Assets [Unity Awards 2018 - BEST ARTISTIC TOOL Winner] Aura - Volumetric Lighting

Hi! Technically, you should really decouple several concepts here : surface lighting and volumetric lighting.

The fact that your bottom box is not lit is because of the shadows in the surface lighting system.

The volumetric lighting is indeed about lighting the "air"

Thank you for your compliments

 

I was just confirmed by the Asset Store team that Aura would be accepted on the Asset Store along the GitHub page

See you soon!

 

Thanks for the info. And congrats on the asset store submission!

 

Thank you!

 

I stopped using Unity a long time ago, and switched to Unreal. One of the features that prevented me from going back to unity were the volumetric stuff UE4 has, which I cannot live without.

However, after seeing your plugin, and the fact that it will be free, I immediately thought about going back to unity just because of your work and how cool the results look (among other things like writtable shaders and things like that).

I have some questions though:

1. Will it be compatible with enlighten/global GI?
2. Will it be compatible with baked lightning / probes? I mean something like UE4's volumetric lightmaps.
3. What are the plans for 2018.1+?

 

Hi! Wow I am honored that you'd consider switching back to Unity after you saw Aura. I hope it'll live your expectations.
It's a fact that Unreal has currently more advanced default rendering features but we should keep in mind that Rome wasn't built in a day and that Unreal is much older. I am glad if I can contribute to step Unity onward the path of rendering.

1. I have a plan of investigating in this indeed. For information I already had GI result by injecting SEGI (https://github.com/sonicether/SEGI) data into Aura.
2. Same
3. Continued development

Cheerz!

 

Hello！ Aura is Amazing and make me exciting！
Can‘t wait to use it in my project！

Does it support 2017.1.3f1？

 

The current state is from 2017.2 up (because they introduced multi threading compute shader compilation) but if you are ready to put your hands dirty in the code and step back from some optimizations, it will work by design as soon as 5.4 (because it needs Texture2DArrays)

 

The feature set you got there is really great! Making it open-source is really generous of you I have to say. I'll need to talk to my team members about this.

Just a quick question: I have a project that is really dependent on volumetric lighting in a mixed indoor/outdoor environment. As you've written before, with an injection volume in the scene it should be possible to "punch out" the fog from a global volume. Would it be possible to extend the shapes you mention in the first post with an arbitrary mesh, or is that an inherent limitation? I would require seamless transitions from outdoors to indoors, but in some cases it would be very cumbersome to model the shapes of the buildings with those primitives.

Greetings,
Jan

 

Hi! Thank you for your compliment.

Shapes are parametric, meaning they are not arbitrary. Note that you can combine any number of shapes you want.
There might be a way to use arbitrary shapes but it's not the case for now.

 

Of course, I should've known better. That was the artist in me speaking.

Sure! It's good enough that it's possible at all. Maybe we'll be able to integrate that ourselves if the need arises.

 

Hopefully you'll shared them as well

 

Looks like an awesome asset! Can't wait to get my hands on it!

Would have loved to have this for GGJ this weekend

 

We can arrange that, send ma a PM with the details

 

This looks so great. Can't wait to see how it'll look in my game. Thanks a lot and great job.

 

Thanks, and my pleasure!
I can't wait to see how it looks in your project as wel

 

Wow, this looks beautiful.

I've got a project which puts a lot of strain on my existing volumetric lighting. I have a bunch of huge fishtanks up on cliffs and an ocean with waves, and I'd like to represent those air-to-water transitions as well as possible in my lighting.



In my existing solution, approximating them with particles is weird, approximating them with injection primitives is annoying to set up and way too slow, neither one does well with waves, and no matter what I have problems with volumetrics not being distorted by refraction.



My best ideas for how to go about this are to integrate the volumetrics more tightly into my water shaders somehow, maybe write a special water depth texture and send that to the volumetrics, but I'm not really sure.



Aura looks versatile and, being open source, it seems like it could be a good place for me to start.



I've attached some screenshots of the kinds of environment I'm dealing with right now. Does Aura have any features that could help my water problems? Do you think it could be adapted? Do you have any ideas how?


I'm interested in any thoughts from anyone about this, I'm constantly working to improve my water effects. (These screenshots have several iterations of water effects in them.)

 

Do you have a date?

