Ok why do you wanna make your own game engine again?
Also what do you plan on making your game about?

22 hours ago, SIr Pep said:

Ok why do you wanna make your own game engine again?
Also what do you plan on making your game about?

Why my own game engine:

Experience first of all. I just want know how graphics libraries works. OpenGL, DirectX, Vulcano in the future. I need to know geometrical and math ways to make standart things and, maybe not so standart.

I want to kniw how to write shaders, want use all popular operation sistems and other cool things.

I just love it! I love code and creating of anything.

What about my game is:

This is a mmorpg, but stop laughting:) I think I realy can do it. I have a girlfriend who makes great 3d models and arts and I have 8 ears of programming experience. And I want to do it! Not so fust but I finish it.

It's will be not so standart mmo because you can't play it solo

A lot of people would advise you not to make your own game engine because it's so time consuming, but if you want to know how it work and have 8 years of programming experience that's up to you man
As long as you have a clear cut picture of what you want to do then I'd say go get your dreams!
Also read lots!!! I can't tell you if this is realistic for you and your girlfriend or not because as far as I know 3d isn't as easy...

God bless your development and I hope you have a great day

Also check out @Septopus... He seems to be doing something similar, maybe he could help you

By all means write your own engine, but as Pep suggests, it will take you a long time as it is a big task.

You mention learning OpenGL and DirectX libraries as examples, but be mindful that graphics is only one aspect of the engine.

You can easily go down deep rabbit holes with loading 3d models - this can be challenging based on the specific tools you use and their export functions (Blender vs Maya vs 3dsMax) and what formats you want to code for - i spent a solid month coding a Collada file format interpreter and animation process flow - and another 3 weeks building a voxel based terrain system (not minecraft styled, more like space engineers), loading raw textures (BMP? JPG? PNG? all file formats differ in particular ways), particularly if you're looking to support both OpenGL and DirectX  - there are some truely fundamental differences that can trip you up (LHS vs RHS notations and matrix differences is one example). 

You'll probably spend more time on building a scripting language for your game, and building in the I/O aspects. Owing to you looking to build an MMO, you will need to have strong networking knowledge, coupled with strong multithreading understanding. Building an AI system will also be fun (my current project) and i would recommend starting with a state engine approach to keep things a little more simple.

UX design knowledge (User Experience) is important for an MMO, which will bring this up from the 'coding layer' to the game design and user interface considerations. 

Without knowing what you have, i would recommend building a good library of books on the various topics. I have a former nintendo library Collada book, books on AI, a couple of books on Physics, vector mathematics, linear algebra, game based networking and multi-threading.

You will also need to consider the narrative of your world that you're building, interactions between actors, item design and theme. Politics, etc. 

Haha, yeah I'm working on an MMO, but I'm not also writing my own engine.   Just using Unity over here in the corner... haha 

Welcome to the community @Aleh Lipka!  Looking forward to seeing what you come up with!

As someone working on my own game engine and MMORPG (don't laugh back ?), I'll make a couple of suggestions.  I wouldn't just start off to build a "game engine" first and then the game, build the game and the engine at the same time.  Also don't start putting in features just for the sake of it.  Put them in as you need them. Sometimes it pays to look a little bit ahead, but don't add things way in advance that you may never need, or you may have to re-implement when you figure out what you really need.

Also remember you won't get things perfect the first time through. I know I've made a lot of changes in my code and I'll make many more.  So don't worry about it if you have to throw out code. It's part of the process.  I would do some initial reading and then try to start in on the coding. Getting things started is the main thing.