The first thing to do is to be sure that all the required add-ons are enabled in the preferences; in this first recipe, we need the Images as Planes and Copy Attributes Menu add-ons. When starting Blender with the factory settings, they appear gray in the User Preferences panel's Add-ons list, meaning that they are not enabled yet. So, we'll do the following:

Although not strictly necessary, it would be better to have the three (at least in the case of a biped character, the Front, Side, and Back view) templates as separated images. This will allow us to load a specific one for each view, if necessary. Also, to facilitate the process, all these images should be the same height in pixels.

In our case, the required three views are provided for you in the files that accompany this book. You will find them in the templates folder. The Import Images as Planes add-on will take care of loading them into the scene:

We used a Python script, which is an add-on, to import planes into our scene that are automatically UV-mapped with the selected image, and inherit the images' height/width aspect ratio.

To have the textures/templates clearly visible from any angle in the 3D view, we have enabled the Shadeless option for the Planes materials; we did this directly in the importer preferences. We can also set each material to shadeless later in the Material window.

We then used another add-on to copy the attributes from a selected object, in order to quickly match common parameters such as location, scale, and rotation:

The template planes aligned to the x and y axis (Front and Side views)

The imported Planes can be placed on different layers for practicality; they can also be on a single layer and their visibility can be toggled on and off by clicking on the eye icon in the Outliner.

The first thing to do is to be sure that all the required add-ons are enabled in the preferences; in this first recipe, we need the Images as Planes and Copy Attributes Menu add-ons. When starting Blender with the factory settings, they appear gray in the User Preferences panel's Add-ons list, meaning that they are not enabled yet. So, we'll do the following:

Although not strictly necessary, it would be better to have the three (at least in the case of a biped character, the Front, Side, and Back view) templates as separated images. This will allow us to load a specific one for each view, if necessary. Also, to facilitate the process, all these images should be the same height in pixels.

In our case, the required three views are provided for you in the files that accompany this book. You will find them in the templates folder. The Import Images as Planes add-on will take care of loading them into the scene:

We used a Python script, which is an add-on, to import planes into our scene that are automatically UV-mapped with the selected image, and inherit the images' height/width aspect ratio.

To have the textures/templates clearly visible from any angle in the 3D view, we have enabled the Shadeless option for the Planes materials; we did this directly in the importer preferences. We can also set each material to shadeless later in the Material window.

We then used another add-on to copy the attributes from a selected object, in order to quickly match common parameters such as location, scale, and rotation:

The template planes aligned to the x and y axis (Front and Side views)

The imported Planes can be placed on different layers for practicality; they can also be on a single layer and their visibility can be toggled on and off by clicking on the eye icon in the Outliner.

Although not strictly necessary, it would be better to have the three (at least in the case of a biped character, the Front, Side, and Back view) templates as separated images. This will allow us to load a specific one for each view, if necessary. Also, to facilitate the process, all these images should be the same height in pixels.

In our case, the required three views are provided for you in the files that accompany this book. You will find them in the templates folder. The Import Images as Planes add-on will take care of loading them into the scene:

We used a Python script, which is an add-on, to import planes into our scene that are automatically UV-mapped with the selected image, and inherit the images' height/width aspect ratio.

To have the textures/templates clearly visible from any angle in the 3D view, we have enabled the Shadeless option for the Planes materials; we did this directly in the importer preferences. We can also set each material to shadeless later in the Material window.

We then used another add-on to copy the attributes from a selected object, in order to quickly match common parameters such as location, scale, and rotation:

The template planes aligned to the x and y axis (Front and Side views)

The imported Planes can be placed on different layers for practicality; they can also be on a single layer and their visibility can be toggled on and off by clicking on the eye icon in the Outliner.

We used a Python script, which is an add-on, to import planes into our scene that are automatically UV-mapped with the selected image, and inherit the images' height/width aspect ratio.

