Chip's Challenge in FPGA?

[NiHaoMike]

More for the fun of it than anything else, has anyone ever implemented Chip's Challenge or a similar tile-based game on a FPGA? (For those who don't know, a FPGA is a chip that can be programmed to run your own custom logic. They're actually very commonly used in retro gaming.)

At the least, it would be interesting to see how much faster it would be compared to the usual CPU implementation, although that sort of speed improvement would only be useful for automated solution finding/optimization. There are some who think it's cool to implement video games as state machines with no CPU.

I have a Spartan 6 LX45 on a Digilent Atlys that is my general purpose FPGA experimentation board (and is the one I will use for this if I get around to it) and an Artix 7 A35T board with USB3 for stuff that really needs bandwidth. (The latter is currently cranking out crypto nonces for my best friend Naomi Wu, using the attached PC to feed it data over USB because the algorithm needs a lot more RAM than is on the board itself.)

[The Architect]

It's a cool idea, but I don't know of anyone around here who's actually tried it.

For anyone who may still be confused, this would be roughly analogous to using the logic elements in CC2 to create an entire video game from scratch! (Though it would run a lot faster than in CC2's 60hz engine. )

[NiHaoMike]

Someone did make a "game within a game" in Rocks'n'Diamonds, but there are way more elements (with maybe an order of magnitude or two higher complexity) and a bigger playing field that would make it quite unfriendly to implementation on FPGA.

On the other end of the complexity spectrum, the Game of Life has been implemented on FPGA many times.

[The Architect]

I'm sure it has. I'll bet there's not a logical system, physical or virtual, in existence that hasn't had the Game of Life implemented on it at some point!