Step aside, Joan! There’s a new lady in town, and if you’re a fan of the Warcraft series, you’ll really dig this contribution.
Max Kor, the artist behind the famous
Night Elf render, has benevolently posted a
tutorial on CGSociety.org. He didn’t (of course) use Blender, and he used a SSS shader, but us Blender artists have been faking SSS for years, and from what I can tell (although I haven’t had a chance to do the tutorial myself yet) it should be duplicable in Blender.
http://features.cgsociety.org/story_custom.php?story_id=4006
If anyone gives this a go, post a link to your work back here so that we can all take a look.
Step aside, Joan! There’s a new lady in town, and if you’re a fan of the Warcraft series, you’ll really dig this contribution.
Holy hell. Thats an amazing piece of artwork. The dame’s sweet as heaven. :)
Hmm. Impossible to animate. Might as well used a pensil and brush.
I know this tutorial and I think its not possible to get such quality with fakeSSS!
Max Kor is also using a MentalRay shader (misss_fast_skin).
In other renderings I saw the use of scattering with 3 (!) different scatter layers!
I know no alike renderings or techniques in Blender. The main reason is, its impossible!
At this point you are comparing an amateur-system like Blender with an professional cinematic system (Maya / Mental Ray). Blender has very much useful features but the implementation and handling is at most like Maya was in 2000.
“””At this point you are comparing an amateur-system like Blender with an professional cinematic system (Maya / Mental Ray). Blender has very much useful features but the implementation and handling is at most like Maya was in 2000.”””
In that case, it’s not an amateur-system but an obsolete professional cinematic system… hu hu hu
But I don’t really agree with that statement.
About the modelling, people are really close to doing the same thing with Blender (visit extinctionlevelevent.com tadaaa!)
http://blenderartists.org/forum/showpost.php?p=845919&postcount=137
All that is missing now is a decent SSS, but as it was mentionned, it can be faked. True, it’s not the real stuff yet. But most of the SSS technique doesn’t seems to be a concern for Blenders devs… it’s better to seek help with yafray or indigo (which has a new test version with SSS I heard…)
But we are eating the distance, there is no arguiing with that.
i wouldn’t compare that with the joan of arc tutorial. joan is a 60 page monster tutorial, this night elf is good but i really wouldn’t call it a “tutorial”. Besides seeing that is probably going to cause a flood of more “how do i fake sss in blender” and “why doesn’t blender have sss” threads at blender artists.
the elf did pretty good.
Joeri: “Might as well used a pensil and brush”
Like telling the classic artists they might as well have used a camera. Every artist chooses their own medium to work in. How can you post a comment like that on Blendernation!?
@ the author: can’t understand why you wrote …of course… Did you want to say, that blender isn’t able to produce stunning images…? Or do you mean, because it’s done with mental ray?
@ all: remember: there is blental, the blender to mental ray plug-in. with this you can use the mental-ray-shader, like fast-sss. and look to the roadmap: the external-renderer-connetction is on its way.
btw: yes, great picture.
S.
like i said…
sss with blender-to-mental-ray:
http://s10.photobucket.com/albums/a116/RCRuiz/MentalRay/Test/
S.
Nice render, and I seemed to have missed the mentalray is free announcement somewhere!
Not everyone is tied into a studio that can utilise these mentalray exporters ;)
Saying that I’ve not played with Blental yet either!
how can somebody be so good and such terrible mistakes with the ears and eye brows?
Man those two elements not only stick out terrible but are also out of proportion.
The rest of the head is just amazing.
@daas
I don’t think that they are mistakes at all.. Max Kor got the idea from War Craft and those night elf characters have such ears and eye brows.. Here’s a link for a male & female War Craft night elf game character picture: http://upload.wikimedia.org/wikipedia/en/e/ef/Warcraft_NightElfs.jpg
Well, I’m glad that the author didn’t spend to much time explaining the modeling process because the head model is fairly standard. I’m glad he explained the texturing in great detail.
About fake SSS. If I understood it correctly, the defocus node can accept any grayscale to blur it. That means that one can paint an SSS map texture layer for the defocus node. For example: the bony parts like the forehead exhibits little scattering, so you could give it a value accordingly, and the fleshy parts like the cheeks should blur more.
That in any case should take care of the blurring effect on the texture that SSS causes. For translucent and forward scattering stuff, I think we should the mappping of the shadow pass and some manipulation with the vector (or was it normal?….) node. I must say that the Fast SSS technique from Matt come very close to the forward scattering effect. I believe that in Blender, believable SSS should be possible.
It would help a lot though if the Z-buffer rendering could be more flexible, in that the Z-buffer (AND invert) are rendered from the lamps point of view. And it would help a lot too if they could be mapped. That would really nail that back scattering effect!
