I'll answer this as honest as possible. There are great designers that do not know a spec of code/art/music or anything other than game design. That being said, I do not have respect for a designer who at least cannot prototype their designs in any way. Design shouldn't be a job title. A game developer is what you want to be.

I am terrible at art, but I got my masters in computer science so I could make games. Do I like programming? No, not particularly. But I wanted to make games so I did it. Do my games have great art? Hell no, But I do what I can when I am working alone on a project. People who are "idea guys" only thrive in larger environments where their uselessness can be overlooked by the sheer size of the team they are on. You should take your craft seriously enough to put in the hard work, and attempt to learn useful skills for game development.

i do not like the term of "idea guy" though I think it describes something

I am curious if 'designers' are expected to be competent in the field of realisation

Of course they are. In the end the finished game is all that counts, all vision, dreams, design doc, etc are merely tools to that goal.

The "Idea Guy" type of person are a dime a dozen. Anyone who can say "Wouldn't it be cool if..." is an idea guy. To be considered a designer of any type, you have to go beyond that. You have to make something. This doesn't mean you have to learn a programming language either. Just write it down someplace. Add to it. Revise it. Rewrite it. Make the Design Document. Sketch things out. Think about how things work and how they can be implemented. Make a physical prototype if possible. Even a boardgame version of your idea will help you visualize it. I often fill half a notebook with concepts and logic trees before attempting any given idea at my computer.

Just know that nobody is going to take you seriously until you can show them something that convinces them you are worth their time. That will most likely be (lots of) money, a well written design, or a playable prototype. Talk is cheap. Take action and you can more easily find others that will help you.

So if a designer is considered a person who is inventing a game in a dream
and making some text on it when coming to the ground of doing this this
my be almost unusable - what with that problem ?

whats up with that, eh?

you have 2 kinds of "designers"

1. "idea guys": wouldn't it be cool if...... but no clue how to do it, how hard it is, or if its even possible.

2. real game designers: wouldn't it be cool if...... that would require x,y,and z. and y might be a problem. we'll need to check into that first.

a real designer will be able to think through the implementation to determine if the idea is BS or not.

"idea guys" are very common.

if you have to deal with them, all you can do is evaluate each of their ideas on its own merits, and determine feasibility. if its not feasible, you must nicely explain to them why. eventually they learn what is and is not do-able.

but most soon discover that their cool ideas aren't doable or aren't commercially viable, lost interest, and disappear. they never were serious game designer material to begin with.

the ones who learn and stick with it can become quite valuable. especially if they have a gift for game design, the way one has a gift for writing, painting, sculpting, composing music, programming, or languages. these are all creative right-brain activities.

You don't have to know what something will look like, you have to know what you want it to look like.

meaning if you want to make something spooky you can just make it dark,

then, later on when the game is a bit further and you can test a level,

you don't check whether it's still dark but you check whether it's spooky.(testing is a big part of making a game)

Idea guys aren't real designers though. Has anyone ever worked with a strictly idea guy in a professional setting? I haven't in the 8 years I've been making games for a living.

The pre-production phase of development is used to develop the vision (i.e. ideas) of the game, and prove out the feasibility. This involves breaking down the vision into more involved design documents, using those to generate initial task lists, have engineers/artists/animators/etc. estimate the amount of work needed based on the task lists, revise the vision and design documents until the work almost fits into the schedule, and repeat.

For things like new tech, or key game features pre-production is usually also used to prototype these things. For example, if "being spooky" is a majorly important part of the game, there is no way in hell a good company would start by "making it dark" and then later make it spooky. They would choose a very small section of the game and they would focus all of their attention into proving they could create a spooky atmosphere before investing millions of dollars building a full game when they still don't know how to make it spooky. That's not to say that you never iterate (you do all the time), but if you can't prove you can do it in a focused area, how can you trust yourself to figure it out for a huge game?