Making a console

It takes a combination of many professions to create a console.
However, creating one has become somewhat easier the last 10 years or so.

Essentially, these days consoles are just computers.
Not entirely consumer class hardware, but similar enough to be called a computer.
They may use a custom made CPU, but still the design will be derived from an existing CPU.
This goes for most components, they're similar to consumer PC hardware, but often tweaked and modified a bit to suit the purpose of the console.
Some consoles like the playstation 3 include very custom hardware like the cell processors. Overall this only has been a moderate success, considering xbox 360 keeps up without it.

But this is not what a console is about. That's just a box of hardware similar to a computer, except every consumer gets identical hardware.
Up next comes the hard part. You can install and run software on your computer because it has an OS.
Consoles don't just install Windows, they create their own OS (often derived from Linux or some Unix flavor though).
Usually a very bare bones OS that only offers exactly what the console needs.
The biggest difference between a console OS and PC OS can be found in the amount of layers (like drivers) between the application (games) and the hardware.
Applications on consoles are much closer to the hardware and often benefit from performance boosts because of this.

Now you have a box of hardware, a special OS, but you still need games.
This is where you create a suite of tools to create and debug games targeting your specific hardware and custom OS.
This will be called the SDK for your console.

And that... unfortunately was still the easy part.
Now you need to convince developers it's worth making games for your console.
This is where you'll truly be in over your head.
It takes a large credible company to strike licensing deals and ensure developers of a sufficiently large audience to make creating a game profitable.

After this there's a large bunch of business stuff to take care of, marketing and more.
Making deals to produce your hardware and a reasonable cost, which usually involves mass producing it.
And getting the hardware to the consumers at a reasonable price.



Overall, to truly make a console you need to be an expert in a dozen of different fields.
Hardware engineers and programmers can take on most of the technical problems.
It's absolutely insanity to take on such project. The sheer amount of work to be done is overwhelming and can not be done alone.

Having said that, it's not entirely impossible if you somewhat redefine the definition of console.
OpenPandora did this. This team went through the entire process of designing and creating a handheld multi-purpose gaming device.
It's somewhat different than what Sony or Nintendo would put out, but it's about as close as you can get without being a multi-billion company.

Check these people out, they did many write-ups regarding their progress and the stages they went through when developing OpenPandora.
They've made almost all the details large companies would excessively guard publicly available.
They also talk about how they outsourced the production of many components, the troubles they go through with getting these into the country and more.

A console theoreticlly shouldn't be that hard to make if you have the write knowledge. It is basiclly an extremely specialized computer sold to the market with identical components. If you have ever taken apart an XBox, then you would see that it has the same basic parts as a regular computer such as a GPU, CPU, Heatsink, CD Drive, ect. Most Consoles today have special cases and components that make it very hard to take apart and modify. I've made several computers in my lifetime and it's not that hard if you have an example to look at.

The harder part is the software required for a consoles. Devices such as Mp3 players, cell phones, and consoles use a specific type of Operating System called firmware. Firmware is a specialized type of OS that limits a device to be used for its intended purpose. Mp3s to play music, cell phones to call people, consoles to play dumbed down games. As the previous poster said, you could simply grap the source code for a linux and attempt to modify it to get your firmware. I wouldn't be suprised if you could get some programmer out on the Internet to go do it for you if you paid them well enough.

If you did get a firmware version of linux working, your next problem would be getting programs to run on your kernel and being able to get controllers and drivers for them to run along with writing games in an API that supports the controller.

So yeah, I think it would actually be pretty hard on the Softaware side of things. If you ever get around to doing it, Good Luck

Have you seen the Uzebox (a simple, 8bit console) or RaspberryPi (a $25 bare-board, linux-based computer with 3D acceleration) ?

Both are very low-cost, simple computers capable of running games software. A retail-friendly version of either could probably be brought to market at a cost of under 50 USD, perhaps even half that -- maybe that's still too high a price, but its a fraction of the cost of a nearly two-generation's old PS2 or any of the current generation of portables.

If that's still too high, you might concentrate on providing a software platform for readily-available, mass-market hardware, like PlayPower (or maybe even better, join in and help them).

I've no idea what Kenya's culture is like, but it bears consideration in whatever plans you develop:
Do people spend much time at home? Do they have a television? If so, is it modern (HD, Widescreen, etc) or old-school (SD, 4:3, SCART/Composite/RF)?
Do people spend much time in transport on buses or trains? If so, are they seated or is it commonly standing-room only?
What's a typical income for a Kenyan family? How much disposable income do they have? What do you expect them to be willing to pay for a gaming device?
What's the situation with internet access (either home-bound or cellular/wireless)? Is it widely available, fast, and reliable enough to serve as a delivery platform, or will you have to deal with retail channels? At retail, what margins would retailers expect, and is theft within the supply chain a significant problem?

These types of questions could help you determine what type of hardware is both suitable and affordable, and how software can be distributed.