To have the textures/templates clearly visible from any angle in the 3D view, we have enabled the Shadeless option for the Planes materials; we did this directly in the importer preferences. We can also set each material to shadeless later in the Material window.

We then used another add-on to copy the attributes from a selected object, in order to quickly match common parameters such as location, scale, and rotation:

The template planes aligned to the x and y axis (Front and Side views)

The imported Planes can be placed on different layers for practicality; they can also be on a single layer and their visibility can be toggled on and off by clicking on the eye icon in the Outliner.

For this and the following recipes, there is no need for any particular preparations. Anyway, it is handy to prepare the two Empties to have markers in the 3D view for the 2.5 meters height of the character; so we'll do the following:

So, now we are going to place the first Image Empty in the scene:

We have used one of the most underrated (well, in my opinion) tools in Blender: Empties, which can show images! Compared to the Images as Planes add-on, this has some advantages: these are not 3D geometry and the images are also visible in the 3D view without the Textured Solid option enabled (under Shading) and in Wireframe mode.

The Image Empties appear as textured also in Wireframe viewport shading mode

Exactly, as for the imported Planes of the former recipe, the visibility in the 3D view of the Image Empties can be toggled on and off by clicking on the eye icon in the Outliner.

For this and the following recipes, there is no need for any particular preparations. Anyway, it is handy to prepare the two Empties to have markers in the 3D view for the 2.5 meters height of the character; so we'll do the following:

So, now we are going to place the first Image Empty in the scene:

We have used one of the most underrated (well, in my opinion) tools in Blender: Empties, which can show images! Compared to the Images as Planes add-on, this has some advantages: these are not 3D geometry and the images are also visible in the 3D view without the Textured Solid option enabled (under Shading) and in Wireframe mode.

The Image Empties appear as textured also in Wireframe viewport shading mode

Exactly, as for the imported Planes of the former recipe, the visibility in the 3D view of the Image Empties can be toggled on and off by clicking on the eye icon in the Outliner.

So, now we are going to place the first Image Empty in the scene:

We have used one of the most underrated (well, in my opinion) tools in Blender: Empties, which can show images! Compared to the Images as Planes add-on, this has some advantages: these are not 3D geometry and the images are also visible in the 3D view without the Textured Solid option enabled (under Shading) and in Wireframe mode.

The Image Empties appear as textured also in Wireframe viewport shading mode

Exactly, as for the imported Planes of the former recipe, the visibility in the 3D view of the Image Empties can be toggled on and off by clicking on the eye icon in the Outliner.

We have used one of the most underrated (well, in my opinion) tools in Blender: Empties, which can show images! Compared to the Images as Planes add-on, this has some advantages: these are not 3D geometry and the images are also visible in the 3D view without the Textured Solid option enabled (under Shading) and in Wireframe mode.

The Image Empties appear as textured also in Wireframe viewport shading mode

Exactly, as for the imported Planes of the former recipe, the visibility in the 3D view of the Image Empties can be toggled on and off by clicking on the eye icon in the Outliner.

As in the former recipe, no need for any particular preparations; just carry out the preparatory steps as mentioned in the Getting ready section of the previous recipe.

So let's start by adding the templates as background images; that is, as reference images only visible in the background in Ortho view mode and, differently from the previous recipes, not as 3D objects actually present in the middle of the scene:

Press 3 on the numpad to switch to the Side view, 1 to switch to the Front view, and Ctrl + 1 to switch to the Back view, but remember that you must be in the Ortho mode (5 key on the numpad) to see the background templates:

The N sidepanel settings to assign the background image to a view

As in the former recipe, no need for any particular preparations; just carry out the preparatory steps as mentioned in the Getting ready section of the previous recipe.

