Kazzo USB rom dumper / dev cart programmer

[Werrock]

No problem, yes it is fantastic, I just need to dig deeper to find a new challenge for lidnariq.

[lidnariq]

Heh. Did you ever try using the dumping script for reading from the 05-00002-010's ASIC registers? Did it even run?

I did encounter another funny/annoying thing when I was writing up the findings: There are five games that used the 01-22000-400 PCB (according to the glukvideo web site). Two with the ASIC, namely Policeman and Strike Wolf, and three without (Puzzle, Volley ball, F-15 City War).

But the latter three are supposedly emulated as mapper 79.

F-15 City War does have at least three different builds, courtesy No-Intro:
* http://datomatic.no-intro.org/index.php ... =45&n=0683 (v1.0, GoodNES says "AVE v1.x")
* http://datomatic.no-intro.org/index.php ... =45&n=0684 (v1.1, GoodNES says "AVE v1.1")
* http://datomatic.no-intro.org/index.php ... =45&n=2699 (Gluk, not present in GoodNES)

The two builds present in GoodNES don't write to $4200 at all ...
edit: Ah, but I found a dump of Gluk's release somewhere else and it does write to $4200 ... as well as to $4120. Mystery solved, I guess.

[Kreese]

Yesterday I recived my unit! Today I will try it out. Excited!!!

First question, will there be any drivers for Windows 10? Now I had to bring out an old laptop, and will try it with that.

Edit: Worked fine. Tried a N-rom 256 cart!

[Werrock]

I dumped F-15 using the same script as for policeman:

f15.nes, mapper 36
Program ROM: size 0x008000, crc32 0x8406ed8b
Charcter ROM: size 0x008000, crc32 0xc470fadc

It starts in the emulator but the graphics is messed up. If I only modify the mapper to 79 (from 36) the game loads fine and the graphics is ok again.

Sorry but I have not had the possibility to try get more data from the Policeman PRG mapper.

[lidnariq]

I dumped F-15 using the same script as for policeman
[...]It starts in the emulator but the graphics is messed up. If I only modify the mapper to 79 (from 36) the game loads fine and the graphics is ok again.

Yeah, that makes sense; currently there are no emulators that actually emulate what the PCB 01-22000-400 actually does.

Relatedly...

Sorry but I have not had the possibility to try get more data from the Policeman PRG mapper.

No rush. Having an incomplete but correct description is already better than the astoundingly incorrect prior documentation.

[Werrock]

I got a request to dump Pipemania fron NA forum as someone states it has not been dumped:
http://nintendoage.com/index.cfm?FuseAc ... 0&sID=1220

Anybody know what mapper this game uses? Should be the same game as Pipe V from Sachen but I cannot cofirm it since I do not own that game.

[Werrock]

Added PCB pictures, I added the tape since I always get nervous when I see EPROM windows:
http://postimg.org/image/ulv2koxpp/
http://postimg.org/image/495ydt2yb/

Very elegant design, only a few patches and some blobs of solder here and there.

[tepples]

Is Pipe Mania another port of the game that BPS ported to NES as Pipe Dream? If so, it joins Othello and Rainbow Islands as games that got NES ports by different companies in different regions.

[Werrock]

Yes the two games are very similar in gameplay.

[lidnariq]

Should be the same game as Pipe V from Sachen but I cannot cofirm it since I do not own that game.

Sachen's Pipe 5 is extremely different, with (according to GoodNES) more PRG and less CHR, and (according to FCEUX) a completely incoherent banking mechanism.

In any case, given that the PCB holds two 32 KiB UVEPROMs, this board is probably similar to CNROM. You should be able to dump the PRG as though it were NROM. I'll look at the photos some more and try to figure out the CHR banking.

edit:

Ok, I can't see all the signals, but I know it's latching CPU D0 and D1 when
* M2 high, R/W low, A13 low
* three other signals high ... probably A14, /ROMSEL, and ... A8?

If you could track down what 74'138 pins 1-3 are, that'd be sufficient to be able to dump the CHR. The board might be compatible with NINA-03, and the silkscreening mentions the NINA CIC stunner (even if it's been replaced here by a messy pile of discretes)

[Werrock]

Here is the pinout:

Code: Select all

 CPU A14 -[01 U 16]- 5V
 /ROMSEL -[02   15]- NC
  CPU A8 -[03 7 14]- NC
 CPU A13 -[04 4 13]- NC
 CPU R/W -[05 ' 12]- NC
      M2 -[06 1 11]- NC
  175P9  -[07 3 10]- NC
     GND -[08 8 09]- NC

Nrom readout:
nrom_readout.nes, mapper 0
Program ROM: size 0x008000, crc32 0xb21f8131
Charcter ROM: size 0x002000, crc32 0xebd5b46a

Sending the rom in PM.

[lidnariq]

Yeah, that's definitely just mapper 79.

Start with the GNROM script again, but replace cpu_write(d, 0x8000, i); with cpu_write(d, 0x4100, i);
and replace cpu_write(d, 0x8000, i<<4); with cpu_write(d, 0x4100, i<<3);


The funny thing about this PCB is that it supports three different mappers:
- Populating both the '138 and '175 makes NINA-03/06 (m79)
- Populating only the '161 makes CNROM (m3)
- Populating only the '378 makes Color Dreams (m11)

[Werrock]

Not fully working, I tried this and got:
test.nes, mapper 79
Program ROM: size 0x008000, crc32 0x713d51cc
Charcter ROM: size 0x004000, crc32 0xeac077bd

Game starts but graphics is messed up at intro but recovers later.

If this was 79 it should work as the F-15 game or? That game had this code:

Code: Select all

function cpu_dump(d, pagesize, banksize) {
  cpu_read(d, 0x8000, 0x4000);
  cpu_read(d, 0xc000, 0x4000);
}

function ppu_dump(d, pagesize, banksize) {
  for (local i = 0; i < pagesize; i += 1) {
    cpu_write(d, 0x4200, i);
    ppu_read(d, 0, 0x2000);
  }
}

I tried this but I got less CHR memory:
test.nes, mapper 79
Program ROM: size 0x008000, crc32 0x713d51cc
Charcter ROM: size 0x002000, crc32 0xa12ffb3e

Still plays the same.

[Werrock]

If I only change the mapper to 113 the games seems to work perfectly using your technique. You can test with the ROM I sent you.

test.nes, mapper 113
Program ROM: size 0x008000, crc32 0x713d51cc
Charcter ROM: size 0x004000, crc32 0xeac077bd

[lidnariq]

Sounds like mirroring issues again? The one you PM'd me works correctly, though, even though it's running as mapper 79 instead of 113.