Misc Project Did the DSi Capture Card End with Katsukity?

  • Thread starter Thread starter KIlly728
  • Start date Start date
  • Views Views 14,178
  • Replies Replies 58
  • Likes Likes 11
FWIW to leave an update about my software work I'm going to put it on ice for now, as cc3dfs on github just commited today, some basic support for CypressUSB Nisetro devices (which is what these DSi's use). Which is very similar to what I had locally in terms of what it's doing over USB. Seems like we won't need to build our own thing. I was able to do some basic testing, and get it to work, so I think i'm going to hold off on just building my own version of something that already seems to work fine. It's behind a flag right now, so a user would still have to compile with extra flags, but seems like they're working on bringing it forward.
 
eBay would scam OP in a second there is no seller protection only buyer protection id use another service

Years ago I was forced to refund and pay return shipping on an upgraded ps4 I sold and was mailed a different system back. eBay didn’t care so in the end I wasted too weeks and $50 to get scammed
There is seller protection, but you need to make sure to have images and video proof to show in case a buyer does try to scam you. Otherwise if you don't have sufficient evidence they automatically side with the buyer.
 
FWIW to leave an update about my software work I'm going to put it on ice for now, as cc3dfs on github just commited today, some basic support for CypressUSB Nisetro devices (which is what these DSi's use). Which is very similar to what I had locally in terms of what it's doing over USB. Seems like we won't need to build our own thing. I was able to do some basic testing, and get it to work, so I think i'm going to hold off on just building my own version of something that already seems to work fine. It's behind a flag right now, so a user would still have to compile with extra flags, but seems like they're working on bringing it forward.

That's fantastic to hear. I didn't know this project was happening and am pleased to see it open source. I'll add it to the greater document as a potential software replacement.
 
That's fantastic to hear. I didn't know this project was happening and am pleased to see it open source. I'll add it to the greater document as a potential software replacement.
I finally got my one working after giving up on it then trying again a few weeks later. The software is quite buggy though and crashes very easily.
 
  • Like
Reactions: KIlly728
I finally got my one working after giving up on it then trying again a few weeks later. The software is quite buggy though and crashes very easily.

Crashing how? I know all the solder joints were very degraded on mine, so it would crash if I bumped the DSi in a specific way.
 
  • Like
Reactions: KIlly728
20250613_062009.jpg


I managed to pick up a capture system. The system is in good shape. The wiring for the capture board is very stiff, with numerous solder joints having a light amount of solder (I'm no expert, but I would have put more solder there if I were installing these). The Chameleon board is "held" in place, loosely, by screws (threading into nothing) acting as posts.. It is somewhat of a miracle a screw didn't come loose and short something. I neat feature I didn't expect is that the mini B can charge the battery (or at least appear to be, given the indicator light).

A bit of a janky setup, but it works... well.... sort of.

20250612_221824.jpg


When to comes to the actual capture of the console, that has been quite challenging. Windows (Win 11) labeled the device as Unknown. So, praying that it wasn't a dead component, I went looking for drivers. Thankfully, Infineon still provides software support for the FX2.

Going through the three capture software programs that I am aware of: NisetroInside's Archived Software, Loopy's DS Capture Software (which I don't expect to work), and Lorenzooone's cc3dsfs.

I am able to get an output from NisetroInside's software, but the current output is a garbled mess. I tried the 3 Mini B cables that I have, and the output is the same. Same result when trying all available versions of the software.

1749838824603.gif


Loopy's Capture Software didn't detect the device. Pretty sure Loopy uses a different capture board, but it was worth a shot.

cc3dsfs should work based on the GitHub page supporting Nisetro capture boards, but, being rather incompetent when it comes to the command line, it is taking time to understand how to get all of the dependencies and other things.

I'll try and get cc3dsfs working before I try anything drastic, but I'm fully ready to reflow all the solder joints. If I feel bold. I may replace the aged wires.
 
  • Wow
Reactions: k66
The wiring for the capture board is very stiff, with numerous solder joints having a light amount of solder (I'm no expert, but I would have put more solder there if I were installing these).
Yeaaaah they're bad. Mine had multiple wires fall off in transit, and even more come loose while repairing the card. I recommend redoing the entire thing if you're able.

My wires were also very bad and fray quickly, so I should find a more reliable replacement.
 
I couldn't exactly pinpoint the problem by the end of the night, but it was clearly not hardware-related.

Installed the drivers and ran the NistroInside program on my desktop, and it ran perfectly.

The "garbled mess" was from my laptop, and after hours of troubleshooting, I couldn't quite figure out what the problem was. I even went as far as copying over the two driver files from my desktop to no avail.

Update Next Day:
It appears to be a Windows issue, or something that I've downloaded is creating the issue. Slapping a new SSD into my laptop with a fairly fresh Windows install, and the capture is a lot cleaner. Not perfect, unlike my Desktop, but far fewer error frames. I hope to pivot away from the NiseTroInside software.

Got Lorenzooone's cc3dsfs program working after spending hours figuring out how to compile the .exe (Bing AI was about as helpful as it was unhelpful), as the release is old, and it works well on my desktop. I still have issues with my laptop, but that's less important.

If you don't want to compile, automatic nightly builds are available.

I'll have the compiling instructions below for posterity.

Update Many Days Later:
After combing through every driver and process on my Laptop, the problem turned out to be in my Laptop's BIOS, where C-states were enabled. Disabling that fixed my issue.

There was also a similar issue on both my desktop and my laptop when using a webcam on the same USB hub. For my laptop, it was due to both devices being on an external unpowered USB hub. On my Desktop, it was due to having a device plugged into a certain USB port on my case, and the other one plugged into a certain USB port on my motherboard. It is assumed they share an internal hub or bandwidth.

========================================================

Compiling the cc3dsfs program.​

For the sake of posterity, for the poor soul who wants to compile the .exe but sees the terminal as an empty magician's hat, I'll post exactly what I did. This was done on a Windows 11 PC.

Install the 3 dependencies: CMake, Git, and G++.

CMake and Git are easy enough to get. Google/Bing, and they are the first result. For CMake, you want the Binary distributions for Windows. For Git, just the standalone installer. Install both.​

G++ is much less intuitive since it is a part of Visual Studio 2022. What you want is the Community version, which is a free download. When you install and open Visual Studio Community 2022, in the Workloads tab, under Desktop & Mobile, select and install "Desktop Development with C++".​
Create a folder, or choose a spot on your desktop to work from.

I'll be making a folder on my desktop just to work from. Going to name it "cc3dsfs".​
On the Window's search bar, search and open "Git CMD"​
Set the root of where you are working from using the "cd" prefix. before the filepath. So, in example, since I made a file on my desktop, my filepath is "C:\Users\MYNAME\Desktop\cc3dsfs". Putting cd before that will look like:​
cd C:\Users\MYNAME\Desktop\cc3dsfs
The command line should now have that file path. If you are unsure of the filepath, you can right-click the file path at the top of the file explorer and copy the filepath as text.​
1749964025667.png
Clone the Github repository.

git clone https://github.com/Lorenzooone/cc3dsfs.git

Cloning the github repository puts everything in another cc3dsfs file, so you can either move the contents out of it or add another "cc3dsfs" to the file path.​
cd C:\Users\MYNAME\Desktop\cc3dsfs\cc3dsfs

Build and compile.

Now the magic happens. The command string on the github page is two commands separated by a semicolon. I wasn't able to run them both at once, so I have to run them one at a time. Note that this can take quite a bit of time.​
Start with the build:​

cmake -B build -DCMAKE_BUILD_TYPE=Release

Once that is done, run the other half to compile:​

cmake --build build --config Release

Profit

The .exe. will poop out in the cc3sdfs file (the second one). Enjoy.​
 
Last edited by KIlly728,
Rather big news happened on the 14th, Stefan from Merki.Net (3DS and OG DS capture card devs and installers) announced on Discord that prototypes of DSI and DSi XL capture cards are being developed and installed.
1750091067605.png


Stefan showcased the boards, what the flex cable looks like, and an output.

Prototype
1750211616716.png

Final revision
1750524176829.png


1750211715489.png

1750211645713.png


The same board will be used for both the DSi and DSi XL, but obviously, the flex cable will be different. The intention is to have the USB-C port replace the charging port.

1750524141198.png


1750211673646.png


1750478570861.png
 
Last edited by KIlly728,
  • Like
  • Wow
Reactions: k66 and rvtr
I will be investigating the feasibility of an HDMI modification on DSi soon.
Currently I attempted DS HDMI on an FPGA with SIP SDRAM but clocking the embedded SDRAM above 80mhz introduces instability. So, external SDRAM and fpga needed.

If I make any progress Ill post it here.
 
I will be investigating the feasibility of an HDMI modification on DSi soon.
Currently I attempted DS HDMI on an FPGA with SIP SDRAM but clocking the embedded SDRAM above 80mhz introduces instability. So, external SDRAM and fpga needed.

If I make any progress Ill post it here.
Is there anywhere you posted about the DS HDMI?
 
Rather big news happened on the 14th, Stefan from Merki.Net (3DS and OG DS capture card devs and installers) announced on Discord that prototypes of DSI and DSi XL capture cards are being developed and installed.
View attachment 511511

Stefan showcased the boards, what the flex cable looks like, and an output.

Prototype
View attachment 511716
Final revision
View attachment 512324

View attachment 511719
View attachment 511717

The same board will be used for both the DSi and DSi XL, but obviously, the flex cable will be different. The intention is to have the USB-C port replace the charging port.

View attachment 512323

View attachment 511718

View attachment 512223
has there been any updates RE this
 
has there been any updates RE this

I have been sitting in Merki's Discord, and the question of DSi capture cards comes up every once in a while from different users. Based on Merki's responses, they are waiting for Loopy to release them before they can install them.

1762716032869.png


Looking through Loopy's Forum, it doesn't look like they've responded or released any information about the DSi/DSi XL Capture Cards.

So, we are all waiting.

Apparently, Loopy has been working on a Capture Board since July of 2024, so that's a piece of information I didn't catch.

1762715800796.png
 
  • Like
Reactions: Dionicio3
The one time I lapse on watching Merki's Discord, the announcement came through that DSi/DSi XL Capture Cards are in production.

1775520653133.png


I'll update when more information comes out.
 
  • Love
Reactions: k66
Cool, looks like I won't have to scrap this project from my TODO list then.

The cited reason appears to be more end-user troubleshooting, and the difficulty of cutting the hole in the outer housing.

I don't exactly like pasting screenshots from Discord all the time, but the platform's walled-garden nature leaves me little choice when citing my sources.

1775523471194.png
 
  • Like
Reactions: k66

Site & Scene News

Popular threads in this forum