think a $39 "joystick" with some built in games and a SD memory card reader for the option of playing additional games from ADF images would sell.
I've had experiecne in the Toy industry, first off. A mass retailer is going to have a problem with a TV game unit that has an SD card interface. First off, sega, Nintendo MS, Sony will raise a stink becuase the whole purpose of an SD slot would be to play games that folks can obviously download for free. They will strong arm whatever retailer wants to carry this device. (of coruse this is the sort of thing you won't read about on Slashdot or game magazines becuase everyone does this monkey business on the down low)
An SD slot will undermid, XboxLive, Sony playstation store and all the classic game bundles arcade games these companies release.
Also lets face it. The Amiga gaming brand is more popular in Europe than the USA. So right off the bat this will be a hard sell to Walmart and Target. Their response will be. We already have Atari, SEGA, Colecovision, REAL ARCADE TV games. Why should we carry this one?
In the USA the Amiga was associated with failure (for those that even remember the brand name). Commodre also to this day has a bad rep for screwing with retailers, and is known as the company that went out of business because they couldn't sell that Amiga computer...
This is an impulse buy item,
So for a mass market item you don;t just release one TV GAME item. First you create a TV game Bundle that targets acade, another with sprots, and cater to specifice genres. You have to release them a few months apart from each other. Then like Jakks Pacific you use the FPGA platform to develop games based on licences for "evergreen" licenses like disney and the like.
I've done reasearch into the ASIC costs, The only way this would ever be released is as an FPGA version. I got costs of something like 200K upfront money for getting chips started. Then theres the issue of Non recoverable engineering. Then there's the issue of Amiga Inc and dealing with the legal costs of licensing whatever parts of the OS you have to. When you get two licensing legal teams talking to each other it gets expensive.
So you take the Angel investor route... or partner with a toy manufacturer..by the very act of approaching investors you're spliing the beans so to speak, now they can take you core idea and hire a low cost dev team in China who will do the cheapest solution (nowhere near as optimzed as Jens)
My point with this rant is that if it were simple it would have been done. As proven before in the Amiga scene. The folks who get things done these days are the expert hobbyist who release projects as opensource. I think EVENTUALLY someone will be able to reverse engineer the AGA chipset.