Maya Tutorials: Game Props Modeling
Maya render farm
Fox Renderfarm, a leading cloud rendering service provider in the CG industry, will share with you the game props model produced by Maya. In this Maya tutorial, we will introduce how to create a game-style 3D model based on conceptual design drawings, from ZBrush carving to high poly and low poly UV texture production.
The following is a brief description of the process of making 3D models of games. I hope that you can have some understanding of the models used in games.
Before making a model, there are usually some conceptual drawings as reference for making. In this case, you can directly put the reference drawing in the production software Maya as a reference.
Before importing into ZBrush for sculpting, you can first do some preparation work in Maya. If the model is relatively simple, such as stone, tree trunk, floor tiles, etc., they are relatively simple. You can use ZBrush to sculpt High Poly directly without making a Low Poly model. If the model is more complex, such as a biological model (animal or the like), you can make a Low Poly export OBJ file in Maya and sculpt it in ZBrush.
Import the low poly model into ZBrush for sculpting. From Blocking, make sure that the scale is appropriate. The most important thing is that the model does not deviate from the conceptual diagram. It is necessary to determine the overall ratio first, and then it is the part of some accessories, and finally the detail part. This process only uses the basic brush.
The number of model faces sculpted by ZBrush is very high, which may be as high as hundreds of thousands of faces, or even millions. Such models cannot be used in-game engines. Therefore, the high poly model sculpted by ZBrush needs to be topologically a low poly before it can be used in the game engine. The topology process uses a topology software TopoGun, TopoGun can help the model Resurfacing and Maps baking, and the details of the model will not be lost. If you are making a single static map, then the model does not need to rig bones, it is not necessary to make low-poly and Normal map.
After the topology is completed, it is the job of the UV. Normal maps can be baked after the UV is split. The UV can be organized using some plug-ins included in Maya, or can be produced using other software. You just need to expand them and adjust as needed. Try to be as reasonable as possible.
Schematic diagram of UV deployment and placement
Before bringing the model to Substance Painter, you need to bake the details of high poly onto the developed UV to make the texture. Baking and texture can be made using Substance Painter or Toolbag, especially Substance Painter has many tools and functions, which is very suitable for realistic and cartoon-like texture production.
These textures represent different information such as Normal map, Specula, Hight, and Gloss. The higher the accuracy of the texture, the better the final display effect of the model.
Import the baked texture into the most commonly used Substance Painter software to create PBR textures. Some parts can be adjusted using some preset textures, and then output different textures according to different needs.
3ds Max Tutorials: Semi-realistic 3D Game Scene Interior Production Sharing(2)
3ds Max Render Farm
Fox Renderfarm, as the best Maya render farm in the CG industry, the tutorial we shared for you this time is from the 3D artist Thepoly, on how to use 3ds Max and Maya** and other software to make a semi-realistic 3D game scene interior production process.
Taking the wooden table as an example below, first of all, we must determine the material to be sculpted and then look for the reference picture corresponding to it. Because the wood in the scene is too old, you need to find some pictures of old wood when looking for reference pictures, for example,
The carving process needs to be combined with actual life, so as to achieve a more realistic effect. In this scene, the wooden table is a bit decayed, so the edge of the table will not be made into a hard and sharp right-angle structure, so you can use a brush to make the right-angle part of the edge more smooth when carving. Reflect the texture to be expressed. Coupled with appropriate deformation, it can more vividly increase the visual impact (the modified place is slightly adjusted with the Move brush).
The wood in the scene has to make a sense of oldness, so cracks are a good way to show that the objects are worn out, and reference pictures are also needed at this time. For example, the cracks on the window plate can be carved according to the cracks in the reference picture. This can reflect the material better, more realistically, and faster.
The final high poly showed as below,
Low polygon model and baking
The final low-poly model uses ZBrush to complete the reduction and then imports into Maya to make the final low-poly model. This can get the low polygon model faster, but the wiring will be less neat. Note that near 90-degree angles, it is necessary to distinguish between smooth groups and split UVs.
The final baking used Marmoset Toolbag 3.
After baking all normals and IDs, import Substance Painter to make materials. It is worth noting that the normals baked in Marmoset Toolbag 3 and Maya use OpenGL mode in Substance Painter, and the normals baked in 3ds Max use DirectX mode. After successful import, the remaining textures are baked in the texture set-baking model texture.
Wall material basic process
First, we make the intrinsic color of the material. The intrinsic color layer can add filters to increase details.
To add dark colors, use the Dirt generator. Add a drawing layer to repair some unsatisfactory areas. Add bright colors and color changes to create a rough effect of wall corruption. Here mainly relies on hand-drawn layers to modify. Continue to add color changes and roughness changes. Try to get as close to the real effect as possible.
Finally, you can add an adjustment layer to increase the brightness. The material made in Substance Painter should be led to Marmoset Toolbag 3, which will be a little lower than the brightness seen in Substance Painter. The wall material is finished.
The furniture in the scene is gray and dark because of the old feeling. The damage and abrasion are to reflect the original color of the wood, so the saturation of the color will be higher than the color of the surface.
