Shinen is using triple buffering for the gbuffer on Fast Racing Neo, bandwidth is not a problem

[zarx]

megafenix said:

the LOD and the passes for the material(also mentioned there by gpu gems)are few of the examples i could put, there are many other examples i can put here were you can see that lowering the passes saves up shader power that can be used for other rendering purposes, but will take some space here that i am not sure everybody will bother to read

And yes, of course that defered rendering is not perfect and besides the problems with huge bandwidth requirements you also have issues like alpha blending and msaa, but there are good solutions for that, instead of msaa for example you can use temporal antialaising, which laos is cheaper than msaa and is better suited with defered rendering, FXAA or Fast Approximate Anti-Aliasing is also a good option

As for the transparency, well you can combine defered and forward rendering, forward rendering would be just used for the parts where you need transparency or can use some other solutions like the ones found here

Again that example is just using lod to reduce shader detail on distant objects. Yes it is reducing the number of passes but it's just not doing that shading on distant objects. So it's not "doing the work in less passes makes it faster" it's "not doing certain effects on distant objects, makes it faster because you are doing passes on fewer objects" you are still doing the same number of passes on near objects. So not really relevent to the topic we were discussing.Which was often times splitting the work up into seperate passes is often faster. Once again you have posted a responce that is completly irrelevent to the topic on hand because it uses some of the same words. And to reiterate I am not saying their aren't cases where batching and combining operations into one pass doesn't make things faster in some cases. Just that it is not a hard and fast rule, and that there are in fact many cases where the opposite is true and splitting up tasks into more simpler passes is often the better aproach.

And yes that is exactly what I said about deferred shading, I am not sure why you are repeating the point I just made.

 

Anyway I am out, peace.