Don't confuse small ideas with small execution. Small ideas rarely works. Also don't confuse number of features with a deep & complex gameplay.

Check my "WizTowerSim": http://www.silverlemur.com/minigames/
Is this game simple or complex? How long would it take you to implement something like that?

You can make a complex & interesting game in a few days/weeks. If you use dirty tricks of course

It's funny because the little game I have right now looks a bit like this or I should say has the same basics. It's some stuff where you build your base and every random amount of times you get a random attack and you have to survive it. At least that's the final goal but it's not finished yet (almost tho when it comes to programming).

I think the main idea is not bad but random events part of it sucks, I wanted to make it multiplayer but I am not good enough at coding networks. So this is the kind of thing that pisses me off a bit.

Do not escalate No multiplayer, you are not looking how to add yourself more work but how to remove some work
If you have a working concept, go for it. Do not add unneeded things.

"No multiplayer" isn't a good general advice. It should be more like "no networking"!
"Multiplayer" doesn't automatically imply networking, since local multiplayers are possible, too. Also, many multiplayer games are much easier to create, since you don't have to implement an AI, you most likely won't need that much content, but the game could still be fun to play.

From the coding perspective "no multiplayer" is a good advice, it makes games a lot easier.

From the content perspective "multiplayer" is a good advice, because you don't need as defined content as in other games.

And multiplayer is a good way to market your game (many people love mulitplayer features).

But one major drawback of a focus on multiplayer is, that you need to build up a working community really quickly, and this is really hard. This would only work if you have either a dedicated fan community (Fan of what ? chicken-egg-dilemma) or enough solo content to attract a community without the multiplayer feature (in this case multiplayer is put ontop of an existing solo game => more work).

The problem with this, is I fear it would limit me for the sound system which is why I do games in the first place. But I don't know much about mods, I'll look into it.

I don't know your special requirements, if you have one or more games which would satisfy your sound system requirements, then check if they support modding. Here is a large database about mods, hope it helps.

Oh I know this website I watch it a lot to get some projects. What I meant for example was, with Unity or UE4 you can add Wwise or Fmod which are audio middleware and help a lot to do great audio. UDK (and UE4) also have a pretty nice audio tools to begin with, but the Unity ones suck. I am pretty confident that doing mods won't allow me to use my middlewares and so basically I need to figure out if stuff used for mods (I don't know the names but for example the Valve one) have good audio tools.

I am not going to quote everyone again because it would end up being a huge post but :

Gian-Reto : Thank you, that's really nice advices. Maybe I should read a game design book myself.

For the multiplayer, I am not escalating, it's something I cut from my features in the first place. But I like the idea of multiplayer without network, for this game it will definitly not fit but I will probably use that for the next one. Like 2 players turn by turn, it will save me from the AI + the network.

Network and AI are the reason why I can't really do games I want most of the time. I like to do strategy games and I am not capable of doing a good enough AI to let the people play against it (like you can in SC2 for example), but also I am not capable of doing the network. So I am kinda stuck with random events and stuff like this.

Network and AI are the reason why I can't really do games I want most of the time. I like to do strategy games and I am not capable of doing a good enough AI to let the people play against it (like you can in SC2 for example), but also I am not capable of doing the network. So I am kinda stuck with random events and stuff like this.

...therefor you should create a mod. SC2 does support modding and as far as I can see with my google, it uses FMOD too + network support + RTS AI out-of-the-box.

Really, many successful games has been started as mods (CS,DayZ,TF,Dota,...)

Gian-Reto : Thank you, that's really nice advices. Maybe I should read a game design book myself..

I can only recommend that. It is really interesting to see how people in other profession go about their tasks, and more so when it comes to a generally rather hard to grasp job as game design.

To me, like to other people I guess, they were a mixture of slackers ("what the hell are they doing all day? Just playing games?") and magicians ("How the hell did they come up with THAT idea? How can this all be so perfectly balanced?")... so it is quite interesting to get a glimpse of how these guys do their job...

On top of that, every programer that writes his own small game for a portfolio or fun will certainly find out quickly how important a good designer can be, even AFTER the initial idea has been laid down. It is exactly the situation you are in at the moment.

A book I found quite good is this: http://www.amazon.com/Art-Game-Design-Lenses-Second/dp/1466598646/ref=sr_1_1?ie=UTF8&qid=1424355098&sr=8-1&keywords=a+book+of+lenses

It tries to bring a little bit more structure to hard to grasp concepts, and I found it understandable for me as an engineer without much game design expierience prior.

I also got this book, which is much shorter (and cheaper): http://www.amazon.com/Theory-Game-Design-Raph-Koster/dp/1449363210/ref=pd_bxgy_b_text_z

Though I don't remember if I thought it was any good, reviews are favourable