noriaki_kakyouin wrote:
In simple terms? Is there a good reason or is it just that Nintendo ROM programmers did it to make things difficult?
The main reason is that they are mask ROMs, meaning to program them, you change the chip mask rather than programming it through the pins.
There are some very rare 24 pin 8K EPROMs, but there aren't any 28 pin 128K EPROMs that I know of.
Incidentally, the maximum amount of ROM you can fit into 24 pins is 8K bytes, and 128K bytes fits into 28 pins. This leaves you a single control pin is all.
Most EPROMs have at least two control lines (/CE and /OE) since one is used to supply the programming voltage, while one is used in programming.
This necessarally limits you to 4K on 24 pin chips, 128K on 28 pin chips, and 1Mbyte on 32 pin chips.
Those 28 pin 128K ROMs on NES carts are a standard pinout, but as stated only have a single chip enable.
As for the CHR ROMs, I am not quite sure WHAT kind of ROMs those are. Nintendo chose them for some odd reason that I cannot figure out. The only guess I have is that they chose them because they have two chip enables. The CHR stuff needs two for /CE and /RD. I suspect these ROMs actually have THREE chip enables total, but I've been too lazy to check
The pinout isn't too standard compared to an EPROM or mask ROM pinout either.