What's new
By:
Filters

Reverse engineering 161 in 1 cartridge to change Rom games

Well at least some good news - all of the chips I pulled (other than those on the daughterboards I can't test until I get replacement daughterboards) have matching IDs to those ack posted earlier!

I have ordered the daughterboards (both single and double) - I believe I have an older 161 in 1 I can pilfer another flash chip from or failing that, my original plan was to have a smaller number of games on the cart anyway

Edit: My older 161 in 1 is populated entirely with 55LV100 so I will need to reduce my game list instead and stick with the original 3 flash chips. I removed all the soccer games except Soccer Brawl, Mahjjong, quiz and Stakes Winner and I think I'm good to go once the new daughterboards arrive

Additional edit: Wow, I knew the flash chips would take a bit to program but it really is a long time

Additional edit:
There has been another casualty unfortunately - my S1 ROM will not program at all. Initially it would crash at around 25% but now it either won't start or it crashes instantly

Looking at the prices these chips sell for it's not feasible to try and resurrect this cart at all. I'll still finish it just so I understand the process, but definitely going to need a new cart
 
Last edited:
I'm now in the testing phase of the Python version of the cart compiling tool. One very weird issue: There is a 1 byte difference between the vrom file generated by the binary tool and my script, but only for sonic wings 3:
1697371110196.png

Nothing in the source stands out... is that a patch for something? The hash matches what I've got, so it must be changed while saving. Any idea?
Mine matches the input file. All other files match 100%.

Edit: Here is the code.
- It IS slower than the exe version. But most people just run this script once, so shouldn't be a huge problem
- for some reason there is a tiny difference 1-byte in the vrom output between the exe version and the script version
- arguments change slightly as I use Python's argparse. Please check the included _run_py.bat
- this is an almost 1:1 conversion of the code. The biggest differences are where the python code can be expressed shorter than the Pascal code, but I didn't spend too much time to properly "pythonize" everything. It can probably be optimised as well.
 

Attachments

  • script.zip
    6 KB · Views: 83
Last edited:
Fluffy said:
I'm now in the testing phase of the Python version of the cart compiling tool. One very weird issue: There is a 1 byte difference between the vrom file generated by the binary tool and my script, but only for sonic wings 3:
1697371110196.png

Nothing in the source stands out... is that a patch for something? The hash matches what I've got, so it must be changed while saving. Any idea?
Mine matches the input file. All other files match 100%.

Edit: Here is the code.
- It IS slower than the exe version. But most people just run this script once, so shouldn't be a huge problem
- for some reason there is a tiny difference 1-byte in the vrom output between the exe version and the script version
- arguments change slightly as I use Python's argparse. Please check the included _run_py.bat
- this is an almost 1:1 conversion of the code. The biggest differences are where the python code can be expressed shorter than the Pascal code, but I didn't spend too much time to properly "pythonize" everything. It can probably be optimised as well.
Click to expand...

The python version generates the same files for me.

Code: 
0a9c450e90313e388d3ac36915c65848  bin/ROM/vrom
0a9c450e90313e388d3ac36915c65848  fluffy/ROM/vrom

19a795da30ba3b5fa242e8a6de8249c1  bin/ROM/crom-2
19a795da30ba3b5fa242e8a6de8249c1  fluffy/ROM/crom-2

1a9f96c495bd6aabef351513def57fdc  bin/ROM/prom-2
1a9f96c495bd6aabef351513def57fdc  fluffy/ROM/prom-2

4a6757f932118fe5cf7dd19f8801e088  bin/ROM/crom-3
4a6757f932118fe5cf7dd19f8801e088  fluffy/ROM/crom-3

7309c62f6b39ea30d396e26f59cba750  bin/ROM/prom-3
7309c62f6b39ea30d396e26f59cba750  fluffy/ROM/prom-3

8039d3e30a446dd71ee7a0bbae62d09b  bin/ROM/srom
8039d3e30a446dd71ee7a0bbae62d09b  fluffy/ROM/srom

c17f1d8b8fe629c1183704c7d04e562a  bin/ROM/mrom
c17f1d8b8fe629c1183704c7d04e562a  fluffy/ROM/mrom

e586f433d2250d9305eb9482aa51d746  bin/ROM/crom-1
e586f433d2250d9305eb9482aa51d746  fluffy/ROM/crom-1

ecf9a06d7a77d8bdcf04ebeaa7498ff4  bin/ROM/prom-1
ecf9a06d7a77d8bdcf04ebeaa7498ff4  fluffy/ROM/prom-1

Is it possible you have bad ram? I had something like that happen to me a few months ago. One specific address location was bad, but it was way up in the memory range. It wasn't I was using enough ram that a program got put into that range that I noticed the issue (program crashed).
 
ack said:
Is it possible you have bad ram? I had something like that happen to me a few months ago. One specific address location was bad, but it was way up in the memory range. It wasn't I was using enough ram that a program got put into that range that I noticed the issue (program crashed).
Click to expand...
Thanks, the memory test reports problems and to contact the manufacturer... sigh, can't they just tell me what is wrong? (The same with Microsoft accounts, where it asks me to contact my IT department - yes, I AM the IT department...)
I would've spent ages looking for some code issue...

