Hi forum.

So I know some C++, more in theory than in practice and was wanting to create a small game, maybe a board game in 3D.

It's quite hard knowing where to start though after the design stage. I am reading GEA at the moment, and I just can't get an overall picture of rendering techniques, for example there are various rendering, algorithms such as phong lighting, and a ton of other graphical techniques such as shadow maps, shadow volume, and then shaders. There seem to be so many different ways of doing rendering that I am lost as where to start.

I can create OGL/DX window, but then that's just copy/paste. Where do I go after that.

If anyone wants to help feel free to pm me. It would be for a learning curve to improve C++ skills and learn about 2D/3D graphics.

So I know some C++, more in theory than in practice and was wanting to create a small game, maybe a board game in 3D

As it was said above, doing 3D now is not a good idea. But:

-You dont have to start with a 2D API, you could do 2D graphics in the 3D space by ignoring the Z coordinate. Then start to use it later when you are comfortable drawing in the XY plane.(You wont have to throw away your code.)

-If you do want to use the modern OpenGL, forget about "shaders" and "buffers" just use one shader and one VBO at first.

There seem to be so many different ways of doing rendering that I am lost as where to start.

-rendering is basically projecting and drawing triangles to the screen. textrues, fog, bumpmap, shadowmap ....thats bonus. Most people start out drawing red and blue rectangles.

I can create OGL/DX window, but then that's just copy/paste. Where do I go after that.

-after that find a "Hello-triangle!" tutorial and copy-paste that too:)

I just can't get an overall picture of rendering techniques, for example there are various rendering, algorithms such as phong lighting, and a ton of other graphical techniques such as shadow maps, shadow volume, and then shaders. There seem to be so many different ways of doing rendering that I am lost as where to start.

So called 'phong lighting' is also called lighting per-pixel. It delivers better results than per-vertex lighting (called Gouraud lighting or shading). When light hits a surface in the real world it may appear more shiny or less shiny depending on the angle of the surface. In 3d graphics this is faked by giving each vertex a vector called a 'normal' that tells you which way it is pointed. So per the lighting calculations that vertex may be more or less shiny depending on the angle of the light and the normal attached to it. This can suck if your vertexes are not evenly spaced, there are too tightly packed, or too sparse. The solution they came up with is to interpolate the 'normal's between the vertexes and run the calculations for every pixel (aka fragment). Hence, it is called per-pixel lighting.

Shadows are a PITA. In 3d graphics lighting geometry is relatively simple. Casting shadows well and efficiently is much more difficult in non-trivial cases. I'm still new to graphics...so all I know of shadow mapping is where you calculate where the shadows are (maybe real time via render to texture or ahead of time in some tool) and put into a texture which is applied to the geometry in the scene that is going to receive the shadows via texture blending.

I would focus on making a 2D game though first though...which you can do with OpenGL by the way. You just won't be using the 'z' coordinate or the depth buffer. Even if you are using a pre-canned shader 3d graphics can be a bit like drinking water from a fire hose.