Punch-Out corrupted graphics

MaximRecoil

Well-known member
Joined
Oct 11, 2003
Messages
3,105
Reaction score
86
Location
Maine
punchoutcorruptedgraphi.jpg


I know that the problem is somewhere on the CPU board (CHP1-01-CPU), because I've swapped its CPU board onto a fully working Punch-Out boardset, and the corrupted graphics followed it there.

The corrupted graphics are the only issue; i.e., the gameplay and audio are fine.

The problem is not related to any missing components either, because this boardset used to work 100% and then these corrupted graphics happened all of the sudden.
 
Last edited:
Look like an eprom issue or a decoding eprom chip issue.
Have you verified your eproms?

I have no way to verify the eproms, though I suppose I could swap the eproms on the CPU board with the working boardset.
 
As I mentioned above, I know that the problem is on the CPU board. When the problem first appeared, the boardset was a Super Punch-Out. Since SPO is the main game for my machine, I removed all of its SPO components and traded them with a working PO boardset. I then had a working SPO boardset (newly converted from PO) and a PO boardset (newly deconverted from SPO) with corrupted graphics.

There are 7 eproms on the CPU board, and 6 of them get changed during an SPO conversion/deconversion. That only leaves 1 eprom that could possibly be the culprit, if indeed it is an eprom issue at all; and that's the eprom at location 4K. So just now I swapped the 4K eprom from the corrupted graphics PO boardset with the 4K eprom from a fully working PO boardset. No change.
 
The PO and SPO boardsets use dma meaning the video memory exisis on the CPU board whose contents get copied over to video mem on the video PCB. Most likely this is an issue with the memory datapath on the CPU PCB for this memory or something with the dma circuitry itself.

I would start with checking and/or reflowing the edge connectors, as the DMA address and data path flow right through there and could cause a problem like this.
 
Last edited:
The PO and SPO boardsets use dma meaning the video memory exisis on the CPU board whose contents get copied over to video mem on the video PCB. Most likely this is an issue with the memory datapath on the CPU PCB for this memory or something with the dma circuitry itself.

Do you have any idea which component or components would be the culprit? Even if it could be narrowed down to a dozen or so components after eliminating the components that it couldn't possibly be, I could just replace them all ("shotgun" the problem so to speak, since I know very little about troubleshooting logic/memory/etc. chips/circuits.

Here is a high resolution picture of the CPU board, and here is the CPU board schematic, which I've combined into one large image (it is 4 separate images in the PO manual, which makes it difficult to use).
 
Last edited:
The culprits would be the video RAM (8C and 8D), the DMA addressing logic (5C, 6C, 7C for starters) and the DMA datapath (1A). If I had to shot gun I would do it in this order:, checking between each step to see if it was fixed

Check/reflow edge connector P3 and P4. These handle all the DMA addressing/data info between the top PCB and other PCBs

Replace 1A

Replace 5C,6C,7C

Replace 6A, 7A

Replace 5A 5B 6B 7B

Don't think I would replace 8C or 8D though. The fact that your info screen and player screens are both messed but would lead me to believe it's an addressing/data problem and not the memory itself.

If all else fails, I know a guy who can fix it for ya. ;)
 
Last edited:
P3 and P4 connectors are okay (each pin shows good continuity with its respective solder joint, and all the solder joints look good).

The 1A chip is marked MB74LS374. On a Double Dragon bootleg parts board I see a GD74LS374 and an SN74LS374N. Are either of those compatible?
 
Okay, I read that the 2-letter prefix on the TTL logic chips just denotes the manufacturer, so I removed A1 from the PO board, added a socket, and pulled the GD74LS374 from the parts board and installed it in the PO board. The good news is that there are no more corrupted graphics. The bad news is that there are still some major problems. This is the result (for some reason I can take focused pictures of the bottom monitor, but I can't get a focused picture of the top monitor to save my life ... in reality both monitors are in good focus):

po1wh.jpg


So on the bottom monitor there are screwy colors, and the "Nintendo" text on the ring post is missing. The top monitor is in black & white, and there are a lot of missing graphics. Here are some more missing graphics on the top monitor from other screens:

po2wa.jpg


Out of the three PO boards I have (1 converted to SPO), 1 of them had a color issue that I assumed was bad color proms, so I guess this particular board ended up with the bad color proms after I did all that swapping around to end up with 1 working PO, 1 working SPO, and 1 garbled graphics PO (which is the one that's the subject of this thread).

However, bad color proms wouldn't cause missing graphics would they? I'm guessing that not only does this board have the bad color proms, but also whatever components went bad to cause the issue that is the subject of this thread.

Should I continue on with ...

Replace 5C,6C,7C

Replace 6A, 7A

Replace 5A 5B 6B 7B


... or do the results from swapping A1 and the possible color proms issue change things?
 
Last edited:
They could, if the graphic is there but the graphic color is the same as the background color!

BTW, good job on the progress so far. I have some PO boards to check/repair so I am following this thread.


However, bad color proms wouldn't cause missing graphics would they? I'm guessing that not only does this board have the bad color proms, but also whatever components went bad to cause the issue that is the subject of this thread.
 
They could, if the graphic is there but the graphic color is the same as the background color!

BTW, good job on the progress so far. I have some PO boards to check/repair so I am following this thread.

You may be right. Either way, I just confirmed that replacing A1 did in fact fix the CPU board, because I swapped it onto the working PO boardset (remember that the corrupted/garbled graphics followed over to the working PO boardset when I did that before, confirming that the CPU board was the problem) and this is the result:

po3tp.jpg


Perfect.

So the CPU board is fixed (thank you ajcrm125!), and now I guess I need color proms (assuming that's what's causing the missing graphics and wrong colors pictured above).
 
Last edited:
I verified that the color proms are bad (and are the cause of the screwy colors and missing graphics in the picture above) by swapping them into the other board. Who can burn me some new color proms?

In case anyone is wondering, the color proms are located on the BAK board, locations 6E, 6F, 7E, 7F, 8E, 8F.
 
Last edited:
I verified that the color proms are bad (and are the cause of the screwy colors and missing graphics in the picture above) by swapping them into the other board. Who can burn me some new color proms?

In case anyone is wondering, the color proms are located on the BAK board, locations 6E, 6F, 7E, 7F, 8E, 8F.

I believe my Data I/O Unisite supports those, but I don't have any blank spares off hand so if you can procure some and send them to me...
http://www.atomicmall.com/view.php?id=693595&mba=0
I can program them.
 
Back
Top Bottom