Re: [stella] FPGA2600 progress

Subject: Re: [stella] FPGA2600 progress
From: Glenn Saunders <mos6507@xxxxxxxxxxx>
Date: Sun, 26 Sep 2004 12:19:50 -0700
At 09:12 AM 9/26/2004, you wrote:
I don't think the author of 2600-on-a-chip did anything wrong here,
because I don't recall him ever asking others to help under false
pretenses that the end result would be open source or anything

Remember that without economy of scale I'm not sure an FPGA 2600 can facilitate, let's say, a portable 2600. If the FPGA core necessary to make this happen is $100 then you are really not going to be able to make these things affordable. What's the pricepoint we're talking about?

To many of us a portable 2600 with LCD screen is what we're really after, right? And while old 2600's are easy to obtain still, they suck power too quickly vs. modern chips.

You need big bucks to do an actual chip mask off of the VHDL and make a final _cheap_ core chip that can crank hundreds of thousands off an assembly line.

I'm still crossing my fingers that the joystick thing coming out is going to have a chip inside that is hacker-friendly from a hardware perspective to cannibalize in a lot of different ways. But at the very least it should be possible to case-mod it and add an LCD to it. It's just that you might be limited to joystick games only.

similar. But it sure would be nice to have a 2600-on-a-chip project
that *was* open under a license that guaranteed it would stay that way.

There are also legal considerations to think about related to patents/copyright.

Other than that he must have used the schematics that Curt gave out, which is proprietary Atari information. Without them I don't think a truly 100% accurate 2600 clone could be created.

(I heard rumors that the TV Boy people were somehow related to the group that did the 1-chip 2600 design for the Jr. and therefore had priveleged tech info from Atari Corp. Unfortunately they fell off the face of the earth otherwise you would have expected them to make a new system off that design by now with the Jakks craze.)

I still have no idea how Coleco or Mattel did their 2600 clones if all they could do was reverse-engineer using 20-year-old tools.

The 2600 system patents, while still on file, must have expired by now, so I think the schematics are in the free and clear. Here is something I found online regarding patent expiration:

Because of recent changes in the law, an existing patent can fall in one of
three categories in terms of its expiration date. If the patent application
was filed after June 7, 1995, the expiration date is 20 years from the date it
was filed. If the application was filed by June 7, 1995 and issued before June
8, 1978, the expiration date is 17 years from issuance. The "best-of-both-
worlds clause" comes into effect if the application was filed by June 7, 1995
and issued after June 7, 1978, in which case the term is the later of 17 years
from issuance or 20 years from filing. These terms are accurate for both
Utility and Plant Patents, Design Patents expire 14 years from the date of

Regardless of these changes, the TIA patent must have expired. I know the TIA patent was filed in the 70s but I think it was finalized in the early 80s (I can't find the link offhand, it's easy to confuse it with the one for the 400/800). But there is no way there is a date on it any more recent than maybe '83.

Current Thread