Which route should I take?

[{\/\/} King {\/\/}]

Hello all, I know that this question has been beaten to death, but not in the context of certain questions I have and I am just trying to make sure I use my time as efficiently as possible.

 

So my question is this: I am an intermediate level c++ programmer and a decent sprite artist (imho) I am going into game development with myself and hopefully eventually a few others. I am unsure of what path I should take to get some of the ideas I have for games off of the ground.

 

Without getting too heftily into the details, I enjoy the camera orientation of Metroidvanias and Battletoads, I like the use of elements from rpg games, I have ideas for interesting system implementation into games (for example, a submenu from your main menu that is a bookcase that is a representation of your character's training of various skills. The book case can be replaced with various other bookcases (3 spaces instead of 2, etc..) and the books that go into the spaces each contain a few miniature skill trees pertaining to specific subsets of skills such as a lancer book, or a book with fire skills etc. that could be found as enemy drops ala' castlevania. So that the player can sort of create their own sphere grid (ff10) with some of the skills being passive changes to primary attack that can stack and effect each other (think the binding of issac if it were a metroidvania for this aspect) and the game would require some sort of p2p networking to get groups of up to 4 to load to stages similarly to the way Monster Hunter Freedom Unite or Left 4 Dead or Diablo 2 (any of those models really)) 

 

I realize this is a massive undertaking, and is not the first idea I am pursuing by a long shot (due to those and other complications) but with these ideas being my motivator, would it be better for me to use a framework like monogame and then import the libraries necessary to do the missing core functions (like box2d, etc) or use Unity and try to teach myself C#? (being that I have moderate proficiency in C++) OR just write my own engine from scratch in C++?

 

I have heard so many conflicting viewpoints about the value of programming your own engine from scratch on monogame, as well as conflicting reports as to the limitations of Unity.

 

Thank you for anyone who reads this massive wall of text and provides me any insight.

Edited July 7, 2016 by {\/\/} King {\/\/}

[{\/\/} King {\/\/}]

Just in case anyone else has any kind of similar questions (IE: if I know c++ what should I use to get the best bang for my buck time wise (to create a 2 dimensional game), while still being streamlined enough that I'm not reinventing the wheel or recreating APIs from scratch that have been made 1000 times, while also able to get in closer to the metal to optimize and/or write extraneous functionality for "gameplay systems" [non-generic game functionality created from scratch])

 

The answer I have gotten through other forums almost across the board has pretty much been this: "Unreal engine 4 is free now, use that if you already have precursory knowledge of C++ (I do on what I would consider to be a low intermediate level, but I learn and practice daily) and 100% across the board I have been warned: "Don't use Monogame or any framework or even worse program your own engine from scratch via C+, to do so adds countless hours to the amount of programming necessary before you can even begin working on your game, which kills drive and passion (and stymies creativity) "Make games, not engines." (is a phrase that comes up A TON)"

 

...Although the response I got from the Unity forum was more along the lines of "If you chose Unity you would have to learn C#, but it is really similar to C++ and its always fun to learn a new language, plus 40$ / month is nothing, I'm sure as an indie developer bankrolling everything on your own and doing all of the work essentially without paying yourself while also paying adobe a similar monthly, paying regular-a## life bills, and helping to support your elderly and under-prepared parents you would love another monthly bill!" I much prefer the investment model of UE4, as if the code you give me for free makes me any money any more easily than I could have done so on my own I have no problem giving you a percentile (5% I think?) But to have a $1500 entry barrier or $40 monthly entry barrier is still an entry barrier, and I have minimum resources until I can get a prototype working for a kickstarter or other such investment, so I have begun pushing my research in that direction.