Re: R6502AP

On Nov 27, 2014, at 6:43 AM, Anders Carlsson <anders.carlsson@sfks.se> wrote:

> Marko Mäkelä wrote:
> 
>> Isn't it ironic how Nintendo is allowed to copy from others but not vice versa? As far as I understand, it is not a copyright violation to copy a physical item if some minor details change. It would be legal, provided that no patents, design patents or trademarks are violated either.
> 
> I am not a lawyer, but which intellectual properties belong(ed) to the MOS 6502? As far as I know, one can't claim copyright on hardware, only patents. Since the 6502 doesn't contain any microcode, but a PLA, I suppose there was no software copyright that could apply as well. To be honest, didn't the 6502 itself emerge out of the 6800 and at least to a beginning was controversial?
...
> Given these assumptions, I suppose a CPU design that indeed contained microcode, to which extent it can be copyrighted just like higher level computer software, would have been harder for Ricoh/Nintendo to copy.

You can copyright mask designs now, but at the time, you could only get patent protection. The US was the first to make masks copyrightable in 1984. Software wasn’t copyrightable here until 1980.

http://en.wikipedia.org/wiki/Integrated_circuit_layout_design_protection
http://en.wikipedia.org/wiki/Software_copyright

Also, Japan back then was like China today. They would happily copy a product, even look-and-feel or software, in order to get into the market. 

The controversy MOS had with Motorola had a number of angles:

1. Pin compatibility — the 6501 was pin-compatible with the 6800, but there is still nothing wrong with that today. It let you reuse a socket but you’d have to rewrite all your software for the 65xx. The 6502 has an integrated clock generator circuit, which made it cheaper and simpler to use by itself, so it would have beat out the 6501 for new designs. Peddle claims socket compatibility was the biggest aspect to their dispute, but I think that’s an engineering-centric view of a legal/business matter.

2. Patents — the 65xx was designed by ex-members of the 6800 team. They reused some general ideas but improved on them. It would be interesting to see a side-by-side comparison of the two architectures. It was and is a common practice to patent as much as possible about chip designs, so Motorola likely had some general patents that covered almost any microprocessor (just like Intel). These weren’t tested in court because they were kept for defensive purposes, but the MOS team were attracting the attention of their former employer by undercutting them in the market.

3. Stolen documents — through the lawsuit, Motorola discovered that one of the MOS engineers had taken documents with him when he left. Peddle claims he was furious about this and had told them not to bring anything with them. This didn’t trigger the suit, but it gave Motorola the edge in settlement negotiations.

4. Pricing — MOS was undercutting the whole CPU market. Within a few years, the 6800 pricing had come down and Motorola was upset. This is probably the biggest issue which led to the legal trouble.

> The various articles about the Famicom keep coming back to the fact that there was no 6502 development software in Japan, so Nintendo had to roll their own. This does seem like a false statement as HAL Labs made several games for the VIC-1001 which clearly predates the Famicom, even on the prototype stage. It is quite obvious that in Japan, they most reluctantly would import and adopt some foreign tools (e.g. cross assemblers etc) from USA or Europe, so disregarding what Commodore themselves and through partners did in Japan, perhaps there really wasn't anyone doing 6502 coding before the Famicom.
> 
> Perhaps though Nintendo rolled their own development software to not raise suspicion from outside, in particular if they were going to use a special CPU with built-in sound that resembled/copied an existing design but without had it licensed. I don't know how tightly various software firms worked with the hardware manufacturers, i.e. if MOS et. al would have gotten a word of notice if some third party software developer in the US sold software to Nintendo in c:a 1982.

Nintendo were (and are) control freaks and the Japanese market was highly isolated. The native 6502 platforms of 1982 were the Atari, Apple II, and VIC-1001. The first two were US-centric and the VIC had only 22 columns and required RAM expansion to do anything useful. At the same time, the Japanese manufacturers were just starting to get traction with their PC-series. It makes sense that they chose a local model for their dev system.

There really isn’t a need for a complex explanation.

-Nate


       Message was sent through the cbm-hackers mailing list