Neo Geo MV-1: Vertical lines on sprites?

Mauricio

n00b
Joined
Feb 16, 2018
Posts
29
Hey folks,

Long-time reader, first-time poster here! I posted this at KLOV and it was suggested that I try here as well. I have searched for possible relevant threads, but nothing quite matches this, so here we go...

I have recently acquired a Neo Geo MV-1 PCB, and a Metal Slug cartridge. I tested it fine for about an hour one day, and the next day vertical lines appeared on the sprites. See the attached screen image. Here is the troubleshooting I have done:

The cartridge and slot contacts are clean.

I removed the cartridge slot PCB from the metal bracket and cleaned the white plastic connectors with blown air and some alcohol. They were somewhat dirty.

I have not opened the cartridge yet. The cartridge exterior is in excellent physical condition, and I don't want to damage the hololabel if possible. I currently do not have another cartridge to test... but I am looking to get one.

The service menu (without a cartridge) looks normal. (not surprising, as this is almost certainly a sprite / C ROM issue and the menu doesn't use sprites) Similarly, the FBI warning is fine.

Here is the interesting part: Touching a single pin (with a finger, under the board) that goes to the cartridge slot mostly clears up the vertical lines. With some continuity testing (using a multimeter), I confirmed that this particular pin is connected to pin C12 of the NEO-ZMC2 chip. See the attached composite photo, and this video:

https://www.dropbox.com/s/e5cvcvetj343q99/metalslug.mov?dl=0

Perhaps unrelated: There is a wire / solder patch on the board that connects the EDGE pin of the LSPC-A0 to (I think, but not confirmed) the corresponding EDGE pin of the NEO-ZMC2. For some reason, it is routed through an unused OR gate on the a 74LS32P (quad OR). Again this is possibly unrelated, but I am curious about it. See the attached image.

Do you have ideas for further troubleshooting or know a possible cause / solution? Thanks!

Some references:
https://wiki.neogeodev.org/index.php?title=Graphic_glitches
https://wiki.neogeodev.org/index.php?title=NEO-ZMC2
https://wiki.neogeodev.org/index.php?title=LSPC-A0
 

Attachments

  • issue.jpg
    issue.jpg
    583.8 KB · Views: 55
  • touch.jpg
    touch.jpg
    234.3 KB · Views: 53
  • patch.jpg
    patch.jpg
    516.3 KB · Views: 53

furrtek

Super Sidekicks 2
Joined
Feb 6, 2011
Posts
106
Glad it was only a contact issue.

The patch is interesting. It looks like it was a factory fix, I've seen that kind of green wire for other bodges in cartridges.

They obviously had last minute timing issues with that EVEN signal, so they ran it in the OR gate to add some delay as a fix.
Funny thing is that they realized the fuckup and incorporated two delay choices to choose from in later systems, just in case the issue appeared again.

Is it possible to use an un-annotated picture of the wire patches on the neogeodev wiki ? Credit would be given.
 

Mauricio

n00b
Joined
Feb 16, 2018
Posts
29
furrtek: I have decided to go ahead and take and process photos of the full board (with PNG transparency), since it looks like the dev wiki doesn't have PCB scans for the MV-1 yet. Have a look at the "top" file here and let me know what you think. I will do the rest (bottom + slot board) if that looks good and is useful.

https://www.dropbox.com/s/pz03sv77je5eon8/mv1_top.png?dl=0

(I don't have PM privileges yet, but sharing this way should work for now.)
 
Top