nesdev.org

I have been trying to build a snes cartridge, I using a am29f040b flash, I connected all the pins where they should be, but I can't get this thing working, I tried programming the full rom, then I tried programming the rom cutting the 512 bytes header and then I use snestool to cut the header and it cut the first 200 bytes header, but it never work.
I have the fifa 97 mask rom with 64 kb sram, I trying to build a FFMQ.
someone please tell me what am I doing wrong, what do I have to do to the rom before programme the chip.
thanks

What are you using for a pinout?

Are you sure FIFA isn't HiROM where FFMQ is LoROM? If that is the case you'll need to shift ROM A(15-18) to SNES A(16-19) and leave SNES A15 disconnected.

64kilobits or kilobytes ? Because FFMQ uses only a normal 8 kb SRAM (but that's 64kilobits so I don't know if you're correct or not).

FFMQ uses a SHVC-1A3B board like most early games with battery backup. It would be better to find a cart with a similar board for the modification.

fifa 97 is a lorom with 64 k (8KB) sram, EA-1A3M-30 board, about the pinouts I did connect all the pins that are the same and connected we# from the flash memory together with oe#. I left snes A19 disconnected.
it shouldn't work with ea-1a3m board?
about the rom, I should cut the first 512 bytes header, right?

Well, 1A3M is pretty much the same as 1A3B exept that a MAD-1 is onboard allowing for more ROM than without it where discrete 74xx logic chips have to do the decoding.
I only know about SHVC-xxxx boards, I've never seen a EA-xxx board. Usually on the Nintendo code format, the first code indicate the console, so a EA-xxxx wouldn't be SNES, but I guess that's not it.

I seem to remember reading that after the Sega Genesis came out, Nintendo of America rewarded publishers that stuck with the NES by letting them manufacture more of their own cart components. For instance, Acclaim/LJN made the boards listed as 51555 (TLROM clone), 53361, 54425, and 55741 in boardnames and boardtable.txt, and Konami/UltraGames made the 351258 (U*ROM), 351298 (U*ROM), 351908, and 352026 (TLROM clone).

Perhaps EA-* boards are like the corresponding SHVC-* except manufactured by EA. It's likely that EA got a similar deal because it was already making its own Genesis carts with a distinctive longer case.

I have NBA 97 mask rom and it is EA-1A3M-30 board too...
I thought that the pins from the mask rom are standard for hirom and lorom carts. isn't taht right?

You need to connect #WE to VCC, not to #OE. Otherwise the flash chip will be in write mode every time a read is attempted. Make sure you re-flash the chip before trying again, you may have corrupted the data stored on it.

Kind Regards, Andy.
http://vectrex.playntradeonline.com

I am still trying to build a FFMQ cartridge, now I am using a shvc-1a0n board, without a sram and mad 1.
before trying to put the flash chip on this board, I did put the rom chip from a NBA 97 cartridge, it work just fine, I did try with a FIFA 97 rom chip and it works too (both of this game ''lorom'' use 64k sram and mad 1), then I think that I don't need sram or mad 1 if I just want to play a game, without saving.
is that correct?
I am using that board because I haven't find SHVC-1A3B board yet.

have anyone ever used an am29f040b flash ''plcc 32'' (the little square one) to build a snes cartrdge?

the pinouts that doesn't match from the am29f040b flash and snes mask rom are:
pin snes mask rom flash chip
1 A17 A18
2 A18 A16
24 A16 OE#
30 A19 A17
31 /OE WE#
I connected all the same pins together and left A19 from mask rom disconnected, I did try connecting WE# to VCC or /OE.

some people told me to connect WE# to VCC and others to /OE, but where should really go WE#?

I found a webside where they use the same chip (am29f040b plcc32) to build a gameboy cartridge http://www.reinerziegler.de/readplus.htm they connected WE# to audio in and a pullup resistor of 47kohm between WE# and VCC.
should I try to connect a pullup resistor between WE# and VCC?

I used HEX EDITOR NEO to cut the first 512 bytes header (offser 0 to 1f0) then the first line looked like this:
4ad1 13ab d55a d5c2 6773 81cf 964b abec
is that ok?

I found that if I am using a board like SHVC-1A3B I should cut the track between A21 from snes mask rom and pin 13 of mad 1 and connect A19 from snes mask rom to pin 13 of mad 1.
is that corect?

I really want to get this thing working, I have been trying like a month.

thank for any help

Hi,

I've built several repro carts for the SNES with an AM29f040b, thet definately work:) There's a picture of a 'Little Magic' cart i made at;

http://www.digitpress.com/forum/showthread.php?t=120604

/WE definately needs to be tied to VCC. Make sure the ROM image you are trying to use is exactly 512k in size.

Double check all of your wiring, it may also be worth checking that the mask rom pins match up to the corresponding lines on the edge connector of the cart PCB.

Kind Regards, Andy.
http://vectrex.playntradeonline.com

I finaly get this thing working
thanks for all your help
regars

Is there a reason roms bigger than 512k do not work with this chip? is it the missing a19 pin?

I am using a svhc-1a3b test board and can get f-zero and smw working but not roms bigger than 512kb like zelda or RTDL etc , rtdl boots up but no graphics after a garbled splash screen, just clear sound. I want to convert my japanese version of smw to rtdl. I have the 29f040 on hand but I do also have 27c801's on order, just would like to figure out why the 29f040 doesn't work.

Thanks,

Yes, it is the missing A19 pin. The 4 in 29F040 stands for 4 Mbit, or 512Kx8-bit. It's possible to put a game on two 29F040 memories by feeding A19 through a decoder circuit to generate separate enable signals.

Ok, I thought that would be the issue. Sounds like I'm better off to wait for the 27c801's to arrive.

Thanks