So let's start by adding the templates as background images; that is, as reference images only visible in the background in Ortho view mode and, differently from the previous recipes, not as 3D objects actually present in the middle of the scene:

Press 3 on the numpad to switch to the Side view, 1 to switch to the Front view, and Ctrl + 1 to switch to the Back view, but remember that you must be in the Ortho mode (5 key on the numpad) to see the background templates:

The N sidepanel settings to assign the background image to a view

So let's start by adding the templates as background images; that is, as reference images only visible in the background in Ortho view mode and, differently from the previous recipes, not as 3D objects actually present in the middle of the scene:

Press 3 on the numpad to switch to the Side view, 1 to switch to the Front view, and Ctrl + 1 to switch to the Back view, but remember that you must be in the Ortho mode (5 key on the numpad) to see the background templates:

The N sidepanel settings to assign the background image to a view

First, let's prepare the scene:

We are now going to move and extrude the vertex according to our template images, working as guides, and therefore generating a 3D geometry (thanks to the Skin modifier):

The Skin modifier is a quick and simple way to build almost any shape; its use is very simple: first, you extrude vertices (actually, it would be enough to add vertices; it's not mandatory to extrude them, but certainly it's more handy than using Ctrl + left-click to add them at several locations), and then using the Ctrl + A shortcut, you scale smaller or bigger the influence that these vertices have on the 3D geometry generated on the fly.

If you have already tried it, you must have seen that the more the complexity of the mesh grows, the more the generated geometry starts to become a little unstable, often resulting in intersecting and overlapping faces. Sometimes this seems unavoidable, but in any case it is not a big issue and can be easily fixed through a little bit of editing. We'll see this in the next chapter.

First, let's prepare the scene:

We are now going to move and extrude the vertex according to our template images, working as guides, and therefore generating a 3D geometry (thanks to the Skin modifier):

The Skin modifier is a quick and simple way to build almost any shape; its use is very simple: first, you extrude vertices (actually, it would be enough to add vertices; it's not mandatory to extrude them, but certainly it's more handy than using Ctrl + left-click to add them at several locations), and then using the Ctrl + A shortcut, you scale smaller or bigger the influence that these vertices have on the 3D geometry generated on the fly.

If you have already tried it, you must have seen that the more the complexity of the mesh grows, the more the generated geometry starts to become a little unstable, often resulting in intersecting and overlapping faces. Sometimes this seems unavoidable, but in any case it is not a big issue and can be easily fixed through a little bit of editing. We'll see this in the next chapter.

We are now going to move and extrude the vertex according to our template images, working as guides, and therefore generating a 3D geometry (thanks to the Skin modifier):

The Skin modifier is a quick and simple way to build almost any shape; its use is very simple: first, you extrude vertices (actually, it would be enough to add vertices; it's not mandatory to extrude them, but certainly it's more handy than using Ctrl + left-click to add them at several locations), and then using the Ctrl + A shortcut, you scale smaller or bigger the influence that these vertices have on the 3D geometry generated on the fly.

If you have already tried it, you must have seen that the more the complexity of the mesh grows, the more the generated geometry starts to become a little unstable, often resulting in intersecting and overlapping faces. Sometimes this seems unavoidable, but in any case it is not a big issue and can be easily fixed through a little bit of editing. We'll see this in the next chapter.

The Skin modifier is a quick and simple way to build almost any shape; its use is very simple: first, you extrude vertices (actually, it would be enough to add vertices; it's not mandatory to extrude them, but certainly it's more handy than using Ctrl + left-click to add them at several locations), and then using the Ctrl + A shortcut, you scale smaller or bigger the influence that these vertices have on the 3D geometry generated on the fly.

If you have already tried it, you must have seen that the more the complexity of the mesh grows, the more the generated geometry starts to become a little unstable, often resulting in intersecting and overlapping faces. Sometimes this seems unavoidable, but in any case it is not a big issue and can be easily fixed through a little bit of editing. We'll see this in the next chapter.

So, let's suppose that we want the arms to be posed more horizontally and widely spread:

Creating the rig (that is the skeleton Armature made by bones and used to deform, and therefore, animate a mesh) for our character's base mesh is really simple:

By clicking on the Create Armature button, the Skin modifier creates a bone for each edge connecting the extruded vertices, it adds an Armature modifier to the generated base mesh, and automatically assigns vertex groups to the base mesh and skins them with the corresponding bones.

