Modeling: ZBrush, 3dsMax, xNORMAL 2
Render: Marmoset toolbag
Post Processing: Photoshop
Translation : Michele Marcelli (Michele71)
MAKING OF CONAN
"Giorgio Palombi (aka Fracture) leads us to the discovery of his personal Conan the Barbarian, where with wise management tools and programs, takes us on a journey of shapes, colors and lights. Absolutely to follow and appreciate. Good reading!"
Hello to all
Welcome to the making of Conan, my entry to the Comicon Challenge 2014.
I'll try to explain the phases of the process creation trying to focus what I found most interesting and useful to the community. Thanks to Treddì for the hospitality and to you for your attention!
I finished the project in about two months, working between the leisure, and my job always trying to cut corners to save time... The software that I have used are:
3ds Max, ZBrush, Photoshop, and Marmoset toolbag xNORMAL 2.
According to the regulations, I chose a cartoon character portrait reacts while injured after a bloody fight, and my choice fell on Conan.
In particular, I'm a fan of the books of John Buscema and Alfredo Alcala, such as clothing and overall look I chose the more stringent section of Claudio Castellini and then I tried to mix thoroughly ...
I did some sketches in order to take possession of the character better and to look for a first pose.
Totally abandons the stage I think is important because often the right idea mature observing the first sketches after some time, jump out strengths, weaknesses and new ideas. In my case for example, I decided to change position only after the legs to show some more details.
I started in Zbrush trying to turn a human body into a standard barbarian and exaggerated comic strip, stretching his limbs and exaggerating proportions. Then I switched to sculpture in ZBrush and I took as reference the comics to the more mass using Move, inflat and for the plans HPolish. For details surface I used a series of alpha (skin pores) and a combination of Slash and Dam standard brush to define wrinkles and folds. I greatly exaggerated the details because I prefer the normal map modular decided later.
Weapons and objects were "aged" and worn using the slash brush and some alpha (dirt and scrath).
The thing that I think is important at this stage is to try to make some strokes big enough to be seen in the middle distance and depending on the size of the mesh. That give more realism I tried to do as if I draw cuts in china, with sections of varying size.
For the hair, eyebrows and undercoat I initially created a basic "solid" in zbrush, just to control the mass and the more general flow, then I used fiber mesh in ZBrush. I tried to find a middle way not to go too much on polycount, then I positioned the automatically generated UV and I adjusted the texture with alpha channel.
For the body I used the classic retopolgy of Zbrush, which I think always the most reliable and fast, flanked by 3DsMax to the finish. For everything else I decided to do a bit of freestyle since we had not any limit of polygons.
In particular, I tried to use a plug-in for ZBrush very useful decimation master.
This tool allows to drastically reduce the polycount of a mesh using a triangulation that you can modular getting different polycount. Its feature is the speed and accuracy in maintaining the masses, and the details in the case of external render. Although the result is not good for a game engine, it is still a loophole useful to populate the scene for a render static. Weapons, arrows and base are molded in Max
For the mapping I used 3dsMax, but sometimes I also made use of the plug in zbrush "UV Master," as for studs and chains.
To use UV Master to better prepare the seams in 3DsMax, do the unwrap in ZBrush, I go back up to better organize the space. In the end, I used a 4000px map for the skin and 3 other textures 2000px for clothes, weapons and props.
I usually use the "render to texture" of Max, but this time I decided to try xNORMAL.
It has proved a very useful tool because it manages memory better in hi poly models (millions of polygons) and avoids embarrassing situations in Max From here I have always exported the spread (bake vertex color, ie polypaint), cavity, AO, normal.
For the AO I had some problems initially, after some testing I realized that to work just delete the vertex color from the model hipoly.
For the skin I used the tool "spotlight" in ZBrush, which I projected directly a photographic base. Then I cleaned the surface photographic detail preserving tonal values, getting my base skin.
All the details jump out of the sculpture below, to get them I created a cavity map + AO by xNORMAL and I then composited in Photoshop. For objects I used directly photoshop after exporting maps.
Once ready the low poly (in this case perhaps add mid poly) I created a biped and skin it in max to create a series of poses. In the end I chose the one that best suggests aggressiveness and dynamism, to the detriment of drama wound .. Conan in a knee or partially sick I was not really convinced.
The fabric below the belt and hair were positioned after installation (just to avoid jerky sessions at max).
MATERIAL, LIGHTING and RENDERING
For some time I started to fiddle around with Marmoset 2 , which in addition to being an excellent real- time engine has further developed its computing power and the quality of its rendering. For details , I refer you to the official page , I 'll try 'to summarize my workflow . Initially I used neutral enough illumination to change my shaders , at this stage I went back and forth between marmoset and photoshop editing diffuse , normal, specular and gloss maps.
For each material, it is possible to modify the " diffusion" in " Skin" , this is a great way not only to have the material of the skin, but also to add translucency and " fuzz" materials such as fabrics , hair and fuzz .
For the lighting, I initially chose a sky between presets (HDR) then I lowered the Global Brightness and I started to modify light by sampling the values directly on the map ( very intuitive ) , and then I added some spotlight in a natural way to lighten some parts too in the shade or natural reflexes to give metal and eyes.
Once you have a good render with alpha I simply created a background in photoshop and added the effect of paper consumed typical of a cover 70.
I hope this overview is helpful, if you have specific questions you are welcome :-)
Thank you all!
For more info about the author of Making Of see: http://www.giorgiopalombi.com/