DarrenF
Well-known member
I wasn't looking forward to this one, but it turned out to be pretty satisfying.
Game: MK boot
Problem: no sound
-Pulled sound code ROM and read it. It was IDENTified as being an MK audio ROM.
-Put the EPROM back in backwards (note to self: don't rely on bootleg sockets to be installed in the proper orientation)... EPROM toast.
-Programmed a new 27C020 with mks-u3.rom file and installed, to get back to where I started...
-Probed CPU address lines (with scope). Looks like CPU is just cycling thru all of address space. (Each higher address bit is just a square wave at half the frequency.) It's like it's not executing code (or rather, all NOPs).
-This is a bootleg, but luckily they more or less followed the Midway schematics for the audio CPU, RAM, ROM and address decoding, so I was able to use it (except for the chip ID numbers and locations, of course). I focused on the address decoding at first. Probing, everything looked fine. And piggybacking didn't change anything.
-Then I realized that the CPU can only address 64K, and the EPROM is 256K... so it must be banking different parts of it in and out. Back to the schems... that's what the LS374 (U16 on a legit MK) does. Piggybacked it... got sound! Soldered it in, and done.
Even though the bookleg was VERY different in appearance from the original (the sound stuff was on the main PCB, not on a seperate board), the original schems were very useful. The bootleggers didn't appear to change anything unless they thought they could do it cheaper... and in general, I'm sure they'd want to just copy the original design as much as possible.
Game: MK boot
Problem: no sound
-Pulled sound code ROM and read it. It was IDENTified as being an MK audio ROM.
-Put the EPROM back in backwards (note to self: don't rely on bootleg sockets to be installed in the proper orientation)... EPROM toast.
-Programmed a new 27C020 with mks-u3.rom file and installed, to get back to where I started...
-Probed CPU address lines (with scope). Looks like CPU is just cycling thru all of address space. (Each higher address bit is just a square wave at half the frequency.) It's like it's not executing code (or rather, all NOPs).
-This is a bootleg, but luckily they more or less followed the Midway schematics for the audio CPU, RAM, ROM and address decoding, so I was able to use it (except for the chip ID numbers and locations, of course). I focused on the address decoding at first. Probing, everything looked fine. And piggybacking didn't change anything.
-Then I realized that the CPU can only address 64K, and the EPROM is 256K... so it must be banking different parts of it in and out. Back to the schems... that's what the LS374 (U16 on a legit MK) does. Piggybacked it... got sound! Soldered it in, and done.
Even though the bookleg was VERY different in appearance from the original (the sound stuff was on the main PCB, not on a seperate board), the original schems were very useful. The bootleggers didn't appear to change anything unless they thought they could do it cheaper... and in general, I'm sure they'd want to just copy the original design as much as possible.
