@vejn: For Gorgon I combined a Dielectric Material and a Glass Material with a mix node, factor input = Custom Fresnel node I built based off CynicatPro’s tutorials here: https://www.youtube.com/watch?v=xMpDTk3DTW8 . I used this method to create the Material’s surface. For the volume I combined a volume absorption node and scatter node through an add shader.
In terms of realism, the main issue I think you’re having is your lighting. Try a different camera angle and setup DOF (create and empty, place it on the front surface of your object facing the camera, set camera target to the Empty, and play with the F-STOP/Radius values until you find something nice), and then use an HDRI environment texture as the input to a background shader in the world material. From there, just start throwing lights around everywhere. LOTS of lights! 10, maybe 15. Solo the lights so you can see what they are doing individually and tweak them to your hearts desire. Physically Based Rendering and in general, photorealism, is entirely dependent upon accurate reflections and ray scattering, which is hard to simulate if there is nothing for the materials to reflect.
You want your render to look good without any compositing first. Once it looks right, head over to the Scene Properties in the Properties window (next to the World Settings), and turn on color management. Gleb Alexandrov has a great tutorial on Blender’s Color Management settings, here: https://www.youtube.com/watch?v=aByZrsE1J0w . First, turn on film emulation (it expands the range of the blacks, so true blacks are actually black, and emulates the look of real camera film). Then, start tweaking the gamma and exposure settings. You want large dynamic range between dark colors and bright lights. From there, play around with the “Use curves” setting to push contrast, and push/pull the RGB values in your image.
- PBR Materials and Textures
- Camera Angle (set up DOF)
- LET THERE BE LIGHT
- Turn on Render Passes for Z, AO, ENV, MIST, and DIRECT/INDIRECT for all applicable ray types in your scene (transmission, glossy, diffuse, subsurface etc)
- Film Emulation and Color Management
And finally, step 6, Compositor!
Plug the rendered image directly into a Color Balance node. This will allow you to apply final color tweaks, adjust shadows, highlights and midtones, etc. Take the output from here and plug it into a Glare node. Find an effect you like and tweak. Plug this output into Lens Distortion (chromatic abberation is awesome if it’s subtle). If you want more glare, darker ambient occlusion, want to tweak the color of the reflected light, or add volumetrics/mist, use the render passes on the Render Layer and combine them using a MixRGB node with your main source image.
And that’s it. I basically go through that same process when creating any image. It takes a while to optimize the workflow, I’ve spent years learning from different Blender Community leaders like Andrew Price and Gleb. This is sort of an amalgamation of all of their different techniques tailored for myself.