Missile Command Help Wanted

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
Can anyone help me with a Missile Command please?

I've got a board on the bench that was dead. It's now working after some clock issues, and then a duff ROM but I get no yellow background in the last 32 lines of the screen.

I'm aware that the last 32 scan lines are a special area where the CPU speed is slowed by half (625Khz) and quite a bit of funky stuff goes on there related to 3-bit colour ram.

Could anyone narrow what area I should be looking at please as t all gets a bit complicated from there and my understanding of the board is not that great.

To the best of my knowledge I think the clocks are generally OK, and that the RAM is OK since all the tests pass (RAM tests on the Fluke are also OK). I'm not sure I understand the whole 3-bit, vs 2-bit colour operation enough to diagnose further.

Does the fact the yellow background is missing indicate for instance that it's using 2-bit colour when it should be using 3? There is activity on the 3rd colour bit output (into unlatched RGB signals) but it's difficult for me to know if it's the correct activity.
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
Cheers dude. Unfortunately it doesn't help at all since the only thing that's there which isn't already in the manual (or on the schems rather) is the mame source. Sadly mame source is so abstracted by a massively over-engineered macro implementation and lack of documentation that it's difficult for even software engineers like me to work out WTF it's up to! guddler2013-06-20 18:52:41
 

P-Man

International man of mystery
Feedback
4 (100%)
Credits
1,273CR
Check the phase X clock that determines where/when that bit of screen appears.

I had exactly the same problem, it was a bad 7474 @ A7 pin 5
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
Cheers Andy - I'll investigate around that. I had someone round this evening who wanted me to sort out his FPV kit so I had a chance to quickly check that specific chip (which appears to be doing what it should) but nothing more yet.
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
OK, just a quick further question.

The memory map shows 3rd Colour bit region of screen RAM from 200-5FF and 2nd Colour bit region at 6F0 - 3FFF. Both areas test fine on the Fluke but should I see the screen being written to during the RAM test of the 3rd colour region? I don't I only see anything in the 2nd bit region, but I'm thinking that the 3rd bit region may just store the info and the sync would enable the display of those bits or not.

And then hence, back to whether that part of the logic is working or not...

Sound about right? (Appreciate you may not remember all this off hand!)
 

P-Man

International man of mystery
Feedback
4 (100%)
Credits
1,273CR
You wont see anything gudd, if the game can't display the yellow then your ram test wont either.

Check the phaseX clock and what it feeds, stop fannying about with the ram.
 

guddler

Busting vectors like it's 1982!
vacBacker
Feedback
10 (100%)
Credits
4,058CR
That's the thing though, I've spent hours poring all over that phaseX clock and the extend signal and it all looks to be doing it's thing perfectly OK. I can see an overriding 1.25Mhz waveform with a 0.625 riding underneath it (if you see what I mean), so the CPU is definitely being slowed down to cater for the 3rd bit region. I'm not sure how I could tell whether that slowdown is occurring at the right time or not, but the fact it's happening at all tells me i should be seeing the colour SOMEWHERE.

In addition, I can see the signals related to the 3rd colour and they all look OK everywhere - including the 3rd colour inhibit right at the end. So I'm kinda stuck. (And it's doing my head in now
smiley36.gif
)

One thing I spent some time looking at last night is that my clock signals don't go very high in terms of voltage. The peak of the waveform is just nudging about 3.8v, yet this is the crystal circuit that's doing it. If I disconnect the output from the very first buffer it doesn't go any higher so I guess that's just the way it is - I don't have a worker to compare it to.

It's like all the timing for the 3rd colour is correct but the 3rd colour just isn't, so my final thought last night was whether maybe the PROM is OK? I believe it's used for a lookup table that governs whether 2 bit ram or 3bit ram is written and read. Maybe along with other things. But I'll admit I'm clutching at straws here!

I need a break from it for a bit, but the only other thing I really want to get going right now is BattleZone and that's like jumping from the frying pan to the fire!
 

Adrian Purser

Moderator
Staff member
vacBacker
Feedback
2 (100%)
Credits
213CR
I just downloaded the schems (while waiting for my girlfriend to get ready before we go to the beach).

Just had a quick look and I think the first thing I would check is the 166 at P6. I'm guessing that you have already checked this as it's in the area of the schems marked colour. The 3INH signal and LD3COL signals stand out. Maybe check the inputs from the DRAM when the LD3COL signal is active and then follow the signals through the 153 to the colour ram.

I found that the timing on missile command is critical and I found several faults there on my board. so much so that I simulated the whole timing circuit using verilog to generate a waveform diagram of what I should expect to see. I will try to dig it out.

getting nagged to go now so will check again later

Ade
 
Top