Hardware How can you produce GBA Cartridges now'a'days

[trevor403]

In recent years technology has sky rocketed in complexity. I bet if we wanted to we could make a fake GBA cart connecting to an Arduino pushing information off an SD card. But this skyrocket in tech has also given us great access. The most recent open source GBA Carts were made in 2004-2005 and those were very limited to begin with. Now we can get the full sized EEPROM we need for both save and the rom. So is anyone aware of a process for manufacturing new carts? The goal here is to make a cart that can be distributed (maybe not for money legal terms) But for all those people working hard at hacks who want to make their game real. So let's see what we can make in Eagle and what surface mount components would be needed. Thanks!

 

[enarky]

So... why don't you start? GBA hardware is very well documented, what you're asking isn't to hard to do. If you do the grunt work I'm sure you'll find people to join you on improvements.

TBH, your posts sounds a bit like "umh, I have this idea, please do the hard work for me".

Also as long as the EZ-Flash IV is widely available there's little reason in this besides the "because I can" bragging rights.

 

[trevor403]

You misunderstand my tone, anyways i'm a software engineer, this is highly out of my line of expertise but I will look into the way the persistent memory interface works on the GBA cart, lots of other documentation exists for the actual rom storage. Here is an example: http://www.dl9sec.de/gbacart/dld/GBA_Flashcart_(Basic_type)_Board.zip Anyways I am not very familiar with any of this so I can try my best but I doubt I can get anywhere fast. But don't make me out to be what you have said! I am trying my best to research.

 

[jefffisher]

is there something specific you are trying to get onto a card or are you just talking in general.
if it's just for general or one time use or something you want for yourself and not mass produced you can still get a cheap gba flashcard for $10
these http://i43.tinypic.com/9vgvom.jpg
work just like original games as long as there is only one rom on it.
and the $10 price is almost certainly cheaper than you'd spend building something and insanely faster and easier.

 

[The Real Jdbye]

In recent years technology has sky rocketed in complexity. I bet if we wanted to we could make a fake GBA cart connecting to an Arduino pushing information off an SD card. But this skyrocket in tech has also given us great access. The most recent open source GBA Carts were made in 2004-2005 and those were very limited to begin with. Now we can get the full sized EEPROM we need for both save and the rom. So is anyone aware of a process for manufacturing new carts? The goal here is to make a cart that can be distributed (maybe not for money legal terms) But for all those people working hard at hacks who want to make their game real. So let's see what we can make in Eagle and what surface mount components would be needed. Thanks!

You won't be doing that with an Arduino, it's far too slow to transmit data at required speeds
But I don't think it should be that hard to do.

 

[trevor403]

Yeah I just want the authenticity of booting straight to the rom, GBA&NDS cart offers that I know were to buy them. Does that cart boot straight to the rom? Anyways I've been working on a huge Pokemon hack so me and the other devs wanted to make carts for when we finish the final copy. I would love to find something so I don't have to build these babies, thanks jefffisher.

EDIT: Also saving is important, so we kind of need SRAM, but I know some have an EEPROM patch for saving.

 

[FAST6191]

What you are describing are called repros/reproductions and we do not tend to do it around here if you plan to distribute them in any capacity (though as a vanity thing for members of your team is no less unacceptable than anything else we do around here I guess). There are options for those that want it, mainly things like the 3 in 1 expansion pack (you patch the save, write the ROM, possibly disable write pins for 3 in 1 NOR and carry on with life), though the GBA size version is hard to come by and expensive when you do.

Straight up homemade GBA flash carts do exist, and should work for one shot stuff, however they are not terribly well developed either

 

[Darkipod]

This is something I intend to work on once I graduate, originally I was going to do this as my senior design project but that won't be possible as the projects are now assigned instead of thought up.

 

[Helper]

Like Jdbye said, an Arduino won't cut it.

If I wanted to "produce" a GBA "flash cart" (characterized by loading (a) ROM image(s) from a storage device at runtime), I'd probably start with an FPGA (or, perhaps, CPLD) development board, some fast RAM, and a microcontroller hefty enough to read a filesystem on an SD card (or whatever). (And supporting circuitry; voltage level shifters, etc.)

If I wanted to just make a GBA cartridge--truly read-only, one single image burned to the chip before installation and that's it--where's the difficulty in that? Get a PCB, get a compatible chip, burn the ROM, solder it in.

In any case, good luck making anything at a lower cost than commercial offerings. Apart from design effort (nobody's time is worthless), consider the costs of PCB fabrication, components, assembly, soldering (especially if using SMT), and so on; in particular, consider these costs at quantity one pricing.

I'm all for open source designs; but it's still unclear what you're even asking for.

And if you're just asking for somebody who can make some GBA repro carts, I'm sure somebody somewhere offers the service. Hell, even I could probably do it.

 

[Heran Bago]

There is one guy out there who does GBA reproductions, generally English translations, and he's a huge dickhead so I'm not linking or discussing it further.

There are a few more you can find on AliExpress that are much better and cheaper. Mother 3 English etc.

 

[Darkipod]

He wants a new true GBA flash cart. With a fast enough and small enough FPGA (that still had enough gates to do all the logic) you could make one. I was thinking about designing a IC from the ground up in veralog and then sending out the HDL to be fabbed and such. The problem is that this would be very expensive in small quantities, and I dont think I would be able to start a kickstarter. I just dont think I want to start a business with this. I also would probably put a lot of work into it and don't really want to completely open source it to be marketed and sold without me receiving some sort of compensation. Either way, I will design one, and when I do I will be sure to make posts here about how it works, and maybe crowd fund a prototype, which after debugging I would send out to people like FAST and a few other notable forum members to review and to get feedback. From there I'm not sure what I will choose to do.