Robocop Mame

[Leontina Heidgerken]

Visualinspection showed that -all- the filter capacitors were snapped off. (Odd that a bootleg had them in the first place.) To boot, even one of the oscillators was missing. Now seeing how the mame docs said some odd frequencies for different versions, this was a bit of a trial and error. I installed 470uf lytics in the missing spots, then applied power, and started soldering on different oscillators with the power on.

With a 8mhz oscillator I got a stable picture of garbage... I soldered the oscillator in proper then applied power again. The board started with coloured bars, then turned into garbage, rolled it a bit, then stopped. Scoping the cpus, both were trying to start, and the 6502 staid running even after the 68000 crashed. This was a good sign that the CPUs were good.

With the static garbage on screen, my next stop was checking the ribbon cables as these are notoriously bad on bootlegs. Sure enough, wiggling one of them kept changing the colours in the garbled screen.

At this point I pulled the boards apart, and checked for cracked solder joins on them. After finding none, I went ahead and doused the ribbon connectors in good, non-lubing contact cleaner, and cycled them several times till I got a nice firm connection. You can actually feel the headers gripping onto the pins more and more solidly with each cycle.

After applying power again, I got a different pattern of garbage. At this point I was sure something was killing the data bus, but wasn't able to test proper as the CPU kept crashing too quickly. But a more thorough eyeballing revealed an ls245 at the inputs that literally had a hole burnt in it's side with the socket it was in melted to the leg... fun...

After pulling the dud chip, I powered the board up, and got nothing but a white screen... both cpus seemed to be running, but the signals were rather regular on the 68000, it looked like it was running NOPs or something. Then it hit me that these things usually refuse to read code without the input buffers... so with some effort I cleaned the socket enough to get contact temporarily with a new ls245.

At this point, I started resetting the 68000 while probing the main program ram pins, and found them to be rather wonky signal-wise. There were some odd bends in the image on the scope when probing the data lines. So quickly out they came. Popped them in the TopMax, and sure enough, "damaged device or device insertion error". Bypassing the error message the rams did pass the read and write test though... but I've seen that happen with eeproms too... so to be on the safe side, I put in a pair of Hyundai ram that tested fine.

Turning on, I got the same garbled mess, but this time after a few seconds it changed again, and again, and it stopped.... then I power cycled it a few times, when suddenly on it came. Don't ask why or how, but I was greeted with the high score table, sound and everything. Then it stopped at the screen showing robocop's leg. After trying several more times, the board now booted stable, but kept stopping at the same image with the music still going.

After checking around the board with the scope, I decided to lift the leg that I bent back earlier... and boom as soon as the contact broke, the game moved on and ran properly. (Apparently that chip's lifted leg is a "feature" of the board...)

This is where I left off for tonight. The board is running stable, all inputs are working, but there's still a bit of flickery going on with the objects. Probing around with the scope I managed to find a few ram chips that effect the flickering when I touch certain legs with the scope probe. Those are going to get replaced tomorrow, which should finally give me a fully working board...

A game based on the 1990 sequel to the blockbuster film about the titular mind-conflicted robotic law enforcer. Unlike Data East's previous RoboCop arcade game, this one has convenient instant direction-changing shooting controls and co-op gameplay (where the second character is weirdly another RoboCop instead of Anne Lewis).

After starting a new game, the game checks Dip Switch 3-8 ("Game Over Message") and is supposed to send the player to a stage select if that dip switch is set, but there's a branch opcode at the start of the subroutine that effectively dummies this out. Put the following code in robocop2.xml to reactivate this feature:

The game actually tries to run a display debug subroutine every frame during normal gameplay, but the developers put an rts opcode at the start of the subroutine to disable it. Put this code in robocop2.xml to reactivate it:

Like the display debug, this is technically called every frame, but is dummied out by an rts opcode at the start of the subroutine. Place the following code in robocop2.xml to reactivate it:

Press 1P/2P Button 3 to cycle between the Cobra Assault Cannon, Bazooka, Gatling gun, and the default Auto 9. This will also give the player(s) 999 ammo where applicable. No code references the weapon debug subroutine, but the code above will replace the camera debug with the weapon debug.

