL), it should be a perfect match:

If you look at the color bars too you can spot a slighter low brightness of the RED color signal:

Also measuring with an ohmmeter the connection between one lead of the 150ohm RED video signal resistor and ground reveals a short circuit, instead of (about) 10Mohm resistance of the GREEN and BLUE signals, this high resistance should be the internal resistance of an IC probably damaged.

Looking at the schematics only one IC can be responsible of this bad connection, so we must take care of it, it’s a 74LS05 Hex Inverter. Promptly removed from the board:

… promptly tested, ad promptly failed, the output at pin 6 is shorted to ground (pin 7) and obviusly the relative logic unit fails the test:

I replaced the IC and tested the board …. SUCCESS!

For correct color balancing i replaced all three 150ohm resistors too:

… another case closed! Enrico should be happy this time too! 🙂


It’s the round on the repair bench for an undisputed arcade masterpiece … Double Dragon!

It’s for sure the godfather of all the cooperative beat ’em up of all time. It was developed on 1987 by Technos Japan and distributed in North America and Europe by Taito.

The board was sleeping in my personal collection probably for over 20 years, until few days ago when a friend of mine Sasha asked me for this game. I “quickly” found it and i left it on the bench for testing.

The board was running fine, it can be coined up and both player controls were in working condition, except for a partially missed audio. The speech syntesizer section a couple of OKI M5205 seems working but the game is lacking of FM music, so it need further investigation and troubleshooting.

The speech FX were working so the coulprit should be before the amp stage on the audio chain of the board.

The audio section is populated by a classic arcade tandem: a Yamaha YM2151 FM synthesizer and it’s trusted and partially scratched DAC friend YM3012. Both chips were already socketed so they can be easily swapped.

I started the troubleshooting at the end of the audio chain, i put the oscilloscope probe on the YM3012 DAC i looked at its inputs and outputs. The IC was getting some kind of data on its serial inputs but the outputs was totally silent. The chip is mounted on a socket so replacing it was a few seconds job, so i swapped it for another working YM3012, ad gave it a try … finger crossed … as usually 🙂

GOTCHA! Now the board is playing music too!

Another case closed … but looking at the board something is keeping my attention high …

a DIP40 IC package marked ENC EL1200AR … mmmhhhh … interesting custom chip i was thinking, but a deeper looking at the integrated circuit reveals that the chip has a scratched/sanded surface with a rebranded label ID. Searching on google didn’t produce any useful information, also other bootleg pictures seemed not to be useful, different bootlegs were made according to MAME source. We can find almost three kind of bootleg boards running almost the same software and hardware they seemed to differs only by the SPRITE CONTROL CPU:

  • HITACHI HD63701 (same as the original board)
  • HITACHI HD6309
  • MOTOROLA M6803
  • MOTOROLA M6809

The boards equipped with the HITACHI chips share the same two PCB sandwich layout like the original one, the boards equipped by the M680x should be on a single PCB.

Dumping the program ROMS and comparing them against ROSICA gave me different matches, most of them were present in both bootlegs and original boards too. The comparisons helped me identifying the RomSet almost identical at the “ddragonb, Double Dragon (bootleg with HD6309)” present in the MAME driver except for the i2 ROM that gave me no match. The i2 ROM is the one labelled “32″ on the picture above. 

If i weren’t wrong i think i should own another Double Dragon board, and i started searching for it wondering if its PCB was made with the same layout, ad i found it. Both boards are identical, but the newcomer isn’t equipped with the unknown EL1200AR but with a scratched DIP40 ic mounted on a socket.

The markings seems to be gone, but using some water some clues can be spotted.

As you can see, superimposing a good known chip marking using Photoshop reveals a quite perfect match, the misterious IC is a sanded and rebranded Hitachi HD6309.

Now it’s time to write down a MAME driver for this ROMSET, with a special looking at the different unknown i2 ROM found.

Looking at other boards pictures on the internet reveals another big question mark laying on the bottom PCB. This board is populated by half of the EPROMS like other boards, very strange … maybe this bootleg was missing something? The board on startup is performing a self test by testing some IC and all EPROMS and they were reported as good, but as we know, we can expect the unexpected by bootleggers!

As you can see the GFX EPROMS (6) on this bootleg are 28 pin IC and according to MAME source they should be about 6Mbit graphics so we should find 12x512Kbit EPROMS, same as the original board too, or 6x1Mbit, but standard 1Mbit EPROMS like 27c010/27C1001 are in a 32 pin package and not in a 28 pin package, with this package we can find at least a 27c512 EPROM but 6 of these doesn’t have enough room for all the gfx…

These custom EPROMs/MASKROMs are marked EN4068 and they should be a 1Mbit devices on a 28 pin DIP package, i tried to read one of them on my LabTool48 Programmer using auto ID the software reports as device Manufacturer/ID 0x22/0x22, maybe it could be useful for someone out there, or maybe this information could be totaly useless too.

Reading it as a 27c512 gives a consistent read but probably only on one half of the data, looking ad the 27c512 pinout no pin is unused, maybe one of its control pin 22(G/VPP) is used as A16 to address up to 1 Mbit data. It needs further investigation … here is the pinout:

Notice: the board is populated also with some custom resistor arrays marked ENC-2.

… but now it’s the time to deep test the board! 🙂

By continuing to use the site, you agree to the use of cookies. MORE INFORMATION

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

To make this site work properly, we sometimes place small data files called cookies on your device. Most big websites do this too. A cookie is a small text file that a website saves on your computer or mobile device when you visit the site. It enables the website to remember your actions and preferences (such as login, language, font size and other display preferences) over a period of time, so you don’t have to keep re-entering them whenever you come back to the site or browse from one page to another.

How do we use cookies?
to remember users' actions, to identify the user login also with third-party cookies (facebook, twitter, google+, etc...) for social related tools, to enhance the performance of the website, they are session cookies, and third-party ones are used for google analytics tools. The cookies will not be used for any purpose other than the one stated.

How to control cookies:
You can control and/or delete cookies as you wish – for details, see aboutcookies.org. You can delete all cookies that are already on your computer and you can set most browsers to prevent them from being placed. If you do this, however, you may have to manually adjust some preferences every time you visit a site and some services and functionalities may not work.