 

Pretty sure before long, free or not, well-meaning people will drag the author down with support requests so having a paid support option is not unusual and these improvements make their way back down to github.... just a possibility.

 

Is the SEGI -> Aura bridge something that'll be made available? That would be awesome!

 

Hey guys!

Being able to figure the difference of environments is a constant struggle in video games. Aura surely have the possibility to help in this with injection volumes and locally modifying the color of the light, the density of the floating micro particles and the anisotropy. For the refraction, you'll have to put your hands in the dirt and modify the shaders. But this is a good start

Well, my birthday is February 10th, I might bring some digital cake on that day

I do agree that I won't exhaust myself in exaggerated support, there will be FAQ, documentations and, hopefully, tutorials for the how-to questions. If one needs custom development, let's be realistic guys, I won't be able to do all of them for free ("stay nice but not dumb" ) because it will just take a lot of time.

I'll possibly make a branch for the example but nothing in the master I think

 

On/Off comparison with Aura
Aura will be released in February 2018 for FREE on the Asset Store and on GitHub.

Stay tuned @raphernaelsten

 

AWESOME !!!

 

Thanks

 

This looks really great. Not only that, but it appears to have some behaviours I haven't seen in the other open source volumetric lighting solutions on Github.

 

Was it a good idea or not, I wanted to challenge myself and develop it out-of-the-box and not be influenced by the existing solutions around...

 

Can‘t wait

 

Thanks

 

Awesome comparison!
Can't wait to play with it.

No idea of this was already mentioned. But could you give us a hint about the performance for the final version?

 

That is pretty dang awesome. if you're in the mood to do another one of these videos, i'd love to see some indoor settings. Like, sci-fi XD
I'm being greedy, I know hah.

 

The quality shown on the video runs @~130fps on my gtx960 but it is a very fine detailed for the sake of the video. Lowering the quality would be barely visible.

Sorry I am too busy, rushing for the initial submission

 

This is absolutely terrific, especially the injection volume functions.
Just one question; Like man I don't want to be one of those entitled assholes who go DO THIS DO THAT on free asset threads (above all, free assets with this amount of quality and dedication poured into), but as someone who is not a graphics programmer this doesn't seem like that big of a stretch... could you possibly make it work with Keijiro's KvantStream GPU particles (https://github.com/keijiro/KvantStream) ? They could be so much more efficient for doing some atmospheric stuff than baseline unity particles, like dust and snow especially.

 

No need to do specific work, it'll basically work if you use the particle shaders I provide or if you modify his to use Aura

 

Thing is, KvantStream doesn't react to light out of the box, it has just an unlit line shader to do all the work. I guess there could be just a way to sample the lighting data from Aura volumes, but ehh... anything more complex than changing tint or the blend mode of a shader is far from the reach of my shader writing knowledge (so far) PM me if you as a mortal human being have any more goodwill left in your body.
Hyped for the release regardless

 

After you release the asset please post a donate link somewhere I'm looking forward to playing with this!

 

Is there a posisbility that there will be a subsampling slider so that it might even work on higher end mobile devices such as iphone 7 plus?

but at lower resolution settings

 

It took me less than 10 minutes (including the download of the package)

 

Yeah I've been asked many times. The will be a donation button. Thanks.

Aura is very flexible performance wise. The rendering itself actually is only a few lines in a post process shader Up to you if you want to render the post process halfres

 

God damn it. What a man. Thank you, seriously dude.

 

Also, just curious, what amount of Aura is in compute? Like, is it from ground up DX11+ or can I strip down some of the fancier features to make it work with DX9?

 

The core is Compute, never gonna work on lower API. Sorry :/

 

Nah, it's cool. Even better actually.

 

Yeah, and its the right time for these requirements. Compute shaders have taken a long time to truly come of age, Unity did their bit with things like Metal compute shader support which I always rave about, and even their new HD scriptable rendering pipeline requires compute shaders.

 

Amazing, thanks so much!! Do the point / spot lights work with SEGI through Aura, or were you sticking to area lights using emissive materials?

 

No... Aura just integrates SEGI's results... not the other way around...

 

Ok, I was wondering how it would work. This makes much more sense now!

 

It's February! Woo! Any day now right?

 

Can't wait for this. : D <3

 

Right

 

Looking really cool!! Keep up the good work!!