Interstellar Black Hole (Gargantua)

While messing about with different materials and shaders to create a lensing effect, I found something that I believe to be relatively accurate, though it may not be. Settings are as follows:

  1. Icosphere with flipped normals
  2. Refraction shader with “Sharp” distribution
  3. Full white color
  4. Index of Refraction set to Layer Weight - Facing and a Blend value of around 0.89

This seemed to work for me with no postprocessing. This may be a solution posted in the above link, but I have no idea due to the link not yet having been approved.

1 Like

whoooooooooooooooooooooaaaaaaaaaaaaaaaaaahhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh

(fart)

C&C welcome


No planet yet, but render time is 25 minutes, so there is some hope for animation. Couldn’t get the distortion thinner without loosing inner ring. Still lot of work to do.

Nodes screen for accretion disk. I guess it came out a bit complicated :slight_smile:


2 Likes

looks at node setup Mind blown!
Great job. :slight_smile:

holy crap! that node setup looks like a frakin maze!!! hahahaa

Step in a right way, i think. C&C welcome. Edit: added planet


And guys, don’t hesitate to post your tests with distortion setup from blendswap

1 Like


25 samples, 20 minutes. Just a test.

1 Like

Wow that is looking amazing ! Well done !
The accretion disk node setup is really complicated indeed, but the results really justifies it.
Few posts ago, you posted your method for the black hole lensing. Elegant solution with the refractive planes! I had an idea with a fresnel-driven IOR, refractive sphere. Yours seems to work better though, as the results shows :slight_smile: It’s just that the black hole distortion animation video caught my eye … I don’t think the back of the accretion disk would be showing that much.

I agree, from top view the back of the disk would not be showing that much, rather, i think it would be showing as a bit thinner disk. At a time of doing that animation i used slightly different parameters for distortion. So now it could be little more accurate, but still, my method is a rough approximation.

1080p. Glare node is soooooooooo slow…


3 Likes

Just beautiful!

This is just insanity, one of the best and more realistic attempts at Pro-level Vfx that has ever been seen in the world of Blender.

The only thing right now is the tight gradients of light on the upper left, perhaps if they can be smoothed out a bit…

Ace Dragon, thanks man but i think you exaggerating a bit. :slight_smile:
Gradients in upper left are from glare node, i already blurred it a bit, but you’re right, it needs more.

This is really amazing. I would be extremely thankful if you made a tutorial on even a small aspect of this. This is really really cool. I would love to see a massive 16k render of this on a huge wall. xD Excellent and inspiring work.

Great work! Thanks for the reference on raytracing relativistic and gravitational effects. Another reference is a 2011 Belgian paper that applies physical lenses to simulate gravitational lensing [http://vela.astro.ulg.ac.be/themes/extragal/gravlens/bibdat/engl/DE/didac.html].

16K is lil’ bit too much, huh :). It’ll take at least a week to render it and i could be making steaks on my GPU. And i don’t think it’ll have enough details for that big resolution. But i could render a 4K.

That’s a good paper. On my time of working on this project i have seen many researches on grav lensing. The method described in this paper is good for simulatin a long distance lensing, like a lensed galaxy or a quasar far behind a massive cluster (i love astronomy). But for strong grav lensing on short distances, where you need to “trace hyperbolas” this method with a single physical lens would not give accurate results.

I have some thoughts on accurate lensing simulation using refractive surfaces (again) and OSL. It would be slower but will give much more accurate results. But that’s some plans for the future.

So as a still this project is finished. I decided not to use space ship (Endurance) here, but use it in animation which i’ll be working on instead. Thanks everyone for feedback and critique. It was extremely productive.

Here is a final version.


3 Likes

So I was just on the internet like I always am and I found this http://gamejolt.com/games/action/way-to-singularity/44859/ thought you might want to know xD.

He sent me a message out of nowhere dont even know the guy and i dont know why he asked me but he was asking if he could use it for his game (I think) (I didnt reply because idk what to say to that its not my image idk where he got the idea it was unless he was asking about another image… now im confused)

Pretty sure my english is only understandable when spoken lol

TLDR:

IDK lol

“can I have this image?”

…no reply

“KK I JUST TAKE THANKS”

ergh.

Sorry for my mistake, Lord Odin. I just tried to find out author of picture and google give me your name. So, BlackRainbow, now I ask you. :slight_smile: Сan I use your work in my game?

a dude you already did ?