Help needed to defeat the daemon

[lutorkhan]

I'm quite sure that a daemon lives inside the circuitry of the Cascade PCB I'm currently fixing, there's no other explanation because I've checked and analyzed everything many times but still the PCB is faulty.

Cascade is the Sidam version of Avalanche (schematics are more less equal to the original with minor changes), a game based on the classic Atari 6502 layout.

After the usual troubleshoot (pcb was totally dead) I've got this behaviour:

- Romset is ok
- video sync prom is ok
- watchdog resets correctly
- self test is ok
- game starts

As you can see by the attached picture the screen shows the right things but it seems that there's an error in the video ram (but is not a faulty ram IC). The weird thing is that the game is fully playable except the fact that there's some sort of autocoin.

Prior to begin the "brute force" method (= replace every IC) is there anyone that can have a look at the schematics and give me a suggestion on what to check?

Many thanks for your help.

P.S. I've also tried what Vernimark calls "Arcade magic wand" (Hp logic comparator) but, in my opinion, is useful as a raincoat when you are crossing the death valley.

[italiandoh]

P.S. I've also tried what Vernimark calls "Arcade magic wand" (Hp logic comparator) but, in my opinion, is useful as a raincoat when you are crossing the death valley.

I agree with you about the logic comparator. If anyone is searching for one, please let me know and I will sell you mine.
About your game, is the clock timing correct ?

Matteo

[lutorkhan]

About your game, is the clock timing correct ?

Good question.
The H/V sync pulses appear to be correct but a slightly wrong value in the main clock could explain the problem in the ram bank selection.
Tomorrow I'll sacrifice the lunch break to check it.
Thanks Matteo

[lutorkhan]

Unfortunately the daemon is not in the clock circuitry, only five days remaining before the "brute force method" .

P.S.
Needless to say, if someone has a Cascade PCB I'll be glad to buy/trade it. Cascade PCB is very easy to be identified since both "Sidam" and "Cascade" are marked on the component side.

[speleo_de]

If your H & V Clock circuitry is working (1H-256H & 1V-128V), and I use a Scope to check these clocks usually, then you should check the LS257's left of the Video RAMs (F2,F3,H2,H3) incl the single Inverter Gate and the LS166, bottom right to the VRAMs (E5) in the schematics.

[lutorkhan]

If your H & V Clock circuitry is working (1H-256H & 1V-128V), and I use a Scope to check these clocks usually, then you should check the LS257's left of the Video RAMs (F2,F3,H2,H3) incl the single Inverter Gate and the LS166, bottom right to the VRAMs (E5) in the schematics.

Hi Andreas,

the H/V circuitry is working (two 74161 has been replaced) and has been checked with the scope, the video shifter (74ls166) has been uselessy replaced since the problem is not only in the video output.
The video address decoder circuitry based on the 257 has been totally replaced (there was a problem with just  one decoder), I've also checked the control signals for the 257 (4H, ~4H, ROW/~COL) and they are fine.
The ram  bank switch inverter (4KCS1, 4KCS2) works fine. Now you can understand why I'm feeling so frustrated 

I'm almost sure that there's some sort of sync problem that causes a memory overlap, that's why the credits are always high.
I've made a test pulling out pin 12 both @F3 and @E3 (4KCS1 signal), obviously the watchdog is activated but, sometimes, I can see the screen properly drawn.

P.S. the Asteroids cab you sold me is proudly standing in my living room.

[vernimark]

P.S. I've also tried what Vernimark calls "Arcade magic wand" (Hp logic comparator) but, in my opinion, is useful as a raincoat when you are crossing the death valley.

I gave a Farrari to a farm tractor driver (or a farm tractor to a Ferrari driver...)

[lutorkhan]

DAEMON HAS BEEN DESTROYED!

This time has been very hard, but at the end I've found the problem.
It was the OUTPUTE signal (6@J8) that was always high due to 8@J8 (stuck high), this caused fake VIDEO INVERT (6@F8) that corrupted the video.
Now video is perfect, remains only a sound issue but I'm confident.

I'm really, really happy

[Q*ris]

 

[italiandoh]

DAEMON HAS BEEN DESTROYED!

This time has been very hard, but at the end I've found the problem.
It was the OUTPUTE signal (6@J8) that was always high due to 8@J8 (stuck high), this caused fake VIDEO INVERT (6@F8) that corrupted the video.
Now video is perfect, remains only a sound issue but I'm confident.

I'm really, really happy

Congrats !

Matteo

[lutorkhan]

And finally here's the picture of Cascade up and running.

Thanks to everyone for the support.