Update: Sega GameGear CPLD Cartridge(s)

Just an update for an old post I made.

http://jazz-disassemblies.blogspot.com/2013/09/gamegear-cpld-cartridge.html

Remember the Flash cart that Majesco made with a CPLD as the mapper? Well, I tried again to get a pirated ROM to run in place of the old Caesar's Palace ROM and it refuses to!

Dumping the Binary for Caesar's Palace, it is an exact hex match with the copy circulating on the internet, so I know whatever mapper is contained on the CPLD must at least support ROMs up to 256 KBytes, because Caesar's Palace is that large. I tried the Fan translation of Phantasy Star Adventure which is only 128 KBytes and it will not run. I even duplicated it across the full capacity of the flash ROm just in case one of the higher address bits was tied to something I didn't see, like A18 on a 49F040 being tied to VCC. It still refused to run...

Doing some research on the GG and different BIOS versions, some GameGears check 0x7FF0 for a line of text, which is in fact present at the location it needs to be. I have tried everything and yet it still refuses to boot up. I am rather upset by all of this, which is why I am going to make my own flash cart for the GG! BLARG!

VileTim and others have made CPLD versions of the original 315-5912 Mapper chip included in a few Sega Master System and Sega GameGear carts. Obviously the goal is to design a board that makes it so we can use commercial and modern components rather than salvaging and destroying official devices. The first point I am trying to make is to design a board that supports 512 KByte ROMs, homebrew or otherwise. Then I would like to add SRAM and a battery because I am hopeful for a music tracker that runs on SMS and GG in the future.

LSDj is getting old in my opinion and there are other devices that make unique and varied sounds, such as the 76SN489 inside of the GG and SMS.

Anyhow, I have now completed a version with an official mapper that 'should' support up to 512 Kbytes when using a 49F040 ROM.



I am also working on two simultaneous versions that include SRAM and a battery, but they are going much slower because I am running out of room for traces...






The board containing a 49F040 still holds up to 512KBytes, but the TSOP can only hold up to 128KBytes. This downgrade is because the TSOP is much more available than the 49F040 and much easier to route traces to.

The two latter boards are also a little taller than the other board. The first one I designed was based on an actual official game, garfield or something, which does not fill the entire shell. The other boards I measured just how much more I could fit inside of the shell and still have it close without any cutting, which is important to me.

The second biggest concern of my boards is that I hope to get them to support Sega Master System Homebrew and ROMs. The jumper between cart pins 41 and 42 is an SMS/GG selector. when it is bridged, the GG runs in GG mode, but when it is cut, it runs in SMS mode. I have done no (successful) testing of this just yet, but I will surely post about it once I do.

Stay tuned

GameGear CPLD Cartridge

After researching PLD for my Atari dev cart, I remembered a time when I was trying to figure out a Gamegear flash cartridge. It is simple enough, but I had little luck because I was still very unlearned at the time when it came to Addressing bits.

Anyhow, I came across a Caesar's Palace Cartridge by Majesco which utilizes a CPLD! At the time I figured that I would never see one again, nor would I care at all if I were to. Now I care very much because they are pretty damn incredible devices!

The pictured CPLD is by Xilinx and now that I think about it, I really should locate it once more and dump the program on the CPLD. I dumped the game ROM, but it was exactly bit for bit the same as one I ... erm ... 'found' online.



below is some pictuyres I took when I was trying to make that flash cart. I really wanted to play the Translated Phantasy Star on actual hardware, but I never got to before dropping the project.



My drawing is flawed in that the highest address pins are connected to VCC and I programmed the ROM at the lowest bank which would place it at a point where the highest Address pins would all be low. Haha, silly me. had I simply filled in the extra ROM space with the ROm over and over again, it would have worked. I'll try again in the near future, no worries.

Below are more pictures of the bare board. Good times. lol
See how I ruined a few pins by pulling the pads up? That was a long time ago and I am proud to say that I can do much better.






Even though the game gear is a terrible handheld, battery- and longevity-wise, it uses some great technology. Some things I like the most are the fact that it has all of the fingers on one side of the board, such as the gameboy. The atari the fingers on two sides, which gets annoying when designing a board. You'll also notice that every component is Surface Mounted; caps and Mask ROM included. This makes the size of the complete board much smaller and able to fit inside of the cartridge shell much more comfortably.

I may try to make a flash cartridge in the near future. One such cartridge that will use a CPLD so we can address higher memory addresses and will utilize saving features with a battery just because it is easier. If I stuck with those parameters, it should be able to support more official games as well as any homebrew titles.

I really like looking back on past projects and seeing how I could have done them better with what I know now. I am still rather new to EE, but I have every last bit of motivation and hunger for learning. I hope I can go back to school soon. I would really like to pursue my Master's degree next if not at least another BS degree.

Thanks for reading,
Jazz