Just wanted to share this thought with you because it both cost me quite some hours but brought me even more.

Picture yourself as a hobbyist or indie, you have a great idea, sky's the limit.
But there are always moments you hit a big challenge which demotivates:
- modelling or creating a level
- some difficult math or equation
- playing a game is a quick escape, or maybe stop the project because the idea wasn't worth it anyway (or was it?)
- etc. Etc.

I believe it's one of the main reasons lot of us don't actually finish projects.
This in combination with no check lists or high over set goals, marking even small achieved goals can give a fullfilling feeling.

What to do?
- forget the idea, delete the function, delete the mesh and just start over. It gives you new energy, less faults and keeps you motivated. Besides this, the end result surprised me more then often.

No scientific or difficult lesson, but it really can bring a lot.

I've mentioned this before.

You must have (a) a list of action items to avoid wandering off and doing inconsequential tech things, (b) a team size of more than one so you have people that depend on you and motivate you so you're unlikely to slack off or get bored, (c) an actual finished product you're aiming for rather than just an idea to more clearly define your action items and keep your team motivated on a tangible goal, and (d) a realistic target timeline the team is expected to adhere to so as to avoid procrastination.

If you're just working in isolation on a vague idea, you've already failed. A few people pull it off but those people are exceptional, not the norm.

Not really DirectX -- or even development -- specific, so I'm moving this topic over to The Lounge. :)

Thanks, I fully agree.

@SeanMiddleMitch: I think it depends on what you want to achieve/ how big your goal. I'm a passionate hobbyist without a team and apply it with success. For example for a new feature in my engine, a tech demo or a "level"/ scene.