The bones of this Armature work in Forward Kinematics, which means they are chained following the child/parent relation, with the first (parent) bone created at the Root location we had set at step 3 of the Getting ready section.

Note that the bones of the Armature can be used not only to rotate limbs, but also to scale bigger or smaller parts of the mesh, in order to further tweak the shape of the base mesh.

So, let's suppose that we want the arms to be posed more horizontally and widely spread:

Creating the rig (that is the skeleton Armature made by bones and used to deform, and therefore, animate a mesh) for our character's base mesh is really simple:

By clicking on the Create Armature button, the Skin modifier creates a bone for each edge connecting the extruded vertices, it adds an Armature modifier to the generated base mesh, and automatically assigns vertex groups to the base mesh and skins them with the corresponding bones.

The bones of this Armature work in Forward Kinematics, which means they are chained following the child/parent relation, with the first (parent) bone created at the Root location we had set at step 3 of the Getting ready section.

Note that the bones of the Armature can be used not only to rotate limbs, but also to scale bigger or smaller parts of the mesh, in order to further tweak the shape of the base mesh.

Creating the rig (that is the skeleton Armature made by bones and used to deform, and therefore, animate a mesh) for our character's base mesh is really simple:

By clicking on the Create Armature button, the Skin modifier creates a bone for each edge connecting the extruded vertices, it adds an Armature modifier to the generated base mesh, and automatically assigns vertex groups to the base mesh and skins them with the corresponding bones.

The bones of this Armature work in Forward Kinematics, which means they are chained following the child/parent relation, with the first (parent) bone created at the Root location we had set at step 3 of the Getting ready section.

Note that the bones of the Armature can be used not only to rotate limbs, but also to scale bigger or smaller parts of the mesh, in order to further tweak the shape of the base mesh.

By clicking on the Create Armature button, the Skin modifier creates a bone for each edge connecting the extruded vertices, it adds an Armature modifier to the generated base mesh, and automatically assigns vertex groups to the base mesh and skins them with the corresponding bones.

The bones of this Armature work in Forward Kinematics, which means they are chained following the child/parent relation, with the first (parent) bone created at the Root location we had set at step 3 of the Getting ready section.

Note that the bones of the Armature can be used not only to rotate limbs, but also to scale bigger or smaller parts of the mesh, in order to further tweak the shape of the base mesh.

Note that the bones of the Armature can be used not only to rotate limbs, but also to scale bigger or smaller parts of the mesh, in order to further tweak the shape of the base mesh.

Let's prepare the mesh and the view:

Well, in our case, everything went right with the Skin modifier, so there is no need for any big editing of the mesh! In effect, it was enough to delete the first Mirror modifier (that we actually used mostly for visual feedback) to get rid of all the overlapping faces and obtain a clean base mesh:

The "clean" mesh with new Mirror and Subdivision Surface modifiers

In the preceding screenshot, the base mesh geometry is showing with a level 1 of subdivision; in Edit Mode, it is still possible to see the low-level cage (that is, the real geometry of the mesh) as wireframe.

There are a couple of triangular faces (that, if possible, we should always try to avoid; quads faces work better for the sculpting) near the shoulders and on the feet, but we'll fix these automatically later, because before we start with the sculpting process, we will also apply the Subdivision Surface modifier.

To obtain a clean half-body mesh, we had to delete the first Mirror modifier and the vertices of the right half of the mesh; to do this, we had also added a middle edge loop. So, we obtained a perfect left-half mesh and therefore we assigned again a Mirror modifier to restore the missing half of the body.

