In Blender (today …) you have multiple very-powerful built-in rendering systems: BI, Cycles, OpenGL, Game.
(And all of this is in addition to interfaces to several external rendering systems.)
Each of these built-in renderers takes its own (very different!) approach to the same fundamental objective of “producing 3D images (and portions thereof).” Yet, each one fits into the same data-processing pipeline. And, each one shares 3D-world aspects such as camera, 3D coordinate system, and so on. Thus, they can all produce renders which will coincide exactly with the work-products of the others … and this is very important.
When you look at any final scene in a movie, what you’re looking at probably isn’t “the single final-output of one render.” Almost certainly, it is the combined (composited …) output of many different renders, any of which could have been done in several different ways. “And then, on top of all that, it’s the product of many committee meetings.” :yes:
So … no strategy for rendering a 3D scene (or, portion thereof …) can ever be called, “useless.” The notion truly has no meaning. It’s just another blade in the Swiss Army® Knife that’s called “Blender.” Part of your creative (and project-management) decision making has to do with which blade(s) to use, when, and why. All of the “built-in” rendering strategies, very-different though they all are, are nevertheless “readily at-hand,” “under the same roof,” and “easily compatible.” (What’s not to love?)
The audience, when they stare open-mouthed at your amazing graphic artistry, will never know, or care, just how you did it. (But they will know if you didn’t meet the deadline, because they’ll be staring at a blank screen.)