Should I do what I desire, or do something I know I can?

If this is a hobby thing that you're doing in your own free time, why not do something you actually want to and look forward to? Why spend your time doing something you're not interested in?

You may not be able to make the game right this very second (because of skill level or whatever), but you sure as heck can work your way up to it (and make parts/prototypes of it, as you suggest) and learn a ton in the process until you can ultimately realize your goal (or move on to a new goal).

If this is a hobby thing that you're doing in your own free time, why not do something you actually want to and look forward to? Why spend your time doing something you're not interested in?

You may not be able to make the game right this very second (because of skill level or whatever), but you sure as heck can work your way up to it (and make parts/prototypes of it, as you suggest) and learn a ton in the process until you can ultimately realize your goal (or move on to a new goal).

That's what I thought, but I didn't know if it was practical in application.

I honestly do not think I'll ever finish the game, I know myself, but, I want to TRY. I don't mind doing really rough write-up type deals of it, if that's what needs done, so your suggestion sounds good.

It's a pretty thin line, though. Like Cornstalks said, it's easier to put time into something you enjoy doing. But if your idea is way beyond your means, it might not be good to go all-in on it. You'll have to assess how closely your abilities match what you want to do, and plan out a reasonable way to achieve it.

If you don't really have much game programming knowledge, keep it simple and follow tutorials to get started then add or change features to cement your knowledge of how they work. Pick something relevant to what you want to do: A platforming adventure game could start off with a side scrolling tutorial. A top down shooter has some common elements with pong.

BOTH


Clinton

As Haps already said, it really depends on how big the gap between your skills and the talent needed is.
A good way to figure out whether you'll be able to cross that gap in the process is to try and construct the general structure of the program. Can you imagine the general way the program is going to work? Can you name your shortcomings and know where you'll have to learn more? Do you know where you'll have to work modular, so that you'll be able to insert the more complex code later on, when you're capable of writing it? Or are you just completely lost and hardly know where to start?
A project you care about is a great way to learn new things (also, you're working on sth. you care about), but it won't do you any good if there's no chance of finishing it. Keep in mind that there'll be a lot of hardships you don't even think of yet.
And finally, don't let me scare you. If you think you can do it and really want to: go for it. There'll always be someone willing to help you with problems you encounter as long as you're willing to put in the same effort.

bw,
Tobl

Always two good things to ask yourself: "do I like doing it?" and "am I learning something from this?". (The third one is: "do I get paid enough for this?" ;) )
I myself have always favored the "things I like" route. It's not very productive, but makes for excellent pastime nevertheless.
That being said, it's always educational to look around and try stuff. You should actually try stuff you think you can't do. Builds character >_>