It is currently Thu Aug 22, 2019 3:14 pm

All times are UTC - 7 hours





Post new topic Reply to topic  [ 84 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next
Author Message
PostPosted: Sun Jul 14, 2019 5:43 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 8533
Location: Seattle
rainwarrior wrote:
Finally, one thing I don't think has been mentioned is maybe check the remaining pins of the 74377 and see if they connect to any of the card edge pins of the PCB (the gold fingers on the edge, both sides).
He found that D0-D3 and D5-D7 were connected to the PRG ROM data lines; I'm comfortable saying that all eight lines are connected and in order despite D4's conspicuous absence.

Quote:
but maybe they're in scrambled arrangment here?
Not obviously. He found Q4 connects to CHR A13, and the few visible traces in the photo of the underside connect from Q5 to CHR A14 and Q6 to CHR A15. I can't say about Q0-Q3 or Q7.

The bad dump he sent me included the reset stub (x: lda #$11 / sta x+1 / jmp $8009). The code at $8009 is a credible init routine, so I have to assume that this bad dump is of PRG bank 1, not PRG bank 0.

I have no idea why the Kazzo isn't working; this is pretty clearly the Color Dreams hardware. But as I linked earlier, a previous person trying to dump a Wisdom Tree game had similar problems.


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 5:09 am 
Offline

Joined: Fri Feb 24, 2012 12:09 pm
Posts: 972
A list with all twenty 74ls377 pins would be helpful, showing their function, and their connection to the cart edge or rom chips. Using colored dots for some random pins here and there instead of making a schematic isn't so professional. With that technique it's easy get lost, for example, 74ls377 pin1, is the connection already known, or was it even considered to be connected anywhere yet?

Btw how are dumping tools handling bus conflicts? Are they searching for matching rom bytes, and then write to that address? Or are they just writing by force, hoping to have more amps than the rom chip?

Ps. Using a multimeter works as so: Hold one probe to the pin that you are looking for, then slide the other probe across the cart edge contacts until it makes beep, that shouldn't take hours when searching for only 20 pins.

_________________
homepage - patreon


Last edited by nocash on Mon Jul 15, 2019 10:28 pm, edited 1 time in total.

Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 11:52 am 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 7565
Location: Canada
Just in case anyone reading this thread needs it, the wiki has pinout reference and other information for:

nocash wrote:
Btw how are dumping tools handling bus conflicts? Are they searching for matching rom bytes, and then write to that address? Or are they just writing by force, hoping to have more amps than the rom chip?

lidnariq's script posted above tried to position the writes exactly on a bus conflict table that appeared in the single-bank dump that I presume was shared privately.

When I wrote CopyNES dumping scripts for bus-conflict mappers I would have it search the ROM space for a matching byte. I dunno how easy that is to do with anago/kazzo though.


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 12:24 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 8533
Location: Seattle
nocash wrote:
Btw how are dumping tools handling bus conflicts?
Kazzo scripts assume that the underlying Atmega16 wins all bus conflicts. Prior evidence has insinuated that it's true.

rainwarrior wrote:
I dunno how easy that is to do with anago/kazzo though.
Anago/Unagi don't obviously expose a way for the embedded SquirrelVM to programmatically read a byte and use its contents.


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 4:36 pm 
Offline

Joined: Fri Dec 08, 2017 5:12 pm
Posts: 36
Ok. So from this post (viewtopic.php?f=10&t=19093&start=15#p240720), it looks like CHR-A14 is hooked up to Q5, CHR-A15 (VPP) is connected to Q6, and /ROMSEL connects to CLK.
Also, /WE is hooked up to something. CPU R/W looks like it's in about the right place for that, and that would be expected for Color Dreams.

