effective combining of SSS with Mia_mat ?

[epq]

Did you also hook up the SSS lightmap shader to the mia shading group ?
/eric

[misterJoj]

and turn off the diffuse weight of your Mia

[Libor]

thx for your tips boys! Ill try it asap and post some pics...

L.

[Libor]

So heres my first serious raspberry shader test...

still a lot of room to improve realism (modelling, hairs, surface imperfections etc)

test shots

http://www.cgart.org/data/tmp/berry1.jpg

http://www.cgart.org/data/tmp/berry2.jpg

and the shader network for those interested in...

http://www.cgart.org/data/tmp/berry_shader.jpg

[royterr]

i heard there was 3 methods.

-one of them is to plug the mia into the spec shader of the sss.
-one of them is to plug the sss into the aditional color slot of the mia.
-here's Masterzap's version:
http://mentalraytips.blogspot.com/2008/04/beauty-isnt-only-skin-deep-combining.html

That's an excellent thread, i would really like to know whats the right way pf properly combining SSS and MIA

[InfernalDarkness]

I think they look really great. Not that it needs any improvement, but had you considered using the Translucency slot in the mia_x? We've been using it a lot on leaves and such lately, and it really does help out there. Not sure if it would add any realism to your great berries, though... Keep at it! Perhaps post us a render or two with the hairs? I'd love to see how that turns out.

[Galakgorr]

looks nice.

i can't tell from that render but it doesn't look like the white egg-like object inside the raspberry is blocking any of the scattered light... i feel like i should be able to see a dark spot inside the berry where the egg would be. if you add the lightmap shader to the egg (regardless of whether or not the egg is actually receiving any scattering) it should block back scattering through the inside of the berry.

[Libor]

Thanks for your thoughts on this topic guys which is really interesting to me too :)

royter> hmm, sounds interesting I ll dive more into these, thx! ...been whatching ZAPs technique which is pretty much like mine >

" Use a mia_material_x
- Put your current SSS surface shader int' it's "additional_color" input
- turn off the SSS shaders speculars, reflections and unscattered diffuse" *

* misss_fast has option "scattered contrib only" - should be checked on so the shader doesnt contribute anything than sss (mia take care of the rest)


Infernal..> yep I ve realized that when I was playing with different level of tranasparency which greatly affetcs the output but not tried it yet....I should definetely give it a try, Im not even sure if the raspberry is a case of translucency or not in real world ...should examine this definetely!

Galakgorr> me too! Im not certain the ball (which purpose was to examine the scattering) is occluding raspeberry or not! Do you think its neccessary to plug lightmap shader to its SG node too?? anybody knows?

Again thanks for your interest and inputs!

L.

[royterr]

Thanks for your thoughts on this topic guys which is really interesting to me too :)

royter> hmm, sounds interesting I ll dive more into these, thx! ...been whatching ZAPs technique which is pretty much like mine >

" Use a mia_material_x
- Put your current SSS surface shader int' it's "additional_color" input
- turn off the SSS shaders speculars, reflections and unscattered diffuse" *

* misss_fast has option "scattered contrib only" - should be checked on so the shader doesnt contribute anything than sss (mia take care of the rest)


L.

What about
diffuse weight,front sss weight, back sss weight, can i set them each one of them to 1 and stay phisicaly accurate? or should they all make 1 when added together?

[Galakgorr]

Galakgorr> me too! Im not certain the ball (which purpose was to examine the scattering) is occluding raspeberry or not! Do you think its neccessary to plug lightmap shader to its SG node too?? anybody knows?



i'm almost certain that if you want the ball to affect the scattering, it needs to be part of that same lightmap. try connecting the lightmap to the ball's shading group and see what happens.

[JSlotosch]

That´s exactly the same thing I also try to figure out, what combination and settings are the best between mia_miterial and sss. If you do a render with the skinplus in Max, its the same like combining both mia_material and fast_skin_plus´in Maya, but the result looks some way different.

I´m also still not satisfite how it looks in Maya, compared with the result in Max.
Do you guys know where I can find the slot "local enviroment" in Mayas mia_material or misss_fast_skin? In Max with it´s modified misss_fast_skin shader skinpuls, this slot is under "Reflections (from A&D material)", but I can´t find it in Maya.