Before finally importing into Marmoset Toolbag 3, the lights must be set before rendering. Establishes the main light, and the auxiliary light supplements the light source to maintain the brightness of the picture. And to turn on high DPI and oversampling, turn on GI global illumination and AO environment occlusion.
The final effect showed as below,
I hope this learning and sharing can help you.
Behind The Scenes: Sagittarius(2)
Maya render farm
Three main buildings were built according to the position in the original painting, and then the entire background was spliced out with these three buildings. This step is relatively simple.
Then put the objects made at this stage into the main scene, adjust a few materials, such as metal, leather, SSS materials, give the corresponding model, add an HDR skylight, and test the entire scene.
Now it's time to break the symmetry between the character's head and the monster's head model. This step needs to enrich the details on the skin, such as adding some wounds or something.
At this time, you need to split the head UV. Use ZBrush to export the displacement map to view the effect in Maya.
The ZBrush export displacement method is as shown above: first connect the displacement map exported by ZBrush in Maya. You need to open the subdivision level in the object properties. In this way, the object can restore the display effect in ZBrush.
When the model with refined objects is put into the scene, you can start to adjust the lights. Decrease the intensity of HDR light and use it as a secondary light source, with a film light source as the main light source, and adjust the effect of the light and dark boundary line in the sun with parallel light. The effects at this stage are as follows.
Use of character XYZ
Here you can start to try the smallest level of detail processing on all objects, such as the XYZ replacement texture of the skin, the metal texture and damage traces of the equipment itself, and the skin after the XYZ replacement, as shown below:
The effect of making detailed equipment:
The model is almost finished. Now it ’s time to split the UV. Here, UDIM is used to facilitate textures, a set of UDIM, a set of background, a set of skin, and then the cumbersome UV process.
SP texture production
After splitting the UV, the textures are created. All textures are made in the substance painter. A set of UDIM models are directly imported into the Substance Painter. As long as the options shown in the figure above are checked, they will be automatically recognized. When making textures, it is mainly due to color changes. The color changes should be rich but they cannot cover his original main color. Of course, the roughness changes are also very important. The following picture is the effect of Substance Painter.
The skin needs to be color maps, roughness maps, and rough range maps in the Substance Painter. The color maps are obtained by mapping the real face skin (the same method as XYZ). The roughness can be converted using the previous XYZ three-channel conversion then overlay some random ALPHA textures. The rough range is to control the position of the highlight, and it is hand-drawn. Generally, the forehead, the tip of the nose, and the highlights on the cheeks are more obvious.
Maya texture connection
After the three pictures are drawn, you can add the rough range in Maya, and paste the roughness below:
The color map is attached to the arrow in the figure below, and then radius gives a red color (light transmission effect).
We also need to apply the XYZ displacement and ZBrush-derived displacement that we created earlier to the displacement channel,
After the first pass of the texture, you can test the rendering, and then you need to adjust it according to the picture effect. Then the monster's head also needs to draw a map. Just like the human head, it needs three pictures of color, roughness and rough range.
The next step is to replace the wings. The method of wings uses patch stitching. First, make a single feather map in three or four different positions, and then stitch it. What it looks like at this stage,
Xgen hair production
Let's start adding hair, XGEN for hair, each part should be separated into different Maya files when making hair, and then integrated into the main scene. Mainly how to import into the main scene, first clean the Maya file of the Xgen hair that is made separately, leaving only the growth bodies and the grown hair, and then directly import the cleaned Maya hair file in the main scene, and the imported file will be prefixed with a name. The prefix cannot be deleted, otherwise an error will occur.
When rendering, you need to add AOV channel as a later auxiliary layer.
The entire scene is divided into the foreground and the background for rendering separately, and the later scenes are stacked together to facilitate the final effect adjustment. Finally, the final picture is rendered,
And the detail,
Behind The Scenes: Sagittarius(1)
Maya render farm
Fox Renderfarm, a powerful but affordable maya render farm, brings you a sharing of behind-the-scenes of Sagittarius made by Maya** from 3D artist Zihao Zhou. This article introduces how to use the animation CG model production process to restore a 2D original painting Sagittarius, and tells the whole creative process, especially sharing some special places' processing methods and skills.
This artwork contains many elements in the early stage, so it is necessary to consider clearly in the planning of the early stage of production, otherwise it will affect the later model production. Zhou divides the whole work into several parts,
- Character skin: human head, body, horse and monster head.
- Equipment: protagonist equipment, wings and soldier equipment.
Construct the entire scene with simple geometric objects, and make some adjustments after determining the position of the camera lens, and try to place all the objects as well as possible. The location of the light can also be tested to find the most suitable place. It is necessary to make some model objects in ZBrush and put them into the scene as soon as possible to enrich the scene. Don't pay too much attention to the lights and details.
When everything is there, you can start making more accurate models to replace things in the scene. During the construction process, gradually adjust the shape or the position of the wrong part. In the production of this scene, I used to separate the equipment into separate Maya and ZBrush for production. There are two main reasons for this.
- There are many symmetrical objects in the equipment (although it will definitely break the symmetry later), but the previous symmetry can be improved efficiency (making high-precision models directly in the scene will be inconvenient) .
