Maxwell Render | Official public forum

i searched and find examlpes made by maxwell but the result not realistic ; i know maxwell can do it but how ?
mental ray or vray have special material for skin ; i hope see good examlpes with details

The issue is that we can map the transparency of a SURFACE, but there is no way to MAP trans within a volume, SO you can make a nice sss skin like Meinbender do, but the skin must have very uniform MAPS especially the transparency, SO very few details in the maps is the key - very boring / very uniform / very blurred maps are needed.

Excuse the pervy model, it had the most uniform skin tone maps, so I used this model for a test;

Still quite slow, 40 mins render;

http://vimeo.com/22342702

eric nixon wrote:The issue is that we can map the transparency of a SURFACE, but there is no way to MAP trans within a volume, SO you can make a nice sss skin like Meinbender do, but the skin must have very uniform MAPS especially the transparency, SO very few details in the maps is the key - very boring / very uniform / very blurred maps are needed.

Excuse the pervy model, it had the most uniform skin tone maps, so I used this model for a test;

Still quite slow, 40 mins render;

thank you ; the maps i used is from daz studio and only two maps ; diffuse and bump ; as you say the maps is very important ; actually i will use only the two maps ; i tried alot but all the result with maxwell not good
the skin looks good and the render time also interesting ; i think in my laptop take long time
i create simple material with two( BSDF: the skin and second BSdf : wet skin); it have not good settings but give me the result what i need

thank you

Hervé wrote:http://vimeo.com/22342702

i see the video in cartoon network channel ; very beautiful

You have to use a lot of different maps to control scattering, speculars, colour and more with Maxwell but the trick comes to weightening them properly to eachother to get a good results which renders at acceptable speeds. It's not easy!

I made some tests recently and they are okay, more work can be done to improve it even further but i got this far..
http://maxwellrender.com/forum/viewtopi ... 48&t=41679

/ Magnus

Maximus, your tests illustrate the problem of trying to map the trans of a volume very well.

more work can be done to improve it even further

No this is not true..

Maximus3D wrote:You have to use a lot of different maps to control scattering, speculars, colour and more with Maxwell but the trick comes to weightening them properly to eachother to get a good results which renders at acceptable speeds. It's not easy!

I made some tests recently and they are okay, more work can be done to improve it even further but i got this far..
http://maxwellrender.com/forum/viewtopi ... 48&t=41679

/ Magnus

thank you
i see your work its good result ; i use perhaps the same maps as you .
diffuse map ; diffuse map with hue ( to red color)
density map ( one for scatterind and one for thikness)
i will try more

the skins are not the same ; for that we need create base settings
and the skin human have a verry little hair ; that give the skin special effect in head for example ; blood ...
i also tried with blood bsdf + skin bsdf

No this is not true..

Seghier, and for anyone else reading. When you use a weightmap or even a diffuse map you are therefore making a change to the transparency, because the values are connected. Maxwell will try its best and you can get somewhere with very subtle blurred maps.

Try thinking more about light passing through substance-volumes, VS the mapping of polygon surfaces, - this is light-smulation and material-simulation, its not like V-ray style renderers... think less about finding a magic hack.

I hope you understand.

Eric, yes there are problems with this but at the moment i cannot see how to work my way around it. Thanks!

seghier, thanks. You should use a similar approach when producing your skinmaterial. Paint all the maps, create a atleast dual layer bsdf material within a single layer Maxwell material. Start out simple in a neutral lighting scene at the correct scale, create the basic colourschemes first for both layers, add in the basic scattering, add each map at a time, testrender until you turn blue, add maps and adjust all the time until you reach satisfactory results, do about a million testrenders again in different lighting conditions to see if your material works as expected. If not, start over. If it works, go live with it.

/ Magnus

eric nixon wrote:

No this is not true..

Seghier, and for anyone else reading. When you use a weightmap or even a diffuse map you are therefore making a change to the transparency, because the values are connected. Maxwell will try its best and you can get somewhere with very subtle blurred maps.

Try thinking more about light passing through substance-volumes, VS the mapping of polygon surfaces, - this is light-smulation and material-simulation, its not like V-ray style renderers... think less about finding a magic hack.

I hope you understand.

thank you ; i will try . in maxwell fire don't give exact result and rendering take long time ; i made some tests but i don't finished the work
vray or mental ray and some rende engines we can easily setup the skin material ; there exist alot of tutorials and tips and special skin shaders
i test a free gpu engine can render human body and skin quickly with excellent result
----

Maximus3D wrote:Eric, yes there are problems with this but at the moment i cannot see how to work my way around it. Thanks!

seghier, thanks. You should use a similar approach when producing your skinmaterial. Paint all the maps, create a atleast dual layer bsdf material within a single layer Maxwell material. Start out simple in a neutral lighting scene at the correct scale, create the basic colourschemes first for both layers, add in the basic scattering, add each map at a time, testrender until you turn blue, add maps and adjust all the time until you reach satisfactory results, do about a million testrenders again in different lighting conditions to see if your material works as expected. If not, start over. If it works, go live with it.

/ Magnus

thank you for the tips
i will try to get the best result ; my problem with waiting long time to render the skin
metal or glass for examlpe have settings and we don't test to know if it is good or not ; maxwell right now don't have special skin material and we need test to get the right material !

You can still use Fire for your preview to get a rough approximation of your materials properties. Then do a quick finalrendering of it while you continue to tune your material some more. That's what i did to save time and i have only one slow i5 PC with 8gb ram.

Start off with the simplified wizard interface in 3.0, create a red wax SSS material, start tuning that into the skintype you want as the base characteristics are about the same.

/ Magnus

eric nixon wrote:

sometimes i imagine that my laptop will explosed like that