> If you can't succeed doing it for 10-20 hours a week as a hobby, it's doubtful working even 80 hours will work.

Not necessarily. Speaking for myself, sometimes it's all or nothing. I can't imagine having the same enthusiasm and persistence for a hobby than for a full-time effort, particularly if the latter involved burning a few bridges.

I'm not arguing it's rational, or that one should blow one's savings on it, but sometimes full-time is your only choice.

I can only think of 1 successful game that was built as a part time hobby (Cave Story).

Development of even a simple game takes so much time I don't know how practical it is on 10 hours a week. Game studios have armies of employed people on 60+ hours a week each.

Dwarffortress - started out as an ASCII game and end as an ASCII game. Actually, it's two games in one, and also insanely complex.

Game studios that employ armies of developers usually have tons of artists and spend ridiculous amount of time on graphics and assets. That's why a massively complex game like dwarffortress are never created. So there's tons of niches for complex games with ASCII graphics. (Sorry, pixel arts also takes ton of time)

As everyone else already said, control your scope or otherwise you never finish the game. Dwarffortress got where it is today by simply not aiming for great 3D graphics.

There's plenty of permissively-licensed tilesets out there. Dwarf Fortress really had some trouble because it was built on ASCII until a fan was kind enough to write a graphics library : https://github.com/Baughn/Dwarf-Fortress--libgraphics- . Writing an ASCII rendering engine was probably more work for a lesser product than going with a library to begin with would have been.

Also, Dwarf Fortress isn't done yet. It quite possibly ends with beautiful procedurally generated graphics (though I can't imagine that would happen with Tarn keeping it a solo operation.)

Not quite: Dwarf Fortress has always used OpenGL (http://www.bay12forums.com/smf/index.php?topic=28841.0) to render tiles that just happen to be ASCII characters onto the screen. The fan effort was to port it to SDL, making it much more efficient.

Not many household name games are built as a part time hobby. I'm not saying wait until you're a multi millionaire :)

You can easily make a decent Flash game or experimental iPhone game part time. My first Flash game, with over 200 million plays, was made in two weeks including the time it took to learn Flash.

The additional hours will get eaten up by business stuff and maintenance once you're full time anyway.

I dogfooded my own advice, for what it's worth.

Would you mind linking to your first flash game, for the curious?

It looks like this is the game, made in 14 days, by a team of 6: http://www.bigblockgames.com/games/coffeehero/

This is the one, at my first company:

http://www.miniclip.com/games/stunt-pilot/en/

Two people, two weeks, normal working hours (Coffee Hero was an absurd crunch).

Very much depends what sort of game you're considering. Anything approaching the AAA space, agreed. But mobile games / Web games / anything with carefully controlled scope, not so much.

You may or may not have heard of the 48-hour game development event GameJam? http://globalgamejam.org It's possible to create something playable even in that time.

Also, Minecraft was originally developed part-time, according to Wikipedia.

Working on something serious part-time is hard, and I'm impressed by people who manage it. But it's doable.

I think Minecraft was developed this way too.

"Game development is sort of like being a musician. If nobody likes your songs, you can't make up for it in volume."

Love it.