Do it separately in each Maya file. It can save some time for testing the material effect later, you can put the main scene lights into a single Maya equipment in the file, you can increase the speed of rendering. Of course, in the end, the scene effect in the lens shall prevail.
In this step, you can sculpt the protagonist's face, muscles and monster head (except for the symmetrical sculpture of muscle, human face and monster face, which is convenient for making the overall effect, and then break the symmetry later), the muscle part uses an existing body. Use the mask and move tool in ZBrush to adjust the desired appearance, and put it in the scene to match the position of the body. The head of the protagonist and the head of the monster are also placed in the placeholder model before the scene is replaced. The horse's body also needs to be sculpted with muscles, but it will be covered with a large area of hair in the later period, so the sculpting need not be too detailed.
Equipment is produced separately. When this large piece of equipment is determined, the next important thing is the pattern on the body, and there have been many versions of this pattern. In the production process, depends on the topology pattern of the equipment. Here I use Maya's own topology tool to first select the model to be adsorbed, click the magnetite, and then click quad draw to draw on the model.
Then add a line in the middle of the drawn shape, and put the added line in the direction of the normal (to turn off the suction command) so that the pattern is there, then only need to extrude the thickness, pay attention to the corners wiring.
In addition to a larger pattern, some small patterns attached to the surface of the equipment are needed to enrich it. This level is actually to find some suitable black and white pictures of the pattern on the Internet for stitching to make a continuous square picture, and then expand the equipment. UV, stick the pattern to the position corresponding to UV.
Then paste the black and white pattern into the ZBrush replacement channel (need to flip the V direction), adjust the extrusion height, and take a look at the effect. Of course, the patterns on the equipment can be stitched using some online pattern materials.
The production of the armor part is more troublesome, because the shoulder armor looks like a broken metal effect, so first make the shoulder armor large, and then use the ZBrush grouping to divide the shoulder armor into multiple groups, and rewire the regrouped part. The effect of broken metal can appear. Of course, it is still necessary to manually adjust the gap between each piece of metal. By default, the gap size is the same, which is not good-looking.
Wings can first find a wing and place it to see the effect, and then replace it with your own wings.
How to Set Up a Maya Render Farm
Maya render farm
Maya is an excellent 3D production software owned by Autodesk. It is very famous in the animation design industry and belongs to the top design and production software. Especially in animation production, game characters, film special effects. For example, Blue Sky's animated movie Spies in Disguise in 2019 and Netflix's science fiction series Lost in Space. The issue of rendering is inevitably involved in animated film production. I believe that many people are worried about the rendering time is too long and the delivery time is too tight, but using the render farm is a good choice, so how to set up a Maya render farm?
Spies in Disguise
Lost in Space
How to build a Maya render farm? First of all, there are many issues to consider when setting up a render farm, such as the number of machines, the configuration of the machines, the network, labor, software, purchase of Maya's license, time cost, security, and your budget. You can refer to the following article: Is Building Your Own Render Farm The Best Option?
If the project budget is tight, and you don't want to spend too much time on setting up and managing the render farm, you can consider using a commercial Maya render farm. After all, it can be used anytime, anywhere, and there is no project or the project has not reached the rendering progress when you don’t need to vacate your resources or ask someone to maintain it. Especially in the process of using your local render farm, if you encounter unexpected problems, you can also get the assistance of professional technicians. You don't have to spend time researching the problem yourself until you find a solution. During this period, if you are in a commercial Maya render farm, Some projects are estimated that you have finished rendering. The Top Benefits of Online Render Farms
Of course, choosing a commercial render farm also requires choosing a large and well-known professional online render farm, which guarantees safety. In addition to security, the number of machines and the quality of service can be provided very well. For example, the TPN-Accredited cloud render farm, Fox Render**farm, in addition to providing massive rendering nodes, 24/7 services, there are professional TD personnel to solve technical problems for you at any time. At the same time, Fox Render**farm is the rare render farm in the CG industry that supports both CPU and GPU rendering. Various mainstream software, renderers, and plug-ins are supported, as well as personalized customization services. Fox Render**farm provides multiple submission methods for users to choose from. In addition to web page upload, rendering, monitor, and download, offline software applications can also be used for asset uploading, render**ing, monitoring, and downloading. It's very easy to use and it only takes a few steps.
If you are still hesitating to build local Maya render farm or use a commercial render farm, why not directly get a free trial of $25 on Fox Render**farm**, which can be used as a real experience reference.
Behind the Scenes: 3D Game Scene
Maya render farm
Today, we will share a production process of a 3D game scene. This project uses the standard PBR process, and uses 3ds Max to complete the basic modeling, determines the proportion of objects. Finish sculpting in ZBrush for high poly, topology in Maya, baking in Marmoset Toolbag 3, finish making PBR materials in Substance Painter, and finally use Marmoset Toolbag 3 for rendering.
The production is divided into three parts: 1. Modeling 2. Material production 3. Rendering
Before production, determine what materials, objects, and overall proportions are in the scene, and then determine the production order according to the proportions. Most of the scene in this scene is walls and stones and wooden furniture, so they are the focus of the production.
