The reality of the Ouya console doesn’t match the hype

I'd like to add one more thing as well, but don't want to go back and edit: I find that Ouya, by having a -functional- hardware prototype that serves most of the core features it will promise as a piece of hardware, is already well advanced beyond the state of most kickstarter projects. The most frustrating thing for me about kickstarter is how many projects are -conceptual- at the kickstarter phase.

I refuse to give someone money because they're unwilling to take the risk to build the first version of something with their own sweat, blood, and tears before begging for cash. Why should I pay you just because you have an idea? Show me that you can fully implement it. Better yet, show me that you're done and you just need to raise the cash to make it happen. There seems to be something in the start-up world these days that just having the idea and the merest inkling of how to achieve it should be all that it takes to get people to throw money at you.

The great thing is that pretty much every game on steam supports it and even recognises it as a 360 Pad so game tutorials will display the correct labels and colours for buttons. Works great under Linux too.

What surprises me though is the number of people to whom this ability is surprising "You're playing a PC game with a joypad? What? how?"

It's also surprising that even gaming specialist manufacturers like alienware don't offer joypads as an option when building a system.

Don't forget drama.

How on earth would one play "cut the rope" on a controller?

I agree though that it might well be worth the $99 just for use as a media box and to play old arcade ROMs on but the question is whether that is a viable business model for them?

This is a bunch of commodity hardware thrown into a box surrounded by very sketchy details surrounding development,software and the controller.

The console industry is littered with spectacular failures of this type, and I'm before I believe this one is worth investing actual money in I'm going to need more than Yves Behar, Julie Uhrman who's "been in the game industry for years" (exclusively on the software side), and several mysterious people who refuse to even be named.

Especially in the time frame cited; I've worked in both fields and unless they have experienced full-time engineers and developers working on this right now there's no way they're making the deadline. The "we have a prototype" is at best a vague indicator of progress. Do they have a board designed? Has it been fabricated? Do they even have android running on the hardware they plan on using?

If not, are they using off-the-shelf boards? I can't find any Tegra 2 OTS boards available remotely within the price point they list, much less Tegra 3.

And the "some of them would get fired if we tell you who they are" is more disconcerting than assuring. If they do have people with engineering/manufacturing/software development expertise these people do not believe in this project enough to leave their day job. That should speak for itself.

EDIT: for example, see http://trimslice.com/web/order-direct. Their Tegra 2 board with more modest hardware specifications is only available at more than twice the price point touted for Ouya. Note that this doesn't even include the cost of the custom bundled controller, which is itself a substantial hardware undertaking.

First, I've been unable in my (admittedly, cursory) research to discover any prebuilt systems that even come close to their promised hardware specs. So until someone proves me wrong by posting a link to an OTS solution even close to what they're promising, lets assume that the system needs custom design.

This is a whole massive can of worms. I have some experience with hardware design (though not as much as a full-blown EE), and I'd ballpark hardware design alone as a half-year effort with multiple full-time EEs. And that's a best-case optimistic estimate. People who work in software don't seem to understand how hard this stuff is. It's not like you can simply take another schematic, slap the new parts you want on, delete the old parts, autoroute it and call it a day. With parts of this complexity there's a whole host of other design issues that come out of the woodwork. High-frequency switching means if one of your autorouted traces is longer than another, you're going to have timing issues. Crosstalk between layers, capacitance and inductance of the wires themselves can wreak havoc on your busses. You need to intelligently design the ground planes, separate your signals (analog and digital components), etc etc etc.

These are only the issues I'm aware of. I'm sure people experienced with creating boards this complex can fill me in on more. It's not like a hobby arduino project - hooking up traces to debug problems can increase the capacitance of a line enough to cause a new set of problems.

This isn't really directed just towards you, but it feels like a ton of people here are missing how difficult hardware design with components this complex is.