MCR/Tapper Random Sprite Images

KI4SWY

Well-known member

Donor 5 years: 2013, 2020-2023
Joined
Aug 18, 2009
Messages
2,017
Reaction score
229
Location
Brandon, Florida
OK... Got a good one here..

Board worked perfectly fine, and then decided to show customer, and well, all the sudden it got random sprites.. BG images are fine.

Swapped the 6016? on the Video Board, and the CPU Board, and even swapped the NVRAM.

it will play fine, then random sprites then fine, then random sprites.. almost like a flapping data latch in the object ram load section? any ideas where to look , getting schematics from customer tommorrow, so should be easier to source.

does not appear to be the 4 93422 or 23??? ram chips on video board.

will post video later tonight.

Service History: Got board 100% non-working, had two bad z80's, wrong ram on io/ and two dead roms.

sprite ram does test 100% good in fluke 9010 long ram test.
 
Board worked perfectly fine, and then decided to show customer, and well, all the sudden it got random sprites.. BG images are fine.
The video is helpful. Why are you certain that it's an FG problem? The BG is 32x30 blocks. It's too tough to guess from the video, but could it momentarily (like one frame) be showing the wrong block on the background? That would be my guess based on what I see.

getting schematics from customer tommorrow, so should be easier to source.
I don't know if Tapper is MCR2 or MCR3 but schematics will be standard across lots of games. Check here for a great writeup on what to expect from that video circuit.

sprite ram does test 100% good in fluke 9010 long ram test.
If you've got a 9010 you're in great shape. It looks like foreground is F000-F7FF and background is F800-FFFF. Test them both. Based on the writeup above you can figure out how to write to video RAM and put specific pictures on the screen. This will let you tile the screen with whatever you want and better characterize the problem.
 
this happened to mine right after a board repair.

turns out I wasn't getting +5 all the way down to the legs of the ram chips. It measured fine in other spots, but you have to crank it up until you get +5 on the ram legs.

That may not be your problem, but it's worth checking.
 
Looks to me like a charater ROM access problem. I don't have a tapper, but generally a 74LS138 or 74LS139 provides that access, and is probably the trouble component.
 
joey: if you pull the bg roms, its still does it, actually, if you don't run the uut, it does it, its almost like a failing or on the verge of failing latch (i am probably going to go with what hypersport states, and probably a 138/139/157) .. I am getting a good set of original print schematics today, so will be able to actually READ them, i hate staring a schematics on a PC Screen.

tron: yeah, already tested all that, the nvram goes crazy if not getting 5v, but checked all the ram/rom chips for 5v, and tested ok.

hypersport: i agree, now if the pesky chip would FAIL it would be nice..
 
Did you ever figure this problem out? I have a problem with my foreground characters being incorrect (bad position, flipped, wrong graphic, etc). I suspect the FG RAM is corrupted or not accessing correctly, so probably a different issue .. just curious what was going on with yours. Is there an easy way to get a Fluke plugged into the CPU board while it's in the cabinet?

On a side note, it's interesting to monitor this in MAME and see how many objects it uses for some of the animations. E.g. When pouring a beer, there were something like 8 objects going on (upper body, lower body, hand on the spout, beer pouring out of the spout, beer foam, 2 - 3 sections of the beer in the glass, etc).

BTW - The FG RAM range is E800 - E9FF, BG RAM is F000 - F7FF.

LeChuck
 
A couple other questions, now that I'm debugging my Tapper finally:
- Does anyone have a list of RAM chips on Tapper, and what decode ranges they each compose? I've only skimmed the schematics, but the decode logic isn't as apparent as other games I've debugged (I'm mainly wanting to know which chip contains the foreground RAM @ E800 - E9FF).

- I vaguely recall a thread saying that the Fluke RAM tests will fail on MCR games. I notice this when testing my Tapper, e.g. RAM short testing just the single byte at E800, although my game also has issues with the foreground sprites being out of whack. Does this sound familiar? Or are there other known gotchas with the Fluke on MCR games?

LeChuck
 
Back
Top Bottom