Quickly set up their approximate proportions in 3ds Max.
Complete the medium face number model
After establishing the proportion of large objects, continue to build the remaining objects in the scene. And further enrich the details of the model. There are a large number of repeated objects, and several different shapes can be made.
Taking the above book as an example, many different types have been produced, which can reduce visual repetitions very well.
There are two pieces of cloth in the scene. Here we take the pillow as an example,
This production uses the cloth calculation that comes with 3ds Max. The specific process is as follows,
Create blocking objects:
Create a pillow base model (the higher the number of segments, the better the simulation effect),
Squash and find the cloth in the modifier,
The end result is a simple pillow model. (The shape of the wrinkles of the fabric should be well organized, and the top is a simple demonstration. Similarly, curtains may have better effects).
Finally in 3ds Max medium face model,
Next, import the model into ZBrush for sculpting.
High Poly sculpture
Here is an example of a house's walls and stones:
The medium face model was imported into ZBrush because of the irregular shape of the stone. You can use Dynamesh (dynamic mesh) in geometry editing to edit the model.
Brushes 1, 2, and 3 are used for large-scale carving of stones. Brush No. 4 adjusts shape to prevent penetration. Finally, brushes 5 and 6 are used to flatten some edges and corners to make the structure hard. Thus reflecting the texture of the stone.
The big ideas on the wall are the same. Finally, use the surface noise, adjust the curve, and the noise ratio to simulate the graininess of the wall and stone surface.
Finally, add some details such as cracks and bumps to enrich the picture. Use the brush No. 6 mentioned above to suppress it again for better results.
Take the wooden table as an example below, first determine the material to be carved, and then look for the corresponding reference picture. Because the wood in the scene is too old, you need to find some pictures of old wood when looking for reference pictures, for example,
The sculpting process needs to be combined with actual life, so as to achieve a more realistic effect. In this scene, the wooden table is a little bit decayed, so the edge of the table will not be made into a hard, right-angled structure, so when carving, you can use the brush to make the right-angled portion of the edge more rounded, so reflect the texture you want to express. With appropriate deformation, it can increase the visual impact more vividly (adjust the place with the Move brush a little).
The wood in the scene needs to have an old feeling, so cracks are a good way to show that the objects are worn out. At this time, you need to use the reference picture. For example, a crack on a window plate can be carved according to the crack in the reference figure. This can better and more realistically and quickly reflect the material.
The final High Poly display,
Low polygon model and baking
The final low-poly model was reduced using ZBrush, and then imported into Maya for final low-poly model production. This can get Low Poly models faster, but the wiring will be relatively less neat. Note that it is important to distinguish between the smooth group and split UV where the angle is close to 90 degrees.
The final baking uses Marmoset Toolbag 3.
After baking all the normals and IDs, import Substance Painter to make the material. It is worth noting that the normals baked in Marmoset Toolbag 3 and Maya use Opengl mode in Substance Painter, and the normals baked in 3ds Max use DirectX mode. After successfully importing, bake the rest of the textures in the texture set-baking model textures.
Wall material basic process
First of all, we make the natural color of the material, and the natural color layer can add filters appropriately to increase the details.
Add a dark color and use the Dirt generator. Partially unsatisfactory add a drawing layer to fix it. Add bright colors and color variations to create the approximate effect of wall decay. Here, it is mainly modified by hand-drawn layers. Continue adding color changes and roughness changes. Try to be as close to real effects as possible.
Finally, you can add an adjustment layer to increase the brightness. The material made in Substance Painter should be imported into Marmoset Toolbag 3 and the brightness will be a bit lower than that seen in Substance Painter. The wall material production is complete.
And because the furniture in the scene should reflect the old feeling, the overall color is grayish and dark. The damage and abrasion are to reflect the original color of the wood, so the color saturation will be higher than the surface color.
Finally, before importing into Marmoset Toolbag 3, you need to set up the lighting, the main light establishes the main light, and the auxiliary light supplements the light source to maintain the brightness of the picture. And to turn on high DPI and oversampling, turn on GI global illumination and AO environment occlusion.
I hope this sharing will help you.
Behind the Scenes: Demon Warrior Character
V-Ray for Maya
Hey, I’m Dong Chao. I used to work at the Industrial Light & Magic (ILM), Lucasfilm's VFX and animation studio in Singapore. Now I am a professional freelance artist and have long been involved in role-related work. This time I want to share some of the production process and experience of a 3D CG character produced in the project. I hope to help everyone. The software and plugins used in the production are Maya, ZBrush, Mari, UVLayout, Photoshop, shave and hair cut.
- Material collection:
Material collection is an important step. The more time you spend collecting material, the fewer mistakes you will make later. When creating the model of Demon Warrior, a large number of real materials and works references were collected using the Internet. In the design stage, I imagined that he was in ancient times, but did not have any obvious characteristics of the times, so I wanted to make some breakthroughs in the traditional dressing features of ancient China, and portray more mysterious patterns. I also tried brushes with different touches, and finally unified the whole character to make it feel complete. The following is part of the collected materials, put together for your reference.
