Ricoh 2a03 history

[djmips]

In my research, there's no evidence that Ricoh licensed the chip designs from MOS. I've talked directly to Bill Mensch about this and IIRC he indicated that they just copied it. I also have heard Bill Herd weigh in as well that MOS looked at de-capped 2A03 and noted the 6502 core but declined to do anything about that. There was some 'speculation' in CBM hackers list in 2014 that the BCD disable was enough to avoid a patent infringement and perhaps they only hoped to get away with copying the 6502 gate for gate (as shown by Visual 6502) without too much notice. They tend to be very quiet on these things as it probably tarnishes their legacy in a similar way to how they (Nintendo) allegedly stole the Donkey Kong code in order to make Donkey Kong sequels in house. They lost a lawsuit in this regard and the story that they (Nintendo) didn't code Donkey Kong is swept under the rug. In any case, what I am saying is that I don't have any quotable evidence that they copied, only heresay , however unless you have documented proof that they licensed the core from MOS I don't think you should include that 'fact' because there is a strong suspicions that they merely copied.

Some facts:

• the 6502 portion of the 2A03 has an identical layout.
• chip masks were not copy-writable until 1984; 2A03 came out in 1982
• only 5 places in the polysilicon layer were cut to disable decimal mode ( SED CLD etc still work and show up in the flags register)
• "the 6502 <in the 2A03> databus drivers are reduced to remnants which drive an internal bus, eventually driving some logic which in turn drives the real databus drivers."

references:
https://en.wikipedia.org/wiki/Semiconductor_Chip_Protection_Act_of_1984
http://cbm-hackers.2304266.n4.nabble.com/R6502AP-td4659401.html
http://forums.nesdev.com/viewtopic.php?t=9813

Ten part series on the origin of the Famicom (in Japanese) archived
https://web.archive.org/web/20140920211721/http://trendy.nikkeibp.co.jp/article/special/20080922/1018969/

Part seven talks about why the 6502 was chosen (size)
https://web.archive.org/web/20141017171734/http://trendy.nikkeibp.co.jp/article/special/20081002/1019378/

This last part does say that Ricoh licensed the 6502 but from Rockwell (not MOS) which I also find strange since how can a Ricoh license the 6502 design from a 'second source'. I'm only speculating, but maybe Ricoh did have a relationship with Rockwell and acquired the 6502 IP through a back door deal but I don't know if they actually paid anyone at all per unit on the 2A03 in the Famicom/NES. Kind of like how you see a lot of cut/paste code showing up uncredited in other people's software.

[flipacholas]

Thanks for the info, I've amended that section and added a mention to this

[Salduchi]

This last part does say that Ricoh licensed the 6502 but from Rockwell (not MOS) which I also find strange since how can a Ricoh license the 6502 design from a 'second source'. I'm only speculating, but maybe Ricoh did have a relationship with Rockwell and acquired the 6502 IP through a back door deal but I don't know if they actually paid anyone at all per unit on the 2A03 in the Famicom/NES. Kind of like how you see a lot of cut/paste code showing up uncredited in other people's software.

Do you know if Rockwell actually supplied any 2a03 chips for the NES and Famicom? I have an NES that seems to have a Rockwell CPU.