From this post (viewtopic.php?f=10&t=19093&start=15#p240753), we also have D0=D0 through D7=D7 except for D4. That is almost certainly hooked up as well, as there is a trace headed that way.
We also have CHR-A13=Q4. It also sure looks like there is a trace connecting PRG-A15 with Q0.
If the /WE, D4, and Q0 are confirmed this is exactly Color Dreams with support for 64K PRG/CHR (does grounding the VPP line on the CHR-ROM matter? Would require using "| 0x40" if so). Q1, Q2, Q3, Q7 would be expected not connected in this case.

So what's going wrong? A couple of theories:
-First, I'd check the mirroring flag. I believe the Kazzo only tests/and sets the mirroring flag if the dumping scripts specifies this. It looks like vram_mirrorfind is set, so it should be ok, but it's worth checking). From the pictures posted, it looks like PPU-A10 is connected to CIRAM-A10, so vertical.
-Next, is it possible both banks of the PRG have the menu code in them? Maybe the version lidnariq has is using one bank (maybe dumped with NROM script?), but the version being dumped with his modded script is starting (after the 0 write 0x8000 or maybe after completing the script once) in the other bank. This would mean the ROM values he recommends might not match for the bank that's currently active, causing the bus conflicts to be causing issues.

My suggestion is to first check the mirroring flag. And my second suggestion is to make sure that CRCs match for the existing dumps. When dumping with the Kazzo, it always gives the CRCs of the PRG and CHR. Make sure to watch them. If two dumps have the same values, you don't need to test the new dump again as it will be the same. If they're different, however, they need to be examined closely.


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 5:36 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 8533
Location: Seattle
He's not using an NROM script. He's using the Color Dreams scripts ... and also the AOROM script ... and it's the local GUI ("Unagi") that's saying "hey, I've got a file that clearly repeats over and over; there's only 32KB of PRG and 8KB of CHR here". The thing he sent me privately is this thing.

Somehow, the bankswitch writes aren't happening, or if they're happening the latch is immediately reverting to some constant value. The resistors and LEDs would help figure out which, but regardless it doesn't suggest a way forward to get a valid dump. Maybe it's a bug in the Kazzo firmware, but if so, why are Color Dreams PCBs the only mapper that gives us this trouble?

Overpowering the outputs of the latch would be a way forward, tying Q0, Q1, and Q4 to +5V or ground. I wouldn't want to do it long term, but doing this for long enough to run Unagi four times should be ok enough. And the only thing it might damage would be the 74HC377 anyway, which is a commodity part and replaceable without much cost.


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 9:14 pm 
Offline

Joined: Fri Feb 24, 2012 12:09 pm
Posts: 972
Rhetorically, yes. But does the cart owner have a glimpse of an idea about pin numbers or pin names like Q0 or Q1? Currently it might be better to stick with colored dot language, like "wire green dot to purple dot", or whatever.
Or somebody take the time to make a list, I am surprised nobody did that yet, alike...
Code:
74LS377.pin1 (color this) (/WE)      ------- unknown connection ????
74LS377.pin2 (color blah) (pin name) ------- cart.edge pinN (pin name)
etc.
Maybe that would help to clarify that the pins have actual functions, and that it is important to know where they connect to, and perhaps the question marks would motivate the cart owner to find out where the pin connects to. You know, like "oh, man, I had never thought about that pin! Haha. I will check that right now!"

On the software side, it is known how to map the PRG bank for the init/menu stuff. But is it also known how to map other (undumped) banks? If /WE connects somewhere to address bus then it might work only for certain addresses.

_________________
homepage - patreon


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 9:59 pm 
Offline

Joined: Wed Jul 10, 2019 8:17 pm
Posts: 35
Are you guys sure it isn't possible to like romhack the game to make the graphics work?
Maybe what you have to do to get it to "look right" is change some values in the hex editor, unless the file size being inaccurate means that a lot of code was left out from the initial dump?


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 10:22 pm 
Offline

Joined: Fri Feb 24, 2012 12:09 pm
Posts: 972
Dumping depends on wheter you answer open question. For that super simple board, I think that dumping isn't a problem technically, it's mainly a problem about communicating questions about pin connections to somebody who isn't experienced with such questions.
Instead of that... filling missing bits in a hex editor, if you want to rewrite the game from scratch up and then manually key in the code in a hex editor, yes of course that is possible, but is that really what you are up to?