- Basic model
The production of the model is based on a ready-made human basic model. After adjusting the proportion on it, the creation of all props and clothing is quickly performed. According to the size of the volume, start with the large-scale model parts first. The overall feeling of the character is shaped in the time, it is convenient to modify the proportion, and try to be as clean and even as possible on the wiring without having to portray too many details. But pay attention to the large turning and thickness of the armor, so that it is easy to sculpt later. If you need a large number of copies or later symmetrically sculpted objects, you can make them in the center first, and then copy and adjust the position to observe.
Regarding the cloth elements on the character, you can directly model and sculpt according to personal habits, and also use Maya's cloth system to solve. The benefits of the solution are more natural and achieve better folds with a lower number of faces. effect. The disadvantage is that it can't control the details of the folds more accurately than ZBrush. Because the character's cloth shape is relatively simple, the cloth solution method was chosen.
The hair of the character is made using Maya's nHair. In the early stage, the hair shape of the character needs to be built with polygons. The brush function built-in Maya can quickly convert the curve into the overall effect, and then copy the curve to create more rich changes. The second layer of thin curved surface will get more change and more natural hair effect in the later stage.
3.ZBrush high model production
ZBrush sculpting has always been my favorite link. Especially for this kind of project that needs to design a lot of details, I can boldly sculpt in ZBrush. The early materials have helped a lot at this time. When sculpting, generally according to the importance of the character, starting from the whole, first make some parts that account for the largest proportion of the character. Each part is a SubTool to facilitate later modification. Using different brush settings, you can sculpt the desired The detailed effect, whether it is a prop or a character, first make a large structure, and then refine it.
When creating some continuous sculpts in ZBrush, you can use Maya to make some elements first, and then convert them to alpha in ZBrush and map them directly to the model with UV. If the position is not perfect, you can import the model into Photoshop. Aligning alphas without sculpting one by one, it will take a lot of time.
ZBrush sculpting is the most detailed link in the entire model. People who have been using ZBrush for a long time suggest to establish their own familiar UI interface environment and their own custom brushes, which can greatly improve the production efficiency and effect during the production process. When sculpting a character, the collected alpha channel was used to portray the details.
- Model integration and UV
Export the lowest level of all models in ZBrush, and then use UVLayout to get a good UV splitting effect quickly. You only need to adjust patiently here, place the seams where you can't see as much as possible, and minimize the screen. UV stretching of the main part.
The model after splitting through UVLayout can be conveniently managed and laid out in Maya. All the models that need to be copied before sculpting and not sculpting are all arranged, because the number of final parts is relatively large, in order to bake the map faster For the convenience of drawing textures and adjusting materials, try to put objects of the same material in a UV unit.
Import all the determined UV models one by one into ZBrush in the order of SubTool. It should be noted that models exported from ZBrush can no longer be merged, split, or changed in topology. Otherwise, the models in ZBrush cannot be replaced.
Finally, you can set the baking texture in ZBrush. You can choose the texture you want to output. Here, only the displacement texture and the normal texture are baked. Because it is batch baking, the SubTool shows that you want to bake in one. Post the model on the map, ZBrush will automatically bake all and merge. The following are the parameter settings and the final results of the two posters for your reference.
- Map drawing
Next, start to create color maps for each part. First, a set of texture maps for men was selected for the body part. Using Mari's powerful texture mapping, a lot of details on the map are well retained.
Then import all the drawn layers into Photoshop to change the skin, draw some exaggerated details such as pores, dark spots, dust, etc. When drawing details, try to use high-resolution maps for drawing.
Finally, using some patterns made in PS, map Mari to the character's body according to his own design. In the end, it is still necessary to retain the transparent channel of this layer to facilitate later layer material adjustment.
The metal part is first imported into a material, and then the wear, scratches, smudges and other details are drawn in the Mari with a brush or material. Use Mari for hierarchical management. In the process of drawing, you can create a new mair shader. Simple Adjusting the material parameters for observation, you can also create a normal channel, import the previously baked normal maps and observe them to describe the details more conveniently, and then import the PSD into Photoshop for detailed description and color processing. Because our map is finally rendered in V-Ray, we need to set up a simple shader and light at the mapping stage, modify it after rendering and observation.
- Hair production
Use the previously made hair model, convert it into a NUBRS model, copy the spline curve above, and then use the nHair system to create new hair. It is recommended that you create different hairs according to different parts. This is convenient for adjusting different parameters. Try to use fewer hairs when testing hair. White hair has always been a headache, in order to get better highlights and later sub-channel rendering, I used V-Ray hair shaders on the hair system, and created 3 shaders according to different parts. This can achieve more variety of controls, you can hide other objects for quick testing.
The fuzz on the edge of the clothes is a stiff mane in my imagination, so I chose to create it with the Shave and Hair Cut plug-in. The comb function can quickly comb the approximate effect, and then test render.
7.Pose rendering composition
The character itself is a humanoid, so the HumanIK built-in skeletal system, HumanIK, was used to quickly create standard bones. Then, using HumanIK's automatic binding function, a set of bindings with simple functions can be obtained. The final pose can be placed according to the desired effects and props.