Anyway, I hope the script helps people on non-Windows platforms, or on older Windows platforms. I still have an old XP machine because one Flash programmer doesn't work on anything newer.
 
rewrite said:
The seller. Chargebacks with a quality credit card work just fine whether it's Mouser or suppliers from the far east. Heck, we even got to see Amex take care of an entire container of cabinets Yaton couldn't be bothered to ship!
Click to expand...
Just make sure you CYA, and it sounds like you will. Bootleg chips are a huge problem right now and have been for a while. It's especially bad with vintage ICs being literally scrubbed and relabelled in China, and if you're lucky it might be a newer backwards compatible chip at best.
 
I'm making a new menu which is not far from being ready.
It has some slight improvements compared to the chinese menu.
I'll drop it via GitHub soon.
  • A more recognizable look&feel while using the same fonts and elements as the service menu does
  • Auto-repeat directional inputs while holding the controller in any direction
  • Multi-Region: shows US or JAP title-names depending on your hardware or unibios-setting
  • Does not interfer with vortex' work. Gamelists will be added by a batch-file, so everything remains customizable
  • Features game-infos (pub-year, publisher and meg-count, as soon as I completed my list for that task.
I may have plans to improve this further by adding a C and V rom in the future after it was successfully tested.
 

Attachments

  • 2023-10-17 22_58_21-_controller_inputs.asm - NG_MENU - Visual Studio Code.png
    2023-10-17 22_58_21-_controller_inputs.asm - NG_MENU - Visual Studio Code.png
    347.1 KB · Views: 149
Alex said:
I'm making a new menu which is not far from being ready.
It has some slight improvements compared to the chinese menu.
I'll drop it via GitHub soon.
  • A more recognizable look&feel while using the same fonts and elements as the service menu does
  • Auto-repeat directional inputs while holding the controller in any direction
  • Multi-Region: shows US or JAP title-names depending on your hardware or unibios-setting
  • Does not interfer with vortex' work. Gamelists will be added by a batch-file, so everything remains customizable
  • Features game-infos (pub-year, publisher and meg-count, as soon as I completed my list for that task.
I may have plans to improve this further by adding a C and V rom in the future after it was successfully tested.
Click to expand...
Can I suggest to use smaller font, so more game fit on screen, en paginate by alphabetical order ?
 
That is not a bad idea. Maybe we can have tabs at the top, representing the letter, which can be selected from left to right.

And have up and down navigate through the ROMs of that desired letter.

Just an idea. I have no clue how hard it is to make a UI on this system.
 
First step will be finishing the menu with a plain, logical order and high usability.
Then I'll release it on my GitHub, so you can easily test it in MAME before you flash it.
We can talk about improvements when this milestone is reached.
Believe me, you won't need any alphabetically stuff once you navigate through it with all 4 directions.
I admit I don't like alphabetically grouping in lists where not every letter is used,
but chances are good that I'll include a simple favorite system that lets you save favs to your memory card and switch the menu to show them.
 
Rom1 said:
Can I suggest to use smaller font, so more game fit on screen
Click to expand...
You can, but I can't guarantee that I change it.
As a future feature I could think of a way to make fonts customizable, but that's not a task right now.
 
Yea man, just get your idea out there and we can figure it out later. What you are doing is great.
 
I know there was some discussion about ability to do the work earlier in this thread. It's one thing to say you need a 0.6mm nozzle, and it's another to have a visual representation of what that means. So here's what the daughterboard pins looks like scale-wise.

Waiting to borrow @theoddtech 's assistance and FR-301 and nozzle because they don't make them smaller than 0.8mm for my FR-300 unfortunately.

AES cart should arrive this week for measurements @ack , but my MVS ones showed up yesterday.

Screenshot 2023-10-19 114125(1).png
 
Last edited:
I definitely underestimated just how small the pins were

Also - not sure if you will have the same issue, but I found because the nozzle was so small it blocked up a LOT

You won’t be able to use the normal spike because it’s 1mm - I had to use a micro hand drill, it started to block up every 10-12 pins and would typically be completely blocked after a row of 24
 
xodaraP said:
Also - not sure if you will have the same issue, but I found because the nozzle was so small it blocked up a LOT

You won’t be able to use the normal spike because it’s 1mm - I had to use a micro hand drill, it started to block up every 10-12 pins and would typically be completely blocked after a row of 24
Click to expand...
I had the same issue too and also used a micro drill bit like you did to clear it.
 
Last edited:
Not surprising given how small the nozzle is, but good to know it wasn’t just me doing something wrong
 
What's the part number for the hakko 0.6mm nozzle? I have a 0.8mm one (part number N61-02)
 
You must log in or register to reply here.
Back
Top