Jump to content

  • Log In with Google      Sign In   
  • Create Account


SIGGRAPH 2013 Master Thread

  • You cannot reply to this topic
26 replies to this topic

#21 LimeLaser   Members   -  Reputation: 103

Like
0Likes
Like

Posted 04 August 2013 - 02:01 AM

I was actually kinda moved by the Global Illumination with Radiance Regression Functions clip.  However, it appears scenes are static and have a huge number of vertices.  What do y'all think?



Sponsor:

#22 Hodgman   Moderators   -  Reputation: 27814

Like
0Likes
Like

Posted 04 August 2013 - 02:10 AM

I was actually kinda moved by the Global Illumination with Radiance Regression Functions clip.  However, it appears scenes are static and have a huge number of vertices.  What do y'all think?

Yeah IIRC, it's a technique for dynamic relighting of static scenes (with a large precomutation step). For games with static environments and dynamic lighting, it may be useful...



#23 Chris_F   Members   -  Reputation: 1943

Like
0Likes
Like

Posted 04 August 2013 - 01:19 PM



 



I was actually kinda moved by the Global Illumination with Radiance Regression Functions clip.  However, it appears scenes are static and have a huge number of vertices.  What do y'all think?

Yeah IIRC, it's a technique for dynamic relighting of static scenes (with a large precomutation step). For games with static environments and dynamic lighting, it may be useful...

 

 

I just watched the video. It was impressive.

 

Too impressive. What's the catch? blink.png



#24 Radikalizm   Crossbones+   -  Reputation: 2774

Like
0Likes
Like

Posted 04 August 2013 - 03:08 PM

 



 



I was actually kinda moved by the Global Illumination with Radiance Regression Functions clip.  However, it appears scenes are static and have a huge number of vertices.  What do y'all think?

Yeah IIRC, it's a technique for dynamic relighting of static scenes (with a large precomutation step). For games with static environments and dynamic lighting, it may be useful...

 

 

I just watched the video. It was impressive.

 

Too impressive. What's the catch? blink.png

 

 

I'd call the precomputation step and the static environment requirements a pretty big catch tbh.

While the results are very very nice, it's not exactly a 'cheap' method from what I can tell (although I have to admit I didn't check it out in depth)



#25 Chris_F   Members   -  Reputation: 1943

Like
0Likes
Like

Posted 04 August 2013 - 03:22 PM

I'm not sure I'd call pre-computation or static geometry much of a catch considering that light mapping already has those limitations.



#26 Radikalizm   Crossbones+   -  Reputation: 2774

Like
0Likes
Like

Posted 04 August 2013 - 03:40 PM

I'm not sure I'd call pre-computation or static geometry much of a catch considering that light mapping already has those limitations.

 

True, but the last couple of years research has been advancing towards more completely dynamic solutions, with the voxel cone tracing technique being one of the more impressive ones imo.

 

Compared to these flexible solutions to the global illumination problem, having a static environment requirement is somewhat of an issue in my view. But of course, if your game has mostly static environments this could still be a very nice solution.

 

Also in an age where rapid iteration and on-the-fly modification of content is becoming the norm, precomputation passes aren't exactly a good fit :)



#27 Frenetic Pony   Members   -  Reputation: 1189

Like
0Likes
Like

Posted 04 August 2013 - 03:53 PM

I was actually kinda moved by the Global Illumination with Radiance Regression Functions clip.  However, it appears scenes are static and have a huge number of vertices.  What do y'all think?

 

It looks great! Unfortunately, giving it a quick glance over, it doesn't have anything like a constant cost and each additional light evaluated adds to the cost. But the main problem seems to be that it doesn't scale with scene complexity that well. The more of their "RRF's", which is a section of level or model, that you have to evaluate the more your runtime cost is, and the bigger the "RRF" the runtime expands even faster.

 

Thus a small scenes works very well, big outdoor scenes don't, as it doesn't scale and doesn't handle environment (read, skylighting) well at all. And big outdoor scenes are the major problem for GI to begin with. There are solid enough hacks for short range lightbounce, even Crytek's light propogation volumes can work well enough for that, you can raymarch for reflections and voxelize secondary geometry for secondary shadows. It's the typical huge world that needs a good solution.

 

I was starting look up raytracing into signed distance fields, it makes quite a bit of sense from my initial impressions. If you could combine it with voxel cone tracing, which has a constant overhead cost, scales well with scene complexity, and can easily add lights, then the only thing you'd need to handle are dynamic objects. Which could be handled with something like Sucker Punch's ambient occlusion fields and a hacked impostor for reflections, as you don't care about most dynamic objects influence beyond a small area. Not perfect, but then perfect isn't doable yet.


Edited by Frenetic Pony, 04 August 2013 - 09:39 PM.






PARTNERS