Missile Command Issue

LyonsArcade

Well-known member
Joined
Feb 9, 2006
Messages
19,367
Reaction score
3,854
Location
Rock Hill, South Carolina
Hi folks!
I'm working on a Missle Command, and I've never messed with these before. So far, I've replaced big blue, and rebuilt the A/R II board, I'm getting good voltages as measured on the pcb. When I first did this, the game came up and game me a ram error, I replaced that ram chip and the game played 100%. Then it crashed on me. I've checked all 6 eproms, they all test good, and I cleaned and reseated the cpu and pokey.

The issue I'm getting is this: the game plays, but the area that the title appears in (MISSILE COMMAND) stays on the screen, as the bombs blow it up. It never completely goes away... the missiles blow it all to hell, but little fragments stay all over that part of the screen. If you put it into test, it stays there and says rom/map/ram are o.k... if you go back out into the game, it's STILL there. It changes slightly as the missiles blow away bits of it, but some of the explosions stay there, etc. If you start a game, it stays there too. So whatever is supposed to blank that out I guess is messed up. Any ideas?

First thing I'm going to do is replace the sockets on the cpu/pokey, any other suggestions?
 
I've seen low +5v cause "streaking" whereby pixels remain on the screen after
things move across the screen for example. Around 4.6v or 4.7v this can happen
but it sounds like your situation is much worse. But it might be worth a quick
check to measure the +5v at the board to see what you are getting. Otherwise
I would probably suspect something in the MADSEL circuits. The testmode MAP
test of that is a very VERY simple test and thus is not much of an indication it
is really working.
 
Hmmm. MADSEL? I'll check that out Braze, thank you. My 5v is dead on, as measured on the board and at the Dram as well. I've been ruling out the DRAM since it passes the test, and the chips immediately before it, and replacing the stuff in the "DRAM picture output enable" and "picture bit converter" circuits... I was just hypothesizing that since it's only on the top side of the screen it has nothing to do with the 3-color... and I didn't check the MADSEL stuff because I thought since it writes as well, maybe that's what the DRAM check tests?

You know more about it than I do though, so I'll focus on that area next. Thank you for the suggestion!
 
I finished replacing the chips in the 'picture output enable' and 'picture bit converter' circuits, and replaced the 74ls163 that the MADSEL originates from... any other ideas?

To better explain what it's doing, there's a section of the screen, in about the same place that the bricks are in 'breakout'... when you start the game, that area of the screen already has characters and trash in it. The "MISSILE COMMAND" title comes up over it, and then the explosions start tearing everything up. Not all of it goes away, some of the explosions stay on there, along with the title, and along with parts of the original trash that was already there. Then the high score table comes up over that. The missiles tear all of that up, but as long as it's turned on there's pieces of trash, title, explosions, and high scores in that area of the screen (it's from about halfway up, up to a few lines under the scores). Everything plays fine and sounds fine, and it passes the three tests ram/map/rom. I'll keep looking at that MADSEL circuit, though. There's one chip that does MADSEL_____ maybe that's the culprit?

Can I thoroughly rule out the dram since the game is booting playing and passing the dram test?
 
Im not an expert MC boards, but from when I was doing the MC multikit
I had to sort of dig into this a little. I dont have any specific things to
try. IIRC MC has 16K bytes of DRAM and the madsel circuit is part of what
allows it to be accessed in multiple ways, 8-bit, 2-bit and 3-bit accesses
with the 3-bit section being for the scrolling bottom lines. The testmode
DRAM test is only going to test it using 8-bit accesses and doing simple
patterns that are going to have aliasing issues. I think the MAP test does
a very simple 2 and/or 3 bit access, but it might only do one access. I
always wanted to write a full map test but never got around to it :(
The MC board then by decoding special 6502 instructions on the fly creates
a 64K overlay on top of this 16K making it easier (quicker) for the software
to manipulate 2-bit fields of the dram using 8-bit instructions. That is how
the game turns on and off pixels etc. That is why I was suggesting maybe
to look in that area as it sounds like some either dont get updated or the
wrong ones are being turned on. They really maximize how they use that
16K, part of it is for gameplay data (stack, variables), most of it is for
the 2-bit color video ram, and then they squeeze 3-bit color out of the
bottom lines basically using about every single bit of the 16K. Its one of
the more complicated boards I have worked with.
Dick Millikan fixed my boards for me, maybe he has some insight.
 
