Vae_Victis
Banned
The thing is, "ray tracing" is not one defined effect, like "anisotropic filtering" or "screen space reflections"; it's a calculation system that you can apply to a lot of different things.
What you are actually looking at with so-called "ray tracing" should be classified as "hardware-accelerated ray tracing", which then in practice is usually used for "full scene path tracing", or "global illumination"; which are particular implementations of ray tracing that are very computationally intensive, and therefore require hardware acceleration. Screen space reflections, for example, are a form of "ray tracing", but they are light enough that you don't need dedicated hardware to implement them.
On a similar vein, you could do 3D in games before hardware-accelerated 3D, but things could only go so far.
"Ray tracing" is just a meaningless marketing buzzword, the point is what you do with it, how resource-intensive it is, and what the end result is. Which, of course, all need to be judge on a case-by-case basis. IMHO the current hardware for ray tracing is still too immature for a solid paradigm shift on the development side, and the technique behind it to make it really and consistently impactful for the end user is not there yet.
What will happen in the future is very hard to tell right now, it could hit a brick ceiling hardware-side and remain relegated mostly to some specific "support" functions in visuals, it could completely replace traditional light rendering in games in the span of 5-10 years, or some other alternative solution we'll come up with in the meantime could make it go the way of ellipsoid-based 3D graphics.
What you are actually looking at with so-called "ray tracing" should be classified as "hardware-accelerated ray tracing", which then in practice is usually used for "full scene path tracing", or "global illumination"; which are particular implementations of ray tracing that are very computationally intensive, and therefore require hardware acceleration. Screen space reflections, for example, are a form of "ray tracing", but they are light enough that you don't need dedicated hardware to implement them.
On a similar vein, you could do 3D in games before hardware-accelerated 3D, but things could only go so far.
"Ray tracing" is just a meaningless marketing buzzword, the point is what you do with it, how resource-intensive it is, and what the end result is. Which, of course, all need to be judge on a case-by-case basis. IMHO the current hardware for ray tracing is still too immature for a solid paradigm shift on the development side, and the technique behind it to make it really and consistently impactful for the end user is not there yet.
What will happen in the future is very hard to tell right now, it could hit a brick ceiling hardware-side and remain relegated mostly to some specific "support" functions in visuals, it could completely replace traditional light rendering in games in the span of 5-10 years, or some other alternative solution we'll come up with in the meantime could make it go the way of ellipsoid-based 3D graphics.