The rendering uses Maya's V-Ray for rendering. All the lights and shaders use V-Ray's shaders, which not only can get the desired effect faster, but also facilitate the subsequent sub-channel rendering. For lighting, one main light source, two backlights and one auxiliary light, and an additional sky light or metal reflected light can be added according to the effect.
The overall material is mainly metal, skin and leather. V-Ray's 3S texture ball is used for the skin. Simple settings can achieve good results. For leather, due to its characteristics, it is necessary to make a more precise highlight map at the mapping stage, so that the smooth and rough different areas on the leather can be better distinguished by the rendering of V-Ray, and the characteristics of smudges on leather are the only way to Distinguish from metal highlights.
In V-Ray, set all the channels needed to output the final composition.
The synthesis phase is relatively simple, mainly using channels to make fine adjustments. In addition, adjustments are made to layers such as fog, dust, flames, clouds, noise, edge light, and depth of field. The color must be adjusted to make it cooler and lower Saturation makes the picture present a relatively quiet and mysterious atmosphere.
3D Works Mouse! Mouse! Production Analysis
Maya render farm
This work is the result of whim. The mouse is a relatively common animal at home. When you see it, you want to shoot it on the bricks. The mechanical characters of science fiction and the ordinary environment can have a certain association. From the visual performance, through the stop of the instantaneous action, the viewer has a kind of front and back association effect, let's share the production process with you.
Software used: 3ds Max, V-Ray, PS
First, the composition
Firstly, you only need to set up a simple scene. The object is just a basic form instead. The focus is on the choice of the position and the layout of the object. Since the CG single frame rendering is performed, the object layout can be performed according to the camera view after the camera is positioned, and the layout and positioning map are as follows:
Second, modeling and UV
- The production of the hand
Modeling begins after the composition and general object layout are completed. This is a time-consuming basic work, but it is also the key to determining the final quality of the work. It requires basic modeling ability and patience. First select the opponent model, followed by the arm, because the movement of the whole hand determines the "pose" of the work, so I chose to make the first. Then carry out UV deployment and splitting. Specifically as shown below:
- The production of clothing
Because the character has a large movement, the clothes must be matched to have a real dynamic performance. The clothing was first modeled in 3ds Max and then imported into Mudbox for engraving. After engraving, it was imported into 3ds Max. Again, the model finishes splitting the UV for the next mapping. Figure:
- The production of the head and expression
With the correct posture and the appropriate expression, the character of the work can be more vivid and interesting. In the head of the character, I made some settings in advance, and the hat of the character was also carved in Mudbox, and the number of faces was reduced after completion. Level import 3ds Max to split the UV.
- The production of mouse
The mouse must conform to the appearance of the mouse and mechanize it. It is necessary to make some set-ups in advance. Here, only the modeling process and the UV split map are shared.
Since the mouse is a mechanical model, there is no need to rig the bones, and only the "father-child association" can be used to put POSE. The links are as follows:
- Model production of small objects
The model of small objects is relatively simple to make, and some of them can be completed by referring to the network search image. Here, I only share the partially completed production pictures.
Third, the material and texture
Some of the main models of this work were UV-split and mapped in Mudbox. The rest of the small objects are just attached to the map in 3ds Max and a UV Maping modifier is added. It only takes some time to split the UV, which saves a lot of time.
- Arm texture
After splitting the UV in 3ds Max, you can import the model into Mudbox for texture mapping. Mudbox is very powerful in drawing textures. What I use here is the "projection" function in Mudbox. Map rust or dirt directly onto the imported model and map the model in real time.
Character clothing and head can be drawn in this way. What is needed is the splicing of texture.
- Clothing and head texture
It is similar to the method of making an arm.
Mouse texture and chair back texture performance
- Complete the texture of all objects
After the texture is drawn in Mudbox, the texture is exported to the PS, and the adjustment in the PS can be added to the corresponding object channel in 3ds Max.
Fourth, lighting and rendering
- Lighting layout
Once you've finished making all the textures, you can start designing and laying out the lights. This is indoors, and at night, the light from the computer is dark. The main light source comes from:
A. The dimly lit front wall lamp (the wall lamp is not visible here, so only the light is used to express his presence).
B. The computer screen light.
C. The weak light outside the window.
The specific layout is as follows:
- Rendering settings
After the lighting layout, the rendering parameters can be adjusted and tested. The screenshots of the V-Ray rendering settings are shown in the figure:
3.Output the rendering
When outputting, in addition to the output of the rendered image itself, only the output of the Z channel is made.
Fifth, the adjustment of post-production
I chose to add depth-of-field effects with DFusion and color correction in post-production, and imported the file into DFusion to add depth-of-field effects with Depth Blur.
The final rendering is as below:
Author: Liang Xi
The Reason And Solution When Maya Reference Alembic Can Not Be Read
Maya render farm
This article is organized from the leading cloud render farm in the CG industry, Fox Render**farm. If your Maya** can’t read the Reference Alembic, here is a sharing about the reason analysis and solution, I hope it can help you.