Over in the Address decode circuit, I'm getting a strange reading... everything coming out of the decoders is pulsing, except the 4 outputs of E8. The 4 outputs are Out0___ , COL RAM _____ , WDOG _____ , and IN TACK _____ . All 4 of those aren't pulsing, but are high. I swapped E8 and got the same thing, and also swapped C4, the input from C4 to E8 stays pretty high too, the other three inputs (Ba8, Ba9, Ba10 all pulse). No change after swapping those chips, does anybody know if I should see activity on those 4 outputs from the decoder?

Thanks
 
Dug up this 7 year old thread because I have what seems to be the exact same issue with this MC board based on the OP's description....and no record of it being solved. I've been on this board for over a week and I'm sick of it! I feel like I've checked every chip on the board, can't find any stuck pins, logic comparator is coming up clean on all the chips I've compared. Board is immaculate, no broken traces, can't find any crossed lines, replaced all the RAM just in case. Sockets are good. The way the RAM is shared also makes it difficult to mess with video signals without resetting the game.

Has anyone in 7 years run across this issue on an MC board, or at least know what part of the video circuit handles this particular portion of the video? Horizontal lines appear in that affected area at first boot...portion of the MISSILE COMMAND logo remains in the affected area until blown away....explosions only during the title screen also remain if they are in the affected area when it transitions into demo game mode. Horizontal lines can be blown away with explosions in demo mode, and will never reappear until a full power cycle happens....soft reset will not bring back the horizontal lines.

Thanks!
 

Attachments

  • IMG_2271.jpg
    IMG_2271.jpg
    339.7 KB · Views: 28
  • IMG_2272.jpg
    IMG_2272.jpg
    288.3 KB · Views: 33
Last edited:
Man I wish I knew what to tell you, to be completely honest I don't even remember making this post! I've fixed so many games over the years (but only a couple Missile Commands) I don't remember this particular issue. I never actually fixed it, though, because I distinctly remember buying a repaired board here on KLOV for the machine years ago, then sold the cabinet.
 
I wonder if it could be pokey related? I also had a malfunction close to this on my MC but mine was fixed by reseating chips and replacing the big blue. Sounds like you did that. FWIW -- before I finally fixed mine I found and bought a second working board for my machine from another member here. (I have since sold the cabinet and the original repaired board went with it). If you need one I and few other members have working MC boards on the "for sale" section. Hopefully someone has a solution.
 
I've fixed it. Here are the details in case someone else runs across this issue. Some of this may not be 100% accurate, but this was what I encountered.

The 7442 @ N2 has pins 2, 3, and 4 coming out and all going into a triple AND gate. Normally all three of those pins stay HIGH about 99% of the time. Very rarely, particularly when the game is transitioning from one video mode to another (for example, going from the "MISSILE COMMAND" title into the attract game mode) it will pulse out very quickly on one or more of those lines in order to enable the RAM to clear the video data back to zero and erase anything that was drawn on the screen.

I know I checked this chip with my logic comparator early on in the repair process, but I never saw an error because pin 3 was defective but still pinned HIGH. Because HIGH is normal 99% of the time, this would not register a fault on the comparator, but if I waited a long time for a transition and be careful not to blink, I saw a quick flash of the pin 3 LED on the comparator. The blip was also so quick, I never even saw it on my o-scope.

So for my issue, it was pin 3. Perhaps for another horizontal band of the screen it would be pin 2 or pin 4, but that is just speculation.

Anyways, replaced the 7442, and the problem is no more. Hope this helps someone in the future.
 
Thanks. Sounds like a hard to diagnose problem. Nice job in figuring it out, and even better for posting a solution. Worth a Rep.
 
Back
Top Bottom