Great! (The Face)
About Indigo:
Yes, there is SSS, now but it’s not that “new” anymore.
The new thing is, that you can use Bidirectional rendering together with SSS, now.
But: Yet, you can’t give colour-textures on SSS!!!
If anyone can “model” the violet stuff, you could do it in indigo, but I guess, noone is able to, though…
It would need to be ~ 0.1 – 0.01 mm thick (which is 0.0001 – 0.00001 BU, as one BU is translated to one Indigo-meter…)
1. amazing work
2. SSS in blender. real one. now.
3. He lives really close to my place (10 minutes ride) – should I go over and tell him to drop maya in favor of blender?
haha, okay, are we seriously comparing Maya to Blender?! I have used Maya 6.5 unlimited, and let me tell you it runs circles around blender, and the internal renderer compared to Mental Ray, please don’t make me laugh. Both are run by multi billion dollar corporations, there is no comparison, otherwise all these design studios would be using blender.
so what? just use mental ray in blender and enjoy the exact same capabilities. If you want to compare blender to maya then be sure to use maya software and not mental ray because MENTAL RAY is NOT MAYA
Framedworld:
Your comments make it sound like blender is some joke. It really is not the tool in most cases but the artist. Can you show us some of your renders from your Maya product that runs circles around blender?
Thanks
@Framedworld: and how exaclty does it runs circles around blender? Its true that blender has still a long way to walk but it marches but at least i cant tell if a Mesh was created with blender, 3dmax, maya or XSI.
People tend to forget it’s not the tool that make the artist.
I’ve seen people with the complete maya doing utter crapadoodle things while other people can do nice stuff with blender.
Some artists think they can achieve better performance with a Uniball Liquid Gel Finepoint Tech 2000 while real artists can achieve a lot of things with a biro pen…
As for the renderer, compare what can be compared. Both maya and blender internals produce similar stuff IMHO. If you speak of Pixar style rendering, I might remind you that Pixar doesn’t use Maya internal. They use Renderman.
Often, when I speak with professional graphists, they start running out of argument when I ask “why pay a lot for Maya when Blender can do what you need for less?”
The most common argument I can’t counter :
– they fear the interface and having to learn another software
– they think blender is used nowhere in the professional industry
:D maybe someone can find a solution to those…
@ZanQdo: There are shaders with SSS-capabilities in the Maya-Renderer like DT3D or bssrdf (was used in LOTR!).
These are in some scenes better than those of MR!
@aws357: Its not the matter that artists would not change a product because of the GUI! I know a lot using Maya, Max, XSI, Mudbox at the same time. The bad things about Blender are:
– the outstanding bad GUI
– for people who are working for business -> no support
If the Blender-devs are working further in the actual way with the aim to get a tool with “the most” features, they will forget the very important things! And so its a nice tool for getting into the cg-world.
@all: its very interesting for me to start here the discussion “Maya or Blender”! Nowhere else I got so much response!!! I think its a matter of the artists to do the best out of the used product.
What is SSS?
the “move over Joan” was an over statement. Actually, I can’t get much out of THIS tutorial since my modelling skills are still very lacking.
On other news, I think is naive to pretend Blender comparable to Maya, but it is even more naive to pretend Maya will make you a better modeler. I mean, I have access to full versions of Maya at school(Architecture school), 24 hours a day, plus the PLE I keep at home, and it doesn’t make me an awesome CG artist. Besides, I don’t trust AutoDesk, they already screwed up AliasStudio(now that’s a solid modeler) a little bit by taking away its DXF importing capabilities on the last version. NOW THAT THEY OWN BOTH STUDIO AND AUTOCAD, THEY CHARGE YOU EXTRA FOR THE INTEROPERABILITY! Anyway, just a little rant there.
My point is, nowadays with the quality of freely available software, know-how is the key. So I can’t make photo realistic images in Blender? It doesn’t mean I can’t make quality work. It doesn’t need to stand out as the “best CGI on the planet” to be good. There are such things as neat designs, good stories and lovable characters that are software independent.
@bg: bad inteface? Dude, I dig this like 3 times more than I dig Maya’s interface.
Only thing I ask from blender now is decent NURBS support. I don’t think people realize how big that would be in the architecture/design industry.
@DifferentSmoke:
You can’t make photorealistic renders with the blender internal renderer.
But as far as I know, you also can’t with the Maya internal ;)
YafRay is quite and Indigo is very photorealistic, but Indigo isn’t complete, yet:
Animation nearly isn’t possible. (It is, but it wont have motion blur nor will look good, except you’re using MANY MANY processors OR MANY MANY days to render it ;))
(I know, this one was meant for still pics… But I already wrote about that some posts ago)
@ all
“bad GUI” -> Everyone sees that different.
