Rom Hunter mkII

Pagina 5/11
1 | 2 | 3 | 4 | | 6 | 7 | 8 | 9 | 10

Van Manuel

Ascended (15686)

afbeelding van Manuel

03-07-2019, 16:41

Manuel wrote:

Note: I assumed that writes would be in the 0x4000-0xC000 range, but perhaps that's wrong.

Forget about this, I did record all writes in the whole area of the slot. So the writes I wrote about are the writes seen.

Van Manuel

Ascended (15686)

afbeelding van Manuel

03-07-2019, 17:16

Manuel wrote:

From 0x1100-0x1268 is some code (looks like a bunch of helper routines)

The last 5 routines are calls to WRSLT with different addresses. HL contains 77F0, 7FF0, 67F0, 9FF0 and BFF0 for the 5 versions of that routine. So apparently these addresses are also relevant.

Van Manuel

Ascended (15686)

afbeelding van Manuel

03-07-2019, 17:21

Oh, I missed the last part of that code of which the sources are in the ROM image:

420B  ld     a,(ix+#01)   DD 7E 01
420E  ld     hl,#4000     21 00 40
4211  call   #000c        CD 0C 00
4214  cp     #41          FE 41
4216  jr     nz,#423a     20 22
4218  inc    hl           23
4219  ld     a,(ix+#01)   DD 7E 01
421C  call   #000c        CD 0C 00
421F  cp     #42          FE 42
4221  jr     nz,#423a     20 17
4223  push   hl           E5
4224  ld     hl,#5000     21 00 50
4227  ld     bc,#0025     01 25 00
422A  push   bc           C5
422B  ld     a,(ix+#01)   DD 7E 01
422E  call   #000c        CD 0C 00
4231  pop    bc           C1
4232  cpi                 ED A1
4234  jr     nz,#4255     20 1F
4236  jp     pe,#422a     EA 2A 42
4239  pop    hl           E1
423A  ld     a,(ix+#01)   DD 7E 01
423D  ld     hl,#8000     21 00 80
4240  call   #000c        CD 0C 00
4243  cp     #41          FE 41
4245  jr     nz,#4252     20 0B
4247  inc    hl           23
4248  ld     a,(ix+#01)   DD 7E 01
424B  call   #000c        CD 0C 00
424E  cp     #42          FE 42

So the source code in the ROM is actual code as it appears from 0x41BB to 0x424E.

Van gdx

Prophet (2978)

afbeelding van gdx

03-07-2019, 17:33

According the article and the Rom dumps with RHx extension. The RAM of Rom Hunter mkii use ASCII 8k mapper only. Roms from cartridge with another mapper are patched.

I wonder if the internal RAM and ROM are each in a secondary slot or if it is configured otherwise.

Van Takamichi

Master (229)

afbeelding van Takamichi

03-07-2019, 17:43

RetroTechie wrote:

Screw(s) covered by the label is also common. Usually you can feel this by going over the label with your fingers. If used, you have the choice between break things, damage label, or not open it up.

And yes, failed RAM chips is always a possibility. Pretty common actually. No reason this cartridge would be an exception to that rule even if unused most of the time.

Note that failed RAM chips don't prevent you from running test software & check what response you get. If eg. a RAM providing 1 bit failed, that leaves 7 bits that do behave like RAM. Garbage coming out != the usual FFh of 'empty' space. And mapper mechanisms are easy to detect as well.

Thx for the advice, I noticed seemingly two screw heads below "-" and "II" below the label on top of the cart. See if I can peel off the label without damaging. Photo No PIN.

Van Manuel

Ascended (15686)

afbeelding van Manuel

03-07-2019, 18:05

gdx: I see no writes to 0xFFFF so there doesn't seem to be a SSR. There must also be some way to switch from RAM mode to ROM mode.

Van Takamichi

Master (229)

afbeelding van Takamichi

03-07-2019, 18:50

gdx wrote:

1. The fact that the SCC Roms have a different extension may mean that the loader applies a patch to the load. For Quarth or Metal Gear 2, I think the auto-detection of Rom mapper is not effective.

2. testRAM not support the Rom mappers. A "MegaRam" is not a "Memory Mapper".

3. Do not force! There are probably clips like Konami's cartridges.

5. Try with a Konami megarom without SCC sound or a megarom with ASCII 8k mapper

I decided to test with something simple as 16KB Moon Landing from ASCII. The dump became 32KB, 0 to &H3FFF filled with FF, but still ran fine on emu. Tried to play with the utility disk and real MSX. Disk is reading, a few moments passed, and MSX rebooted Sad but instead of the Rom Hunter rom tape menu, MSX-DOS booted from the diskette in the disk. Whatever, there must be a reason all the dumped files that failed/modified from the original has .RHT extension. Files PIN: lowgravityadds16kb

Van Manuel

Ascended (15686)

afbeelding van Manuel

03-07-2019, 21:54

Note that it looks like the program does all writing via the WRSLT call. Does that even make sense?

Van Takamichi

Master (229)

afbeelding van Takamichi

04-07-2019, 11:22

Is that slowing Rom Hunter? Another followup; Rom Hunter can't dump Game Master 2. Only initial 32KB was dumped. No wonder given GM2 didn't exist at 1986. Result PIN: globalwarmingkilledpenguin I looked at other failed dumps and it looks RH could dump only the initial 32KB in all cases.

Van gdx

Prophet (2978)

afbeelding van gdx

04-07-2019, 11:46

You can manually check the RAM and Rom mapper with SHEM.

http://sakuramail.net/fswold/tool.html

Pagina 5/11
1 | 2 | 3 | 4 | | 6 | 7 | 8 | 9 | 10