Odd Asteroids fault

[guddler]

This is one I've not seen before - anyone else?

Initially the board wouldn't boot or do anything. I fixed that, it was a bad LS14 right up at the top of the reset chain near the reset button.

Now it plays and the vectors are totally screwed - nothing new there, but...

It slows down most noticeable by the sound but only, i think as the screen is so screwy i can't really see it. Every now and again it goes so bad that it resets but not generally speaking. The clock frequencies seem perfectly ok. Not spot on down to the last digit of precision according to the fluke but they rarely are, and most certainly don't need to be (you can use a 12Mhz crystal instead of a 12.096Mhz quite happily for instance).

Wow!

I would say any ideas but hopefully I'll have it fixed in a short while

 

[trm]

Do you mean it's slowing down as if you had a slow-mo button like the old Action Replay carts that used to strobe the 68k halt line to control the game speed and it's just decreasing in speed until eventually you could probably see each vector being drawn? That sounds like a pretty cool fault. If you manage to get a good display with the fault present then I'd love to see a short clip of it

Good luck!

 

[guddler]

More of a defender-like slowdown when it can't quite draw enough quick enough. It's not though, Asteroids simply doesn't work like that it'd crash before it did that.

Which is now what it's doing

I've given up for the night now! At least now that the DVG side of things is resetting the board I know I'm looking at a fault in the IC department rather than the analogue output stage. Well, at least! I'm not 100% convinced that there isn't an analogue problem too as I'm getting what I think are odd readings from the +/-8.2v that supplies the sample and hold control circuit. I've trouble shooted that though and it seems fine to me.

So I need to start afresh with a clean mind another day and see what's crashing the state machine side of things now. It's displaying a fair bit before it crashes, but it does crash in exactly the same place all the time so maybe it's something around the bit registers now? Dunno, head's fried at the minute.