What does this mean Libor, "turn off the unscattered diffuse", what is the benefit of this? I always turn the color of the diffuse from the mia_material to black and leave the Overall color and the diffuse color of the skin_shader as they are. Most I use the overal so I leave the color map there and put the diffuse color to white with the strength 0.3 or 0.5.

I think your raspberry looks great so far. I´m not sure if I understand it, but u try to make it transparent? I think it shouldn´t be so transparent, have a look at this pictures: click me (http://images.google.de/images?hl=de&source=hp&q=himbeere&um=1&ie=UTF-8&sa=N&tab=wi) There you can see that u can´t really look through them. I think you only have to add some hairs and maybe something like a grainy fuzzy color map on the top.

Regards,
Josef

[royterr]

did a lot of tests. The best way i found so far war is to:

-connect a miss_fast_shader (with scattered contribution only checked ON) to the aditional color of your mia_material (with diffuse set to black)
- apply the mia material to your object
*i did not found any difference when i connected the miss_fast_shader's lightmap to the Shading Group of my mia_material.I dont see why you are doing this

there is another method :
-connect the mia material (with diffuse set to black) to the specular shader slot of the miss_fast_shader (with scattered contribution only checked OFF ).
-apply your miss_fast_shader shader to your object

-the 2 method give the same result

[Redsand1080]

*i did not found any difference when i connected the miss_fast_shader's lightmap to the Shading Group of my mia_material.I dont see why you are doing this





In 2010, using the method you have outlined here, I have to connect the lightmap from miss_fast to SG of mia_material_x or I get a totally different (and what appears to be incorrect) result. Anyone else noticed the same thing?

Great thread by the way. Glad I found it. Gotta love google search. :)

[Redsand1080]

Does anyone know if the built in AO of the mia_material_x is supposed to work if you use the SSS shader plugged into 'additional color?' I haven't been able to get the AO to work so far, in any of its modes, when I have the SSS plugged into 'additonal color.' I have the diffuse component turned completely off on the mia_material_x since I am using the SSS for my diffuse shading. Could this be the cause of the missing AO? Is the AO directly linked to the diffuse settings of mia_material_x?

Thanks for any assitance! :)

-Justin

[Vectorhead]

Hello,

Can I ask what value from the sss are you plugging into the additional colour via connection editor?

ex: message, output_sss_only?

Many Thanks!

[Vectorhead]

I forgot to say this is for the fast shader x

[AndreHeiter]

Yeah i have pretty much the same question:

As I drag my SSS Shader on to the additional color slot, i have to specify what exactly i want to connect. I tried the Diffuse result from the SSS to the addtional Color from the mia... Was that the correct way? If not, what should i connect ;) Didn't find that problem on the net. Everywhere I looked they just drag it on to additional color... and they are ready to go..

[Redsand1080]

It should be called just 'result' only. Not diffuse result or spec result...just result. Hope that helps.


-Justin

[AndreHeiter]

Hey Justin THX!

[Redsand1080]

Not a problem man. There will probably be two of them. One might be result.a or something like that. Ignore that one. It's the other. Did you get it working?

[EdtheHobbit]

Hey Redsand, did you ever get AO working with the SSS/Additional method?

Hate to raise up a nearly-year-old thread, but I'm having the same problem.

[Redsand1080]

I believe I did. It's been a while since I've used that workflow, but if I remember correctly once I used the diffuse from the mia_material instead of diffuse from the sss, the AO worked just fine. So basically set diffuse to 0 in the sss shader, and use the same value that was previously in the diffuse sss slot for the mia_material. Then tweak it if you want a different look. Can't say 100% positive this worked since it's been a while, but it's worth a shot at least. :)

-Justin

[EdtheHobbit]

thanks for the input -- I'm using the same workflow you describe, but good to know anyways.

I figured out what my problem was -- apparently the "AO with color bleed" mode was the culprit. For whatever reason, it was estimating that my "additional color" contribution was about twenty times as bright as it should be, and creating a glow instead of occluded shadow. Seems to be fixed if I don't use color-bleed on that object.

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.