Let's prepare the mesh and the view:

Well, in our case, everything went right with the Skin modifier, so there is no need for any big editing of the mesh! In effect, it was enough to delete the first Mirror modifier (that we actually used mostly for visual feedback) to get rid of all the overlapping faces and obtain a clean base mesh:

The "clean" mesh with new Mirror and Subdivision Surface modifiers

In the preceding screenshot, the base mesh geometry is showing with a level 1 of subdivision; in Edit Mode, it is still possible to see the low-level cage (that is, the real geometry of the mesh) as wireframe.

There are a couple of triangular faces (that, if possible, we should always try to avoid; quads faces work better for the sculpting) near the shoulders and on the feet, but we'll fix these automatically later, because before we start with the sculpting process, we will also apply the Subdivision Surface modifier.

To obtain a clean half-body mesh, we had to delete the first Mirror modifier and the vertices of the right half of the mesh; to do this, we had also added a middle edge loop. So, we obtained a perfect left-half mesh and therefore we assigned again a Mirror modifier to restore the missing half of the body.

Well, in our case, everything went right with the Skin modifier, so there is no need for any big editing of the mesh! In effect, it was enough to delete the first Mirror modifier (that we actually used mostly for visual feedback) to get rid of all the overlapping faces and obtain a clean base mesh:

The "clean" mesh with new Mirror and Subdivision Surface modifiers

In the preceding screenshot, the base mesh geometry is showing with a level 1 of subdivision; in Edit Mode, it is still possible to see the low-level cage (that is, the real geometry of the mesh) as wireframe.

There are a couple of triangular faces (that, if possible, we should always try to avoid; quads faces work better for the sculpting) near the shoulders and on the feet, but we'll fix these automatically later, because before we start with the sculpting process, we will also apply the Subdivision Surface modifier.

To obtain a clean half-body mesh, we had to delete the first Mirror modifier and the vertices of the right half of the mesh; to do this, we had also added a middle edge loop. So, we obtained a perfect left-half mesh and therefore we assigned again a Mirror modifier to restore the missing half of the body.

To obtain a clean half-body mesh, we had to delete the first Mirror modifier and the vertices of the right half of the mesh; to do this, we had also added a middle edge loop. So, we obtained a perfect left-half mesh and therefore we assigned again a Mirror modifier to restore the missing half of the body.

Open the Gidiosaurus_base_mesh.blend file and be sure to be out of Edit Mode, and therefore in Object Mode.

By checking the Only Render item in the Display subpanel under the Properties 3D window sidepanel, all the possible disturbing elements that cannot be rendered (such as the Grid Floor, Empties, Lamps, and so on) are hidden, in order to give a clean 3D viewport ready for sculpting.

Note that with this option enabled, sadly, the Image Empties we set in the previous chapter to work as templates for references are not visible—instead, the templates we had set as Background Images are perfectly visible in the 3 orthographic views.

Matcaps can in some cases slow the performance of your computer, depending on the hardware; in any case, Matcaps is a very useful feature, especially for sculpting, as you can see the mesh shape easily.

Changing the Lens angle from 35.000 to 60.000 makes the perspective view look more similar to the natural human field of view.

Open the Gidiosaurus_base_mesh.blend file and be sure to be out of Edit Mode, and therefore in Object Mode.

By checking the Only Render item in the Display subpanel under the Properties 3D window sidepanel, all the possible disturbing elements that cannot be rendered (such as the Grid Floor, Empties, Lamps, and so on) are hidden, in order to give a clean 3D viewport ready for sculpting.

Note that with this option enabled, sadly, the Image Empties we set in the previous chapter to work as templates for references are not visible—instead, the templates we had set as Background Images are perfectly visible in the 3 orthographic views.

Matcaps can in some cases slow the performance of your computer, depending on the hardware; in any case, Matcaps is a very useful feature, especially for sculpting, as you can see the mesh shape easily.

Changing the Lens angle from 35.000 to 60.000 makes the perspective view look more similar to the natural human field of view.

By checking the Only Render item in the Display subpanel under the Properties 3D window sidepanel, all the possible disturbing elements that cannot be rendered (such as the Grid Floor, Empties, Lamps, and so on) are hidden, in order to give a clean 3D viewport ready for sculpting.

