nesdev.com

Trying to make myself a Hard Drivin cart. Starting out making a socketed board. What am I missing here?

hard drivin: mapper 64, 128K prg, 256K chr. seems to run ok in FCEU

using klax as a donor, no 7432 chip on board

hard drivin seems to play ok but the grahpics are half garbled. klax (64K/64K) runs fine, shinobi (128K,128K) runs fine. apparently no released games with 256K chr on mapper 64.

prg:
pin 2 (A16) to mapper pin 2
pin 24 (/OE) to GND.

chr:
pin 30 (A17) to mapper pin 29

mapper pinout: http://wiki.nesdev.com/w/index.php/Teng ... O-1_pinout

On the title splash there, RAMBO-1 is using the IRQ to switch between banks as rendering proceeds.
As far as I can tell, the top 1/3rd is CHR banks 4/5/6/7, where bank 5 contains the the "TM ATARI GAMES CORPORATION ©1990 TENGEN" text.
The next 1/3rd of the screen is CHR banks 0/1/2/3, and this time the corruption corresponds to where it's bank 1.
The final 1/3rd of the screen is CHR banks 9/A/B/C, and the corruption corresponds to where it's bank A.
Between all three, it looks like there's something wrong with the connection of PPU A10 to the RAMBO-1, although it's conceivable it's on the other side (CHR A10).

When PPU A10 is 1, CHR ROM A17 is being forced to 1.

I'm about to put this on a cart too. Just waiting for some parts to arrive.

However, that screen looks very similar to what my emulator shows with certain versions of my RAMBO-1 emulation code. And I haven't been able to figure out why it doesn't work. I wonder if it's using some variant of that chip...

_________________
get nemulator
http://nemulator.com

If relevant (may not be, someone may want to check): there was a recent discussion and discovery that the RAMBO-1 mappers' IRQ counter works different than previously thought. It was previously assumed to work like MMC3 (timing-wise), which is not the case.

http://wiki.nesdev.com/w/index.php/RAMB ... _operation

Edit: Never mind, James was the guy who did the timing investigation work. :-) Thread: viewtopic.php?f=3&t=10439

I checked my board over and of what I can see ppu a10 and chr a10 seem fine. part of the traces run under the rambo chip but I can't imagine they would have been damaged during teardown and assembly. I was careful removing the old chips and installing sockets as I have already mangled one board working on this. Also other games I've tried (klax, shinobi) run fine. An issue with the a10 lines would manifest there as well, right? I meant to try the other few games tonight but didn't get around to it.

I'm pretty well out of my depth as far as the technical side here, but I'll try whatever you guys can come up with for potential fixes. Thanks for all the help.

Is there any chance PPU A10 and CHR A17 are somehow connected?

I don't know much about the hardware side; what I can say is that if you emulate "whenever PPU A10 is 1, CHR A17 is forced to 1", you get the emulator screenshot I posted, which looks EXACTLY the same as the screen you posted.

Whatever the source of this defect, it would not cause a problem with Klax or Shinobi because the incorrect data is in the CHR A17 line, which they ignore.

Physically, not anywhere that is visible to me. It's possible they are tied under rambo chip where I can't see. I'll take the rambo of my mangled board tonight and see if it's manufactured that way.

I was just thinking "use an ohmmeter / continuity tester / diode tester"...

oh yeah, uh, that'd be a lot easier.

so thanks to this advice, I got to poking around and figured it out. The rambo pinouts as detailed on the wiki for mapper 64 is wrong, or more likely swapped with the mapper 158 variant (I do not have alien syndrome handy to verify). chr A17 is actually pin 22 on the rambo chip, and pin 29 is apparently CIRAM A10. I moved my jumper wire for A17 and everything is a-ok now. For anyone who stumbles across this looking for instructions for this or general tengen repro help, my wiring is:

klax board, no 7432 chip
prg pin 2 (a16) -> rambo pin 2
prg pin 24 (/OE) -> GND
chr pin 30 (a17) -> rambo pin 22

as a side note, could someone tell me what the two pieces on the board in the green rectangles are? I think the one is a 0-ohm resistor, but I'm not familiar with the other. Assuming that is a 0-ohm resistor, is that the same thing as running a wire between those two points?

Thank you everyone for your help and input. This was driving me nuts.

Easy thing first: the thing that looks like a resistor with a single black band is exactly that, a 0-ohm resistor. (black = 0). And yes, it's equivalent to a jumper, but the package can help prevent shorts, especially in boards without a solder resist like this one.
The other one isn't immediately obvious, but it's probably a capacitor, especially if its two sides are Gnd and Vcc.

Anyway, have I updated the wiki page correctly?
By the way, any chance you'd be willing to test which lockout pins are which on the RAMBO-1?

thanks.

looks right for #64. can't confirm #158 but I'd say that's a reasonable assumption until someone says otherwise.

for testing, do you just want to know which pin on the rambo goes to which pin on the cart edge?

Yeah, that's plenty.

rambo 15 -> cart 34
rambo 16 -> cart 35
rambo 17 -> cart 70
rambo 18 -> cart 71

also rambo 19 is on the same trace as 20, so it's GND for sure.