All of the US and Euro / Asia sets start you off in the street, attempting to stop the crime in progress at Nick's Guns. However, the Japan set (0.11) includes two extra scenes that give a rundown on the events of the first film and lets the player go through the movie's climax.

At the start of the fight, RoboCop will use his visual systems to pinpoint the boss's weak point. The fight itself is a neutered version of Stage 4's ED-209 fight; this version can't shoot rockets.

This stage seems to have been abandoned very early in development. The only object is a pipe that breaks off and swings back and forth, electrifying the nasty sewer water when it gets too low, but the electricity doesn't actually do anything at this point.

A strange version of the nuke vat at the end of Stage 3 with outlines around the main vat and cracks. These graphics appear about 2/3 of the way through the object graphics ROM, while the final vat graphics are at the very end.

09 seems to be rejected stage music; it's worth noting that Stage 1 and Stage 2 share the same track. 15 is a more sedate piece that might have been used for a continue or name entry screen.

The first three lines were used in the first RoboCop arcade game, though the clips here are of higher quality. There is a "Murphy" clip in the first game too, though the one in RoboCop 2 is new and sounds like it was recorded with a different actor.

At the end of the first section of Stage 1, the player has to stop a pick-up from running them over. In the US and Euro/Asia versions, the game tells the player to use the joystick and the buttons, but the Japanese version displays "Hit Button" instead of a moving joystick. Despite this difference, the game behaves the same across all versions - moving the joystick and pressing the buttons will work.

A similar section in Stage 3 where the player is trying to pry themselves off a giant magnet uses the "Hit Button" prompt in every version of the game, but the joystick is exclusive to non-Japanese sets.

Version 0.11 (Japan) of the game has an additional set of scribble-like credits if you manage to beat the game with just one credit. These credits are scrolled on top of a still image of RoboCop 2 from the film that is otherwise unused.

While the images are semi-transparent in the game, the transparency has been removed in these images and replaced with black for ease of reading. Some of the names refer to parts of the staff that worked on the game; these have been filled in when possible based on the endgame credits and other sources.

I've tried all three main C64 emulators: hoxc64, WinVICE and MAME with trying to play "Robocop.d64", which requires left, right, down, fire and jump. It appears the "jump" is default programmed as the fire on controller 2. The space bar works as jump properly, but any time I try to map a controller button or another keyboard key, he always faces left when he jumps (so cannot jump right or jump and fire right, etc.). Mapping jump to a typical "up" button/key does the same thing. Anyone know why this happens or how to fix it? Obviously, left is getting triggered with the mapped jump button, but I can see no reason for this. It happens in both hoxc64 and WinVICE. I'm at work now, but can post the disk image tonight.

Also, MAME did not even play the game. Just pink and white vertical stripes after loading. I've heard it is not so reliable with many games and one cannot even seem to access the GUI key-mapping page (i.e., TAB key does not serve normal menu fuction).

Under MAME, try changing the controller from "joy" to "joybstr" under "Slot Devices" (Accessibly via 'Tab' key menu after first toggling to partial emulated keyboard support by pressing 'Scroll Lock' key). Remember to select "Reset" after changing controllers.

Thanks. I'll have to try another game to see if it works, as I still get the pink vertical lines after the second startup screen of the game. I forgot about the scroll lock. The file is posted above now if you want to see what I mean.

Also, I can load a disk through the file manager into the floppy drive and designate it as "read only" but cannot figure out how to open it by that route. RUN and LOAD "ROBOCOP2", 8,1 do not work by this method. I have to enter the "mame c64 -flop1 /path/to/file.d64" string into the DOS console.

The robocop 2 instructions refer to a console joystick and a standard joystick. On a standard Joystick up jumps but does nothing on a console joystick which has a jump button in addition to a fire button. You can toggle between console/standard joystick in the robocop 2 menu.

3a8082e126