Note that with this option enabled, sadly, the Image Empties we set in the previous chapter to work as templates for references are not visible—instead, the templates we had set as Background Images are perfectly visible in the 3 orthographic views.

Matcaps can in some cases slow the performance of your computer, depending on the hardware; in any case, Matcaps is a very useful feature, especially for sculpting, as you can see the mesh shape easily.

Changing the Lens angle from 35.000 to 60.000 makes the perspective view look more similar to the natural human field of view.

By checking the Only Render item in the Display subpanel under the Properties 3D window sidepanel, all the possible disturbing elements that cannot be rendered (such as the Grid Floor, Empties, Lamps, and so on) are hidden, in order to give a clean 3D viewport ready for sculpting.

Note that with this option enabled, sadly, the Image Empties we set in the previous chapter to work as templates for references are not visible—instead, the templates we had set as Background Images are perfectly visible in the 3 orthographic views.

Matcaps can in some cases slow the performance of your computer, depending on the hardware; in any case, Matcaps is a very useful feature, especially for sculpting, as you can see the mesh shape easily.

Changing the Lens angle from 35.000 to 60.000 makes the perspective view look more similar to the natural human field of view.

As usual, let's start from the last .blend file we saved: in this case, Gidiosaurus_Sculpt_base.blend.

Let's start with the Multiresolution modifier method:

Now, let's see the quick and easy preparation necessary to use the Dynamic topology feature for sculpting:

Again, here you can jump to the next recipe, Sculpting the character's base mesh; in any case, remember to save the file.

The Multiresolution modifier increasingly subdivides the mesh at each level by adding vertices; we have seen that from 2,240 starting vertices of the base mesh, we have reached 143,234 vertices at level 3, and clearly this allows for the sculpting of details and different shapes. The vertices added by the modifier are virtual, exactly as the vertices added by the Subdivision Surface modifier are; the difference is that the vertices added by the latter are not editable (unless you apply the modifier, but this would be counterproductive), while it's possible to edit (normally through the sculpting) the vertices at each level of subdivision of a Multiresolution modifier. Moreover, it's always possible to go back by lowering the levels of subdivision, and the sculpted details will be stored and shown only in the higher levels; this means that the Multiresolution method is a nondestructive one and we can, for example, rig the mesh at level 0 and render it at the highest/sculpted level.

The Dynamic topology setting is different from the Multiresolution modifier because it allows you to sculpt the mesh without the need to heavily subdivide it first, that is, the mesh gets subdivided on the fly only where needed, according to the workflow of the brushes and settings, resulting in a much lower vertex count for the final mesh in the end.

As you can see in the screenshots (and in the .blend files provided with this cookbook), starting to sculpt the character with the Multiresolution modifier or the Dynamic topology is quite different. In the end, the process of sculpting is basically the same, but in the first case, you have an already smoothed-looking mesh where you must add or carve features; in the second case, the low resolution base mesh doesn't change its raw look at all until a part gets sculpted and therefore subdivided and modified, that is, all the corners and edges must first be softened, in order to round an otherwise harsh shape.

As usual, let's start from the last .blend file we saved: in this case, Gidiosaurus_Sculpt_base.blend.

Let's start with the Multiresolution modifier method:

Now, let's see the quick and easy preparation necessary to use the Dynamic topology feature for sculpting:

Again, here you can jump to the next recipe, Sculpting the character's base mesh; in any case, remember to save the file.

The Multiresolution modifier increasingly subdivides the mesh at each level by adding vertices; we have seen that from 2,240 starting vertices of the base mesh, we have reached 143,234 vertices at level 3, and clearly this allows for the sculpting of details and different shapes. The vertices added by the modifier are virtual, exactly as the vertices added by the Subdivision Surface modifier are; the difference is that the vertices added by the latter are not editable (unless you apply the modifier, but this would be counterproductive), while it's possible to edit (normally through the sculpting) the vertices at each level of subdivision of a Multiresolution modifier. Moreover, it's always possible to go back by lowering the levels of subdivision, and the sculpted details will be stored and shown only in the higher levels; this means that the Multiresolution method is a nondestructive one and we can, for example, rig the mesh at level 0 and render it at the highest/sculpted level.

