Spiro thank you so much for your time answering my questions. I really love your passion about game programming and I never said that you say something wrong. You have a valid point. No argue here, what you say is right. However, following your path requires devotion, time and patience and given that I am not a professional game programmer nor I want to become one I am looking for alternatives/shortcuts that will lead me to the final result faster than some others. You see this is going to be hobby project that I am going to accomplish after-work hours, which means that I have little time and too many irons on fire.
Unfortunately, there are no shortcuts to learn how to make games. If there were, we all would have found them by now and no one would be telling you to make Tetris. It's going to require devotion, time and patience whether you want to be a professional or not. I'm a hobbyist and I've been playing around with it for just about 20 years, since my late-twenties. Given my age when I started, I was looking for shortcuts, too. I was a busy guy, with a full-time job, a new wife, and other commitments. I never found them. I wasted quite a bit of time, though, and very nearly gave it all up until I finally buckled down and started doing what I should have done from the beginning. And I can't count the number of people I've seen starting their dream games right out of the gate on this site, who then disappeared with nothing to show for it.
Starting with Pong or Tetris gives you an *achievable* goal toward which to direct the baby steps you need to take. An RPG of the scale which you want to develop is *not* an achievable goal for you at this point. You will very likely give it up in frustration long before you see it complete. Pong and Tetris allow you to learn the fundamentals that you need for any game you develop while still finding yourself with a completed game in a relatively short time. The motivational impact of finishing your first game is tremendous. Then you do it again with Asteroids or Pac-Man. Then you do it again with a simple platformer. Each completed game adds to your knowledge, confidence and motivation. Eventually, the day will come when you are ready to start on your dream game.
Please, don't try to rush. Buckle up and prepare for a long haul. In the end, it will be more rewarding.
Totally on the money. There are no shortcuts, and there are no failures (At least as long as you can still pay the bills because you are doing game dev as a hobby).
Everything is a learning expierience, and you usually learn more from your mistakes than from your successes. Looking for a formula that guarantees success will a) waste a lot of your time because even if that formula would exists, finding it would be a very longwinding process, and b) would make your game all the less creative because you are now creating something like a "me too" product (avoiding the time costly pre production where most of the cool stuff is tested, and some of it is thrown away because it didn't work).
To expand on Aldacrons point above:
You CAN ignore common wisdom and start with a "top down" approach. Try to develop your dream game and cut it down to size when you see its not viable. That means a lot more frustration and much less success along the way though. And you need the ability to both push through when things get boring, AND recognize when the effort needed becomes too much...
Throwing away or at least shelving many days and weeks of work is never a pleasant feeling, but in game development is very necessary even for the pros. If you don't like that, you should not try your hand at game development, for you WILL run into dead ends and have to shelf at least parts of your work from time to time.
Now, if you can live with it but are not fond of shelving many hours of work, do the usually recommended "bottom up" learning approach, start with pac man and tetris, and work through the gaming classics in chronological order.
Will you also "waste" many hours in completing these games that do not interest you? Yes. Will you learn more about modern 2D/3D game development than when jumping in the deep end? Probably not. Will you have a stronger foundation and more courage to tackle more ambitious game development projects built up when you finished your tour in a year or two? Most probably yes. Will you have multiple finished game under your belt by the time your try your hands at your dream game? I would hope so!
The most important thing is: you will have learned the needed basics to really tackle your dream game without the frustration of many dead ends if you start with it today. AND you will have finished projects, and if you can, also released them to the public.
Which gives you valuable expierience in often underestimated parts of game development: Project management (even if its a single person project, keeping track of your project is something many people miss), and releasing a game / marketing (after all, if nobody plays your game in the end, even if its an opensource project, you most probably will not be happy about it).
If you love to bang your head against walls repeatedly, have no problems with dead ends and throwing away parts of your work if needed, and do not mind slaving away for years on your projects without a finish line in sight, then go ahead, take the shortcut. Its also a viable route for people very resistant to frustration... 
just keep in mind, IF you ever need street creds in game developer circles, a lot will come down to your released games. An impressive looking video with pre-alpha gameplay footage might be awesome, but until you release your game, people will not give you much credits for it.
If you look for people to work together on a project, or for crowd funding for example that street cred can make or break your potential relationship with potential partners / backers.
So, where should I start ? Can anybody give me a path to follow that the final destination is an RPG like The Legend of Zelda: A Link to the Past ?
Why isn’t “program constantly, make progressively larger games” good enough?
Why do you think there is a specific path you have to take to get to the ability necessary to program The Legend of Zelda: A Link to the Past?
Do you think only the 3 programmers who followed a specific diet can program this game?
Diets may differ, but people tend to be equally healthy at the end of them.
Just program.
It really doesn’t matter where. Every path leads to the ability to create an action-adventure game like The Legend of Zelda: A Link to the Past. Patience, skill, experience, planning, etc., all come with time. There are no shortcuts. You need to learn the skill of programming over a long duration of time, meanwhile also planning etc., which requires patience and practice. Then, someday, you can make a clone of The Legend of Zelda: A Link to the Past.
I already suggested a Tetris clone. Why are you not making one right now?
L. Spiro
Very good advice...
Trying to develop a game on your own when you are not good at arts, do not like to program and have no budget for hiring artists and programmers is like trying to fly without a plane... no matter how much you move your arms, you will not lift off.
If you are not cut out for programming or do not like it enough to learn it, then by definition you need to find someone who does the programming. The best idea is to hire someone, I think many people also working as hobbiyists or indies have found that to be the best way to do it, there are other options, but it will be harder finding the necessary skills and getting anything out of it.
If you want to do the programming yourself, you better learn to like programming, if you not already are liking it. Fo you will spent A LOT of time programming to get your game done... and even more while learning the skills needed. You shouldn't need a destination, for if the journey is not the reward for you, you will have a VERY LONG, UNREWARDING journey ahead of you!