About the problem When using the alembic cache in Maya, in general, one is directly imported to the scene, and the other is through the reference of Maya, which is referenced to the scene. The problem is in the case of the reference to the scene.
In the process of using the reference abc cache to multiple machines on the render farm, there will be reference’s abc can be loaded normally, and some can not be loaded, open the scene abc will find that the loading is not successful. By looking up the reason, we found that the AbcImport plugin on the machine that can load the reference abc is loaded automatically, but the plugin of the machine that can't load the abc cache is not loaded.
There are now two scenarios, one is to directly import the abc scene, and the other is the reference abc scene.
The situation of without the abc plugin:
After the import abc scene is opened, the loading of abc is normal. The reference abc scene is empty in the scene after opening, and abc does not load normally.
In the case of adding the AbcImport plugin, the import abc scene and the reference abc scene can be loaded normally after the abc cache is opened.
The reason analysis
In the past, we deliberately saved the scene in the format of ma, because the main problem is in ma.
Open the ma scene with the text tool notepad++, we can see that in the Maya scene it will first use the plugin used in the scene to require,
In this way, when the scene is opened, there is a required plugin, which will automatically load the plugin. If there is a node of the corresponding plugin in the scene, the node can read normally. The above is the ma scene of import abc. And in the scene with a reference file:
You can see that the reference file is before they require, which causes the reference abc cache to not trigger the require plugin. Without the AbcImport plugin, the abc reference has problems and cannot be loaded properly.
About the solutions
After finding out the reason, now the solution is to let AbcImport to be loaded automatically before the scene is opened. A few methods as follow:
- On a machine, after opening Maya, load the AbcImport plugin itself, so that when the Maya is opened, the AbcImport plugin will be loaded automatically. The requires of the scene will be triggered to add the plugin.
- Writing an automatic loading of the AbcImport plugin in Maya's userSetup.mel
- Or directly add a ‘requires’ in front of the reference node in the reference scene's ma file.
Requires -nodeType "AlembicNode" "AbcImport" "1.0"
Maya's ma format scene file can be used to see the entire scene in Mel's language description of the scene. In some specific cases, the scene can be edited by directly opening the ma file with a text tool.
Animated Short Film "Survival" Scene Production Analysis
Arnold for Maya
This time, our render farm brings you an analysis of the animated short film "Survival" scene production. As the leading cloud render farm in the CG industry, Fox Render**farm has always been committed to providing artists with affordable, reliable and secure cloud render**ing services. It also encourages outstanding artists to make good works and promote the production of their works, and let more people know them and their excellent works.
This article brings about the production and analysis of the animated short film "Servival" produced by CG artists Li Yicong, Sun Ningning, Xing Luyue, Shi Tao and Li Yongcong. The software used is Maya, Arnold, AE, Nuke and so on.
A brief introduction to the short film
The short film tells that in a city where energy is exhausted after the end of the world, a lonely little robot relies on finding a waste battery in the city every day to live alone, until he finds that the city has no more disposable batteries to use, but he has no choice but to go home. Packing up the luggage, reluctantly pulling his own stroller, leaving the city, embarked on the road to find new energy.
The specific production process
The script is written by members and then determined through multiple discussions based on the technical level and conditions that can be achieved, and then the character design and scene design by the pre-art.
I found a lot of references during modeling, and I used to make up for a mechanical principle for a while. However, the animated character and mechanical design are not exactly the same. Consider the functions of action binding, emotional expression, and plot. Figure 1 is a partial reference of the character. The mapping is mainly used for Ddo and Mari.
One of the scenes is the home of the robot, which lives in an abandoned train station. This lens acts as an empty mirror to ease the rhythm and explain the living environment.
The robot looked for a day's battery, and the harvest was not much. Because it was far away, it almost lost a life on the way home. The lighting is set in the evening, except that it feels very dull, and it also implies that the small robot has lived in this city for a long time. On the other hand, the evening light effect, the light ratio is strong, the angle between the light, the camera and the object is relatively small, and the surface of the object will produce a beautiful Fresnel reflection. The figure below shows the material properties of the railway.
Here you can use three or more textures to represent an object, color map, reflection map, and normal map or bump map. If it is close-up, you need to add a roughness map to make the object produce more rich changes.
Backlighting also works well for translucent objects, such as the dirty old glass on the bridge, the grass on the ground, and the semi-transparent effect of all the plant leaves on the vine. In addition, the grass adjusts the wind output in SpeedTree and then imports it into Maya for material adjustment. Special plants need to be created in SpeedTree. Add force field to shape the plant into the shape and growth trend that you want, and then export. Creating plants in SpeedTree allows you to export wind animations to Maya, and plants can move to increase the authenticity of the scene.
Plants imported into Maya
Create a render layer in Maya, put all the objects into the new render layer, and then add the Arnold base material for layer coverage so that the entire scene is the same material and the texture is ignored. Benefits are,
First, when lighting, it can be not affected by the texture of the surface of the object. It can observe and adjust the position of the light and shadow well, and use light and shadow to compose and shape the volume of the object.