The Dynamic topology setting is different from the Multiresolution modifier because it allows you to sculpt the mesh without the need to heavily subdivide it first, that is, the mesh gets subdivided on the fly only where needed, according to the workflow of the brushes and settings, resulting in a much lower vertex count for the final mesh in the end.

As you can see in the screenshots (and in the .blend files provided with this cookbook), starting to sculpt the character with the Multiresolution modifier or the Dynamic topology is quite different. In the end, the process of sculpting is basically the same, but in the first case, you have an already smoothed-looking mesh where you must add or carve features; in the second case, the low resolution base mesh doesn't change its raw look at all until a part gets sculpted and therefore subdivided and modified, that is, all the corners and edges must first be softened, in order to round an otherwise harsh shape.

Let's start with the Multiresolution modifier method:

Now, let's see the quick and easy preparation necessary to use the Dynamic topology feature for sculpting:

Again, here you can jump to the next recipe, Sculpting the character's base mesh; in any case, remember to save the file.

The Multiresolution modifier increasingly subdivides the mesh at each level by adding vertices; we have seen that from 2,240 starting vertices of the base mesh, we have reached 143,234 vertices at level 3, and clearly this allows for the sculpting of details and different shapes. The vertices added by the modifier are virtual, exactly as the vertices added by the Subdivision Surface modifier are; the difference is that the vertices added by the latter are not editable (unless you apply the modifier, but this would be counterproductive), while it's possible to edit (normally through the sculpting) the vertices at each level of subdivision of a Multiresolution modifier. Moreover, it's always possible to go back by lowering the levels of subdivision, and the sculpted details will be stored and shown only in the higher levels; this means that the Multiresolution method is a nondestructive one and we can, for example, rig the mesh at level 0 and render it at the highest/sculpted level.

The Dynamic topology setting is different from the Multiresolution modifier because it allows you to sculpt the mesh without the need to heavily subdivide it first, that is, the mesh gets subdivided on the fly only where needed, according to the workflow of the brushes and settings, resulting in a much lower vertex count for the final mesh in the end.

As you can see in the screenshots (and in the .blend files provided with this cookbook), starting to sculpt the character with the Multiresolution modifier or the Dynamic topology is quite different. In the end, the process of sculpting is basically the same, but in the first case, you have an already smoothed-looking mesh where you must add or carve features; in the second case, the low resolution base mesh doesn't change its raw look at all until a part gets sculpted and therefore subdivided and modified, that is, all the corners and edges must first be softened, in order to round an otherwise harsh shape.

The Multiresolution modifier increasingly subdivides the mesh at each level by adding vertices; we have seen that from 2,240 starting vertices of the base mesh, we have reached 143,234 vertices at level 3, and clearly this allows for the sculpting of details and different shapes. The vertices added by the modifier are virtual, exactly as the vertices added by the Subdivision Surface modifier are; the difference is that the vertices added by the latter are not editable (unless you apply the modifier, but this would be counterproductive), while it's possible to edit (normally through the sculpting) the vertices at each level of subdivision of a Multiresolution modifier. Moreover, it's always possible to go back by lowering the levels of subdivision, and the sculpted details will be stored and shown only in the higher levels; this means that the Multiresolution method is a nondestructive one and we can, for example, rig the mesh at level 0 and render it at the highest/sculpted level.

The Dynamic topology setting is different from the Multiresolution modifier because it allows you to sculpt the mesh without the need to heavily subdivide it first, that is, the mesh gets subdivided on the fly only where needed, according to the workflow of the brushes and settings, resulