Intro to Cartridge Translation

Hello everyone,

First and foremost, this is my first post here, what looks to be an ever-strong community of classic gamers, modifiers and developers. It's a pleasure to be here.

Well, on the final day of PAX East, a lot of the vendors were looking to sell their things at fairly low prices. But, when I saw someone selling a mint-looking NES with 6 games, two controllers and a lightgun for $50, I broke and got my wallet out. It's been a wonderful dose of nostalgia, as I look for a couple of the games that I owned in my youth.

As I looked around, I noticed that someone tried to sell an unmarked copy of Final Fantasy 3, which I immediately thought to be fake. But as I looked around, I realized that this DID actually exist, somewhere or other. It sounds like http://www.nesreproductions.com made them at one point, but has since stopped. Someone on another thread here, or perhaps another forum, suggested taking the Famicom cart and "flashing" the English fan-translation onto it, which raised a few questions for me, and I was hoping you could help me out.

My number one question: is there an EPROM in the original Famicom cartridge, and if so, what's on it? Is converting the cart to English as simple as getting a correct flash done? All of the pictures I can see online look to be post-modification versions of the cartridge, which may or may not contain all the original components.

Second question: Can the Famicom cartridge be converted to an English NES cartridge, and am I going to need anything else, like a SMB2 cartridge with all the pins (As the owner of NES Reproductions did)?

Third, but irreverent question: Can I order the shell for a NES cartridge anywhere to place this in, if and when it's finished, and what would your recommendations be on creating a label?

If any of this falls down to technical skills, I have a couple suite-mates that are electrical engineers (I'm in school for comp sci, sadly, which helps very little here), so soldering or the like is within my capabilities.

Looking forward to your replies and to be part of the community.

The vast majority of original Famicom cartridges contain a mask ROM, not an EPROM. You'd need to desolder it, rewire the board to use EPROM pinout instead of Nintendo mask ROM pinout (that's usually about three wires' difference), and solder in an EPROM.

You can play a Famicom cartridge in an NES by plugging its PCB into the NES-JOINT adapter board, which you can find inside "heavy" copies of Stack-Up and Gyromite.

But if you just want to play the games, not collect them on a shelf or sell pirated copies, I'd strongly recommend buying a PowerPak.

Ah, noted. That doesn't sound like too much of an issue, I'd just have to open one up and see how big it is. Does this cover the issue of language translation?

Those don't look too uncommon, I may actually have one back at home. I'll have to check. You'd just pull out the game attached to the converter, and place the Famicom chip in?

I was debating that, but for all that I plan to use it for, the experiment is worth the costs. I don't want to play so badly that I'd bite the costs of a PowerPak, I'm just more interested in attempting to get a cartridge converted and working.

It may not make sense for one game, but once you're done with this game you'll be after another game. Then another. At this point, a PowerPak starts to break even. Do you own (or have access to) an EPROM programmer? That alone is already quite an investment.

If you really decide to go the repro way, there's not much to it: get the most similar donor board you can find (use bootgod's database to look up boards, check board details here), remove the mask ROM(s) (if the game uses CHR-ROM you have to replace 2 chips, otherwise you only have to replace the PRG-ROM) and solder the new chip(s). Anything larger than 64KB will need rewiring, so you have to compare the pinouts of the mask ROMs and your chips. If the donor board isn't 100% compatible, you'll have to rewire and/or replace components.

As for labels, I usually go to a place that laser prints to sticker paper, then I put some spray varnish on it, cut it and round the corners with a corner puncher (or nail clipper). If the printer is good, your tools are sharp and your hands are steady, the result looks great.

This is something I considered, and if I ever want to play a non-standard game after this, or if this gets out of hand, I'll definitely go for a PowerPak. While I will play this, like I said, this is more of an experiment brought about by having a NES once again. It's more about if I can do it, rather than it being monetarily profitable.

I do, actually, have an EPROM programmer at my disposal. I would've thrown in the towel long ago if that wasn't on my list of assets. What would that cost, maybe a half grand? More than a PowerPak, at least.

Well, here we have it: the board analysis. Not sure if the wiring gets more complex as the size goes up or not, I'll have to look at the pinouts when I'm home from work. It looks like it's PRG, not CHR. What's on the game's EPROM (or mask ROM), anyway? Not just for this game, for any game that would have one. Is that what's ripped into the .nes files?

What qualifies as a replacement for the mask ROM, will an EPROM from a similarly-structured board work, or should I simply go ahead and find my own IC for this job?

They're not THAT expensive. A good one should cost more than a PowerPak, but there are cheaper alternatives that are sufficient for hobbyists.


The deal with this board is that no US games use it. The closest one is TKROM, which requires the CHR-ROM chip to be replaced by a CHR-RAM one. The problem is that since CHR-ROM can only be read, not all boards have the pin that indicates when CHR is being written (required for CHR-RAM, since it can be written to as well as read from) present in the cart connector. So you either have to find a board that has the pin or hack one in (some people glue a thin piece of metal to the board).


This is not a matter of BEING PRG or CHR... A NES cart always has both chips, but sometimes the CHR is RAM instead of ROM. The difference is that CHR-ROM chips come programmed with all the graphical tiles used during the game, while CHR-RAM is empty and is filled (and re-filled, as often as necessary) with tiles at runtime by the game program.


.nes files are composed of a header, that provides details about the game (such as the mapper it uses and the sizes of the PRG and CHR chips), followed by the contents of the PRG chip and finally the contents of the CHR chip (only in case it's CHR-ROM). Games that use CHR-RAM have both the program and the graphics inside the PRG-ROM. Graphics can be stored in any format, since the program is responsible for decoding and writing them to CHR-RAM. CHR-ROM graphics are always uncompressed, in the format required by the NES, since they aren't processed at all.


Most 5V 8-bit DIP EPROM, FlashROM or EEPROM should work, I've never heard of one that didn't. One thing to keep in mind is the naming scheme for memory chips: the numbers at the end are Kilobits, not Kilobytes! This means that a 27C256 can hold 256 Kilobits of data, which is only 32 Kilobytes. A memory chip that holds 512 Kilobytes of data will most likely end in 040 (representing 4 Megabits).

Another possibility, which involves a more advanced soldering technique is starting with TGROM and soldering a RAM on the wrong side of the PCB, under PRG-ROM. Or jumpering a SMD RAM on top of PRG-ROM.

Either way I'd pick something easier for a first project, wait for INL's flashcart, or use a PowerPak/Everdrive N8.

My current repro boards support TNROM, you just drop in the EPROM directly. The flash version will support TNROM as well, but you would have to wait a few more months for that.

Those repro boards do look quite tempting, I have to say. That would likely be the way that I would go, for this project, if I could.

Admittedly, as others have said, this isn't really as easy as I had anticipated. I'll probably have to do a little more design work before I get this underway. I could see selecting a donor board, and readying new components being a lengthy process.