BattleZone - green dot

BZ's will give a long continuous tone when the VG is bad.

This isn't a documented thing, it's just how they behave. But it can lead you to think it's a ROM issue, when it isn't.

I would dig into the VG. However there isn't a straightforward method for debugging it, as it's one big loop. So you sort of have to feel your way through it, especially if you don't have something like an FPGA Catbox.
Thanks Andrew, is this something I can do with a scope? Have you got a strategy to recommend e.g. following signals through the various ICs or testing outputs of multipliers, decoders etc with a logic probe to identify a dead chip?
 
I usually just use a logic probe to start. And I'll piggyback chips that I suspect might be bad. I rarely use a scope.

Did we determine the board isn't watchdogging? Is it playing blind?

You can lift pin 11 of the LS139 at K5 (VGGO) to disable the VG. If you do that, and it plays blind, the issue is most likely in the VG.

You just have to probe around and figure out what sections are working and which aren't, then try to narrow it down. But as I said, it's one big loop, so you can't trace the logic linearly. Many times you'll see large sections of chips be inactive, if there's a bad chip anywhere in that loop. But you can start by looking for activity at the X and Y DAC's, and tracing backwards from there.

I'll usually start by probing there, as well as the Vector Timer, and State Machine sections (see the schematics), and get a big picture view of whether the whole VG is dead, or if any of it is working. Then take it from there. But debugging the VG is more of an art than a science, and it's one of those things you get better at with experience, the more of them you do. It's not something there's a simple flowchart for.
 
you can eliminate sections that are working. on an asteroids I disable watchdog and run the game . if I see stop signal pulsing I know that the state machine is working since it is trying toi draw lines. if the watchdog is not disabled then you would see items appear to work because they would be resetting and instead of operational. halt should be pulsing with watchdog disabled if not then you have a failure in the vector generator.
 
I usually just use a logic probe to start. And I'll piggyback chips that I suspect might be bad. I rarely use a scope.

Did we determine the board isn't watchdogging? Is it playing blind?

You can lift pin 11 of the LS139 at K5 (VGGO) to disable the VG. If you do that, and it plays blind, the issue is most likely in the VG.

You just have to probe around and figure out what sections are working and which aren't, then try to narrow it down. But as I said, it's one big loop, so you can't trace the logic linearly. Many times you'll see large sections of chips be inactive, if there's a bad chip anywhere in that loop. But you can start by looking for activity at the X and Y DAC's, and tracing backwards from there.

I'll usually start by probing there, as well as the Vector Timer, and State Machine sections (see the schematics), and get a big picture view of whether the whole VG is dead, or if any of it is working. Then take it from there. But debugging the VG is more of an art than a science, and it's one of those things you get better at with experience, the more of them you do. It's not something there's a simple flowchart for.
Thanks Andrew, that's very useful to know. The game is actually playing blind - the player start button flashes and when pressed I get all the game sounds and the controls are responsive I.e. moving the levers produces the correct engine responses, fire button fires, I can hear the enemy tanks firing, flying saucer etc and the game ends when shot.
The debugging sounds a bit daunting, but I will take your advice and give it a go..!
 
if it is not watch dogging you are in the end section of the vg side, counter section and data be transmitted to dacs and output. most of that stuff is straight forward. would work from output back to counters though you will see a screwed up current signal between dac an amp because your tools can't read it
 
if it is not watch dogging you are in the end section of the vg side, counter section and data be transmitted to dacs and output. most of that stuff is straight forward. would work from output back to counters though you will see a screwed up current signal between dac an amp because your tools can't read it
Just a thought, but because both X&Y are affected, is it more likely to be a component that is common to both e.g. the DAC at C9 or the LS273 at C8?
Another thought is that because the test mode is reporting a bad vector generator ROM at B/C3, but the ROM and socket are tested OK, could it be a faulty component in the Vector Generator Address Selector or Stack and Program Counter sections?
I guess that is exactly what you are saying - start at the outputs and work back via the DAC to the counters 😊
 
have you hooked a scope in xy mode to the game? I have reread your post and never seen that info. your dmm should read low on the output because it is a signal that goes plus negative and a dmm averages it. you never used the right tool to analyze the output signals according to your posts. video signals are analog not digital and can not be read with a logic probe.
 
Right.. I forgot about that.. So.. if you pull a RAM, it goes back and beeps properly.. understood.

I've had bad ROM on BZ's that corrupt the other ROMS.. You can try pulling all but N1 and B/C3 (if I have them right) and see if that changes the beep or allows something to get to the screen.. Maybe 157's?

View attachment 801561
I had that issue with my BZ board, where one ROM was taking down a few others.
 
have you hooked a scope in xy mode to the game? I have reread your post and never seen that info. your dmm should read low on the output because it is a signal that goes plus negative and a dmm averages it. you never used the right tool to analyze the output signals according to your posts. video signals are analog not digital and can not be read with a logic probe.
Hi, yes I've connected my scope to the XY test outputs in XY mode and just get a dot in the centre of the screen the same as the monitor shows. I haven't used a logic probe yet, as I haven't had time to start troubleshooting, but what you say makes total sense.
 
interesting both x and y are centered. I do not know that the output would be if you lost the + and minus 15 volts that feed the amp from the dac. I would follow the xy signal back till I get to the first amp after the dac
 
one other idea is that the vg side is not starting at all which is why you have no output.

the more I think about it I would test to see if you have any action on the stop signal in the vector side. I don't have the schematics in front of me and can not remember the what halt is set to to set the flag that the vector side has started. it should at least be toggling if the vector side is working
 
Last edited:
Back
Top Bottom