Second, remove the texture, refraction, etc., the test rendering speed is fast, the adjustment can get feedback quickly. Here you can see that this picture is not completely retrograde, the light source is to the left. At the time, when lighting in the scene, I hope that one side of the locomotive can produce a bright surface, and the lower left corner is too bright, so that the light of the whole scene is too average.
Adjust the lights and start making material adjustments and renderings. If you find that many places need to be adjusted after rendering, you can reduce the sampling of all aspects, in order to save rendering time, after many adjustments to the screen you want, if you want the work more realistic and reduce the rendering time, you can use a Maya render farm.
In Arnold, it is generally necessary to separate several layers，
In the home of the small robot, this is a parting scene, where the warm shades of light are used to make a warm and sad feeling, so use the Maya spotlight to simulate the sun's rays, let it produces a light fog to simulate the atmosphere, which can increase the drama and authenticity of the scene.
Post-production and composite editing
The short film uses Arnold to render a multi-channel synthesis scheme with a total of 13 layers, which is finally synthesized in Nuke and AE.
Making A Robots of League of Legends with Maya and Houdini
houdini render farm
The idea of this robot comes from the game League of Legends, using software such as Maya, ZBrush, Substance Painter, Houdini.
1.Reference I had the idea of making this robot after I saw a reference picture as below.
2.Maya I made the approximate shape of the model in Maya, especially the part of the body structure. Although there is no detail, I try to construct the model.
Now importing ZBrush adjustments, this production process does not use ZBrush's hard surface, but adjusts the large structure of the large model inside, without adding detail adjustments.
After ZBrush was adjusted, it was imported into Maya again, mainly to adjust the details of the model again. In particular, the parts of these edge details need to be adjusted to the position of the wire.
3.UV UV production is very simple, I used the unfold automatic expansion of the UV function, it is very useful in terms of UV extension.
4.Substance Painter After splitting the UV, make a material using Substance Painter. The material of the Apple robot is very simple, mostly metal and luminescent materials. I made a texture using the basic material of Substance Painter.
5.Pose Exporting textures in Maya textures, I made simple settings, Skinning, and Paint Skin Weights in Maya. Because you don't need to do role dynamics, the above steps are as simple as you can.
6.Rendering After the model is finished, I chose to use Houdini for rendering. The material of Houdini is very similar to the PBR material of Substance Painter. Instead of using the Houdini default lighting material, I used light in lighting and created an environment map. Mantra is used for realistic rendering.
The material of the robot's light-emitting part uses the glow node that Houdini comes with, but slowly adjust to the desired effect.
The final rendering.
A Production Process Sharing Of The 3D Works "Redemption" By Arnold For Maya
Arnold for Maya
I am very happy to share with you the production process of the work of Redemption. I have shown a way of thinking and process of making this work. And I also learned something new from the project, I hope everyone can learn something from it. The software used in this work is Maya, Zbrush, Marvelousdesigner, SubstancePainter. Collection of materials Collecting the information you need is a very important part of the project before it is started. This is a very important part. I have been collecting enough information before I start the next step. But some of my friends, they will look for reference materials while making this way, it will take more time and effort, and the effect is not very good. I think that while doing the reference, it is better to take a day or two to collect and classify the required materials, so that you can also collect your own database.
Production 1. Composition The composition of the work is very important, directly affecting the final effect. I use a relatively small composition of the work of "Redemption", which is very interesting in the composition.
- Model At the beginning of the production, I will quickly build a simple model, not too detailed, all models are as simple as possible, the sooner the better. This stage of production I try to control in one day. At this stage, I need to complete the simple and fast, then determine the angle of the camera and arrange the lights.
- High poly workflow The high poly workflow phase is a very important part. Whether it is game production or film and television model production, the high poly workflow stage is very important. After the large shape and the composition of the screen are determined, I will start from the main body of the character and proceed progressively according to the composition and the primary and secondary relationship. According to the principle of subject, local, subject. Some common brushes and some post-production texture aids used by individuals as follow,
- Texture I used the more popular PBR process in drawing the texture. Substancepainter is the most common drawing software for this type of texture drawing. The character face and skin use Mari. Character skin replacement, skin color is made in Mari, Mari is a very powerful software.
- Render and post-synthesis This work uses the Anorld for Maya, and the entire rendering process takes about 4 days. In addition, a separate AOV layer is rendered to facilitate further modification of the color and scene atmosphere in the PS.
The above is the whole process of the project of "Redemption", I will introduce it to you here. I also found some deficiencies in the production, I hope that my production can give you new ideas, we can all learn new things in the project.
BY: Qi Gong
China Film Administration Release New License Logo
Maya 2018 Arnold's Method of Making Motion Vector Blur in AOV Channel
How the Redshift Proxy Renders the Subdivision
Arnold\ V-Ray\ Octane\ Conora\ RedShift, Which Renderer Is The Best?
Arnold\ V-Ray\ Octane\ Conora\ RedShift, Which Renderer Is The Best?
Corona Renderer Learning - Denoising
Is This Simulated By Houdini? No! This Is Max!
Arnold Render Farm | Fox Render Farm