Edit: Thinking about it, I guess 74ls377 pin1 connects to R/W on cart edge?
Edit2: If you were referring to the mirroring flag, yes that would be edited via hex editor.

_________________
homepage - patreon


Last edited by nocash on Mon Jul 15, 2019 10:45 pm, edited 1 time in total.

Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 10:44 pm 
Offline

Joined: Wed Jul 10, 2019 8:17 pm
Posts: 35
Rewriting the entire game from scratch? I'd like to keep my sanity, thanks.
Is it possible that a different, more up to date NES dumping device would dump the game better? Kazzo devices have been out of date for about a year now


Top
 Profile  
 
PostPosted: Mon Jul 15, 2019 11:24 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 8533
Location: Seattle
ShadowMan44 wrote:
Are you guys sure it isn't possible to like romhack the game to make the graphics work?
You're talking about recreating 3/4 of the graphics data from scratch. You have 8KB. There's a 32KB ROM. While this process of inventing things is pedantically a ROM hack, it's almost certainly not what you want.

No, you'll have to figure out some other way to get the contents of the ROMs.


Top
 Profile  
 
PostPosted: Tue Jul 16, 2019 10:23 pm 
Offline

Joined: Mon Jan 23, 2012 11:27 pm
Posts: 168
i have this cart and dumped it with my copynes. i used the mapper 11 script for color dreams/acgi.

here's the famirom info:
Code:
Cheetahmen The Creation.nes
iNES ver.: 1.0
 platform: NES
   mapper: 11
mirroring: Vertical
 prg size: 64KB (512kb)
 chr size: 32KB (256kb)
  prg ram: not specified
  battery: no
  trained: no
TV system: NTSC
boards: (Color Dreams)
CRC32
      prg: A0BB3588
      chr: ADCC1DE8
     data: 9AC319E5
     file: 0D7C08E8


Attachments:
ctc rom info.jpg
ctc rom info.jpg [ 78.92 KiB | Viewed 478 times ]
ctc tile.jpg
ctc tile.jpg [ 31.96 KiB | Viewed 478 times ]
Top
 Profile  
 
PostPosted: Tue Jul 16, 2019 10:41 pm 
Offline

Joined: Wed Jul 10, 2019 8:17 pm
Posts: 35
FrankWDoom wrote:
i have this cart and dumped it with my copynes. i used the mapper 11 script for color dreams/acgi.

here's the famirom info:
Code:
Cheetahmen The Creation.nes
iNES ver.: 1.0
 platform: NES
   mapper: 11
mirroring: Vertical
 prg size: 64KB (512kb)
 chr size: 32KB (256kb)
  prg ram: not specified
  battery: no
  trained: no
TV system: NTSC
boards: (Color Dreams)
CRC32
      prg: A0BB3588
      chr: ADCC1DE8
     data: 9AC319E5
     file: 0D7C08E8


Wait, the rom is now dumped!? Oh man, I wasted like 200 dollars on a shitty bootleg cart for nothing!
At least we have a rom of it now.


Top
 Profile  
 
PostPosted: Wed Jul 17, 2019 12:08 am 
Offline

Joined: Mon Jan 23, 2012 11:27 pm
Posts: 168
if it makes you feel better, i might not have thought to dump it without this thread. i picked mine up secondhand and i think this is the first time i've ever booted it up.

it'd still be good to figure out how to dump yours to verify. maybe the script needs to be fixed or something.


Top
 Profile  
 
PostPosted: Wed Jul 17, 2019 12:12 am 
Offline

Joined: Wed Jul 10, 2019 8:17 pm
Posts: 35
Post the exact script that you used to dump yours, and if that doesn't work then we could confirm that it's the Kazzo's fault.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 84 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group