Jump to content

Animated Spot starring Mantra


Kardonn

Recommended Posts

Yeah, the characters and models are made in Maya or any package of choice really, and rigged + animated in Maya.

 

Then Houdini is the backbone for combining it all together. Creating all the fur/hair, creating the vegetation layouts, shading, lighting. Basically using Houdini for everything it really can be.

Link to comment
Share on other sites

Great look/feel.

I would be interested to hear about your lighting setup.

What is your preferred method of scattering with SSS? Single/Multi? Do you have an attenuation map etc? This past week i have been doing a lot of tweaking in that area and interested to know others experiences with it.

Are you going with geo fur or actual fur? (I cant really tell in the shot),

Looking forward to seeing more.

Link to comment
Share on other sites

Loving the look. Nice lighting/exposure levels.

 

Nice bright and playful. I really like it.

 

Eyes: In every character pipeline I've seen the eyes are excluded from most if not all lights direct reflection in the scene and then have their own parented  light to place the spec where you want it. Use area lights these days. Options for grid, disk. Have an image mask to add custom shaping of the light reflection.

 

Eyes generally lie outside of the normal lighting pipeline just to get that control. Eye light rig HDA parented to the head is quite normal to see. It's a bit tricky to get the parenting right. You want to inherit P but R, you may just want rx and rz. Have to try the various options.

Link to comment
Share on other sites

@abfx

 

Lighting set up is pretty simple actually since my main goal (for outdoor stuff anyway) is just physical accuracy. Envlight with HDR for the sky, Hlight (sun mode with 0.5 shadow angle) for the sun.

 

On the ground in the HDR capture, there's usually a shadow from the camera's mount and operator, so then what I can do is get in there and start to measure the data in the sunlight, and in the shadows facing away from the sun to avoid 'reflection pollution' much as possible in the samples.

 

Then it's a whole process involving making a Grid and a Sphere to cast shadows onto it, making a shader for them and lining it up with the shadows on the HDR image. Once the area on the Grid in the Sphere's shadow is a near perfect color and intensity match to the photographer's shadow in the HDR, you can then begin to solve for the sun.

 

I line up an Hlight sun with the clippy and bad sun in 99.999% of HDR captures, and then dial in its intensity and tint until the sunlight portions of the Grid match the sunlit portions next to the photographer's shadow in the HDR.

 

Once that's done, I have a Sphere and a Grid with a material that perfectly matches the ground in the plate in shadowed sections, and I have a sun that then makes that material perfectly match the ground in the plate in the sunlit sections.

 

It sounds really confusing to write out like this, but it's a nice little technique that's maybe worth a quick video demo some day.

 

Short of perfectly scientifically captured HDRIs with ND3 filters and all sorts of color solves for the filter tints etc., it's the best I can do to have a near perfect sun/sky model.

 

--

 

For SSS I tend to roll with only the Multi Scatter enabled when dealing with skin. I prefer its softer look, feels more 'right' to my eyes. The peach fuzz helps a ton to soften the skin and integrate it all together too.

 

Here's a quick test I did a little while back using that Infinite head scan model: https://i.imgur.com/aTY4eLP.jpg

 

--

 

The fur is just being done with curves in the same way that the Houdini Fur OTL works. I branched off theirs into a new one, but it's the same basic idea under the hood. Ditto with the hair shaders; they all began life as the standard SESI fur shader, and then just been adding to that to get richer strand variation, as well as a 3rd reflection call that's a blend between the "diffuse" and the "reflection".

 

@old school

 

Totally agreed on the eye light rig stuff. I've got some work ahead to get that all into shape...I'm actually even thinking of doing it with envlights linked to the eyes and nothing else.

Edited by Kardonn
  • Like 3
Link to comment
Share on other sites

Thanks for the response.

As for the SSS. I only used to use the Multi Scatter too. Had lots of xray style problems with single SSS. But recently i have trying to just make it work and understand it better. Single SSS really gives models that "fleshy" feel. I haven't posted any examples of it yet, but i will be sure to do so.

Keep up the great work.

Link to comment
Share on other sites

All done and delivered now! Lost some of the photographic feel of it I think...but that's always bound to happen somewhere in between client notes (more saturated, need eye glints, need dad in focus, etc), and the fact that all the layers had to be separated out to hand off for finishing in Photoshop. When it was all layered in Nuke and treated like a proper "VFX shot" a lot of nice blending between elements was happening thanks to various lens corruption techniques I like to use in compositing. Can't do that sadly if standalone layers need to get exported. Also in a perfect world of course I'd go back and do a better eyebrow groom instead of having them be painted on in PS.

 

Overall though very happy with the final:

 

D31OoCn.jpg

  • Like 6
Link to comment
Share on other sites

Here's a look at Peter (the dad) in a shot with his latest shaders, various fur networks, and some simple lighting.

 

The peach fuzz and the clothing fuzz seem a bit long and maybe a bit heavy handed right now...but in motion they read really well and help soften up rim lights really really nicely, so I don't think I should really mess with them too too much. Might opt for shorter and more dense though. The overall softness and tangibility of the surfaces is something that's really important to me, and these fur systems help tie it all together.

 

All lights have their shadow mask set to "* ^*fuzz*" to keep that effect as light on render times as possible.

 

He needs a mask painted for his grey hairs too, get them more along the sides of his head than the top...like me! Also I think I'd like to read the hairs better, it almost seems to turn into a volume in places, and I don't love that.

 

9hvTi7S.jpg

Edited by Kardonn
Link to comment
Share on other sites

  • 2 weeks later...

Here's a shot from the living room set. Still have a lot to sort through for photo mapping though...the Disney based materials in use on this project don't behave properly with the SESI GI_Light, which makes the bounce light wildly inaccurate and really bright compared to what it should be.

 

Right now this is a total hack where I'm deriving the "RAW GI" in Nuke by dividing my GI_Light pass by the diffuse tex inputs, applying a tonemap to curve it into place, then multiplying by the diffuse inputs again. It kinda works, but I'm not thrilled with it. So I'm rebuilding the Disney shader with faked microfacet calculations. From my tests it looks nearly identical to Disney, but it photon maps properly.

 

Lookdev/materials aren't finished yet on this asset, so there's quite a few things with just a solid color applied still.

 

ocWru5Z.jpg

Edited by Kardonn
Link to comment
Share on other sites

...the Disney based materials in use on this project don't behave properly with the SESI GI_Light, which makes the bounce light wildly inaccurate and really bright compared to what it should be.

...

Principled Shader in H15 is not very physical and doesn't preserve energy if you are using non-metallic reflection or coat so I wouldn't be surprised if that's the reason for your problems

 

but your images look really beautiful, great work

Link to comment
Share on other sites

Principled Shader in H15 is not very physical and doesn't preserve energy if you are using non-metallic reflection or coat so I wouldn't be surprised if that's the reason for your problems

 

but your images look really beautiful, great work

 

Yeah I'm actually still in H14 with this because I've got a kind of 'ubershader' wrapper I'd built up around Wolfwood's Disney BSDF work. It conserves energy properly on the edges, but lately I've been thinking of getting rid of that cause I feel like the Schlick/Smyth darkening ends up feeling too dark on a lot of surfaces.

 

Kind of thinking I like the look of Disney's diffuse, even though it can technically return more than 1 energy if your diffuse albedo is 1.0 (I think you'll get 1.08 on glancing angles in a blast furnace). Since a diffuse albedo of 1.0 is impossible in real life anyway though I think I'm cool with what they've done, it has a great look to it. Very vibrant and soft.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...