Ton had a lot different ideas to other 3DApps. And I like those ideas ;)
You need to know many HotKeys, but if you know them, it’s quite easy to make nice stuff with it.
The thing with NURBS and so on: I don’t really know, if they are that bad, but nearly everyone says that, so I guess, yes.
I lack in using them.
@HumoDiferente,
Correct, “move over Joan” was an overstatement used for dramatic effect. Obviously the two tutorials teach very different skills in two very different ways. It’s comparing apples to oranges.
It did catch your eye, though, didn’t it? Therefore, it served it’s purpose.
By the way, in case you didn’t already know, I am tu amigo spiderworm from IRC :) Hola!
David
@ oldcarrot, someone will probably have a better explanation but sss is SubSurface Scattering, which means rougly the light bounces and scatters around in the skin. Skin isn’t a given solid, it’s actually translucent, and light passes through it. hold your hand in front of a light and it will be lit red, around the edges at least. Use a light, not a coin dude :P good to see you here.
@David Millet regarding the joan thingy. Overstatement was an understatement. maybe you overdid it a little.
@ framedworld. May or may not be true, I haven’t used maya, I’ve used max and cinema before though. I don’t know where maya stands next to max, though I have a fair guess. I like blender more. I work for a company where I use blender for visuals a lot. Renders I think do nicely, really impress my bosses and the customers here, I don’t need a $k renderer for that. Tell me you can afford a few k for an app you don’t know will get you a job. I stuck with blender next to a fulltime job and i DID end up getting a kick ass job with it.
@ bg, I used various programs during school, internships, detachment jobs etc, never got any decent official support for those. A number of times I got a “you can’t do that because we locked it” reply.
With the blendercommunity I ussually find what i need within the timespan i need it.
About your last phrase: “I think its a matter of the artists to do the best out of the used product.”
Second that! Creativity is the key.
sorry wrong carrot. I need to eat more of those…
/offtopic
OldCarrot is nothing like Carrot_nl freaky dude ;p
lest we forget a great tutorial on this photo-realistic technique using Blender: http://babelfish.altavista.com/babelfish/trurl_pagecontent?lp=fr_en&url=http%3A%2F%2Fterrier.infographie.free.fr%2Fwork%2Ftuto_portrait%2Findex.htm
(english)
I think this just reinforces how Blender can be used to achieve this, by showing us how to, in addition to the layers in the tutorial, add on yet another layer for color (blood and fat) and specularity (by now, everyone should be using colorramps for specularity, yes?) to achieve photo-realism using Blender.
that image is great and quite realistic, but photorealistic?
I can’t see any caustics of the glasses…
Maybe, the HDRi isn’t good for that though….
It is very interesting what this discussion grow into. IMHO (and I’ve seen, if not actually heavily used allmost all major 3D apps) one can do allmost everything in any app. If you have in mind that early pixar movies look visually stunning even today (no reflestions, no sss, no raytracing) clerly shows that it is the artist and not the app that makes the difference. I am sure that this image would have looked allmost equally great without sss if the artist used all other available light tricks…
just my two cents…
on the comment made by aws357 early on..
hu hu hu..
that was funny.
@bg
– the outstanding bad GUI
Really, I have being designing with every single material or application 15 years now. And I can assure you a really really miss Blender interface ideas for vector editing with Inkscape. What I am discovering now is that the node editor can really fill The Gimp capabilities with Blender usability paradigm.
Just because you don’t like it, it doesn’t mean it won’t work with other people :D
I am an amateur sketcher, and only work with pen and paper. I have never used any digital program to edit or make pictures. I have no idea how hard it is, or what kind of tools you would have to use or anything. With that said:
I admire the person who did that Night Elf very much, because it is amazing. The texture of the skin and hair, and also the contrast between the skin and the metallic bits are just great. The eyes are incredibly realistic.
Still, I think there’s something about the ears that makes them look plastic. And the same thing with the leaves. The leaves are too ..thick, to look realistic. I also think that the “Night Elf eyebrows” look like they have been glued on to her real ones. It doesn’t look right at the point were they start to “lift of” from the face. Perhaps it is because the colour of the eyebrow changes quite a lot. It goes from very dark to light blue in an instant. Maybe it should be toned out instead.
Now, as I said. I have no experience in digital programs, and have no idea if what I said, even makes sense. I was just passing by the site, by accident, really, and said what I thought. Maybe it makes sense, and maybe it doesn’t, and if it doesn’t, I apologize.
Thank you.