Thor17
Offline

Hi,
well, I don't like it to make preannouncement when it comes to projects like this...
However, I'm sometimes kind of a perfectionist^^
That's why I want to collect ideas for the µSD cartridge. Things that MUST BE included and things that can be included optional.
My cartridge will be as cheap as possible but with no compromises related to the functions and the durability(my old DMG is even working these days without problems, it would be a shame, if the cartridge would not work after 25 years big_smile)
Basically it's working with a CPLD that emulates a MBC and to allow the Gameboy to communicate with a SD card.
If I'm done with the work on this project, I'll upload everything about it, so everyone can make her/his own cartridge for private use. It will take some time, so don't ask for the time I need to get it done. It's done, when it's done!!! I'm using a stack memory for storing the games. That is needed, because it is not possible, to load one byte from a SD card as fast as the gameboy is sampling it. The stack memory is pretty cheap. It's including 64MBit flash memory and 8MBit SRAM. I'm using this kind of multi package memory, because it saves a lot of space and makes the pcb layout easier.

These are my ideas:

Things that MUST BE included:

-FAT32 compatibility
-games smaller or equal to 4-MBit will be stored in the SRAM to increase the lifetime of the flash (100,000 write cylces ensured by the manufacturer)
-CPLD does EXACTLY acting as the MBC that the stored game has originally built in (MBC1 or MBC2 or MBC3 or MBC5)
-the name (file name) of the game/program you've played/used the last time will be showed at the top of the list of the games stored on the SD card, so you know what's flashed and so you are able to save writing cycles to increase the lifetime of the flash
-detecting corrupted SRAM and warning(if the battery is running out of juice)
-storing the SRAM data to the SD card (not automatically), storing format will be "romname[NR.].sav" (e.g. "romname1.sav...romname5.sav...") for storing more than one savegame for one game
-loading a specific savegame into the SRAM ("romname2.sav")
-virtual RTC for MBC3 games, it can be configured in the startup menu (if enough space available, I'll built in a real RTC)


Optional things:

-other menu languages than english(I would then implent german and if someone wants to translate the menu, it may be possible to implent other languages)


That's everything for the moment. If I forgot something important, than post it in this thread. No private messages, please.

nitro2k01
Offline
Sweeeeeeden

If you are absolutely sure you can do it, go ahead and try. There are already two projects on the way, one by krikzz and another one by someone else, who I'll keep secret for the moment. (Not me.)

I don't mean to discourage you too much, but if you're not sure you can finish it (given your available time and skill) you may just be wasting your time. The problem for you with doing an open source project is that anyone can make copies and sell (well depending on the license, but assuming GPL) without paying you anything. The problem for anyone producing the hardware is the same, the already thin profit margins are spread even thinner if other people start producing the same hardware.

I would argue that open source works well when the development time matches the user base, ie small user base and low developer cost, and big user base and high developer cost (for example the Linux kernel, Firefox, Apache, MySQL or your favorite large OSS project.) Otherwise the risk is that the project becomes stale because of lack of development effort. This is especially pronounced in a hardware project, since hardware has a higher production cost than the electricity it takes to compile a program.

For this project, I suspect the challenge will be the SD card state machine and FAT interpreter. You want this to both work and be fast enough. And you need a bootloader somewhere that cannot be easily erased so you'll effectively brick your cart. Plus, probably, an endless list of things that need to be fixed.

Well well, enough with the discouragement rant.

12ianma
Offline

I would second this, not to be a jerk or rude.

Thor17
Offline

I knew about the project by krikzz.
It's been two or three years ago, when I was thinking about buying a SD cartridge for the Gameboy and didn't find anything.
Back then, I thought the efforts to make something like that, can't be compensated by money. I'm thinking the same today. But, there is one thing, that can compensate the effort in doing such a project, it's satisfying my curiosity big_smile
I ran into the whole Gameboy development stuff. Of course it is an old system. The computing power is a lot lower compared to the most available microcontrollers big_smile
But somehow, it's really interesting to me. That's it. I don't think, I could earn money with that project. I guess krikzz will do a great job. However it's just a thing I want to do. It will be a long-term project. Even if it doesn't work proper at the end, it's not like I'll learned nothing wink

nitro2k01
Offline
Sweeeeeeden

Thor17, if you use IRC, feel free to join #gbdev on EFnet.

kitsch
Offline
matt's mind

also working on this smile 

along with krikzz (but different project altogether)

right on!  good luck...  it can be a headache no doubt

sound like you've got it sorted out though, i also enjoy working on this old stuff, very cool!

ldax
Offline

Me too smile
I've been working on a cart for a while, got some info here.

I say the more the merrier. As you say, if you do it for money it might not be worth it. But if you do it for fun, I say, go ahead.

Your featurelist looks nice, the small roms in RAM feature is interesting.

bbsfoo
Offline
ldax wrote:

Me too smile
I've been working on a cart for a while, got some info here.

I say the more the merrier. As you say, if you do it for money it might not be worth it. But if you do it for fun, I say, go ahead.

Your featurelist looks nice, the small roms in RAM feature is interesting.


create job!!
your project will be open source?

nanode
Offline
(ﾉ◕ヮ◕)ﾉ*:･ﾟ✧ el ass dee j

Holy shit the 2+ year bump lmao.

I'd be super interested in seeing this come to life. Every cart costs so much now a' days.

BennVenn
Offline
Australia

Did anyone ever get this project off the ground?

Thor17 had the right idea with the stack memory though MOQ was in the order of 1,000pcs and required a 4 layer board due to the BGA footprint... Adding support for all those mappers means a rather large CPLD/FPGA is required, increasing the price and power consumption, not to mention there is a very slim chance of the average Joe successfully soldering a pb-free bga package at home.

Then the equipment to initially program the CPLD and Bootloader... All this makes this project very DIY unfriendly.