So I’ll start it this month… hopefully someone else isn’t doing it too right now.
I spent the last day working on a waterfall effect using shader-side animation of some shader-side billboarded quads. I’ve reached the end of my self-imposed tweaking deadline and this is what I have so far:
Description from the video page:
It's not 'perfect' but I've reached my "tweaking deadline". Things like this, I have to give myself a deadline for endless tweaking otherwise it takes too much time. It's obviously a waterfall now and I can swing back and tweak it some more when there is more of a game. The waterfall is one of the critical story elements and helps tie together some of the other water-related stuff I'm working on.
This particle effect is done as a static mesh using a shader-side particle effect and billboard rotated quads (in shader). It’s the same mesh layered a few times with different textures… though it could have just as easily been different meshes, too.
Three layers at the moment that don’t always sort like they should and you see them pop in front. Good enough for now.
This is using ideas that I designed for Mythruna, so when it’s all properly working then this will definitely be a part of Mythruna’s water system.
Unfortunately, it is probably too slow to be of any practical use.
Hi, looks nice, but in what is it any different than this excellent filter made last year by @thetoucher? I’m currently using it on tens of thousands of geometries and it still runs hundreds of fps a second on a recent GPU.
@.Ben. said:
Hi, looks nice, but in what is it any different than this excellent filter made last year by @thetoucher? I'm currently using it on tens of thousands of geometries and it still runs hundreds of fps a second on a recent GPU.
It is quite different because I have forgotten about that one My one is using brute-force raytracing through light space, which is obviously non-efficient.
I will try to adapt @thetoucher one - it won’t be drop-in replacement, as I have everything in deferred rendering, but hopefully should be simple enough.
@kwando said:
I think I spotted the "brokenCube"-model from dmonkey in there ^^
You would probably spot a lot more common things in the code
I took your code, rewritten it to do deferred lights (3-pass approach) instead of deferred rendering, then rewritten it again back to to 2-pass approach which you have used originally… but after all that, there is still dm_decode(); at the start of almost every shader
So I have been working on this for quiet some time…
I know that the idea of creating your own editor for the MonkeyEngine has been controversial, but it really is a fun project and not only did I grasp the concept of oo programming, but also got to know the engine itself
I might just use this for saying thank you so much to all the developers and contributors.
This engine gave me much motivation to learn how to program and now that I am in college, I am really benefiting from that.
@taintedpyro813 Thank you, but no, I still have to fix a lot of stuff and implement some last features.
And since I thought that I was almost done half a year ago, I doubt that I will finish it soon.
But one of our tasks is to create a webpage, where we can present our college related and private projects. So it will get there eventually… I hope.
@benkibitzer said:
So I have been working on this for quiet some time...
I know that the idea of creating your own editor for the MonkeyEngine has been controversial, but it really is a fun project and not only did I grasp the concept of oo programming, but also got to know the engine itself
I might just use this for saying thank you so much to all the developers and contributors.
This engine gave me much motivation to learn how to program and now that I am in college, I am really benefiting from that.
Wow that looks like a sh**load of work! Nice job
@BigBob said:
Figured I'd take part in this month's screenshot thread...
Finally managed to get SSAO working in my deferred renderer. There are still few light horizontal lines on top of the screen (they are fixed to screen not scene… out of ideas whey they are there) and I need to add temporal blur to reduce flickering artifacts on some of the straight lines, but I’m happy with effect in general.
My one is using brute-force raytracing through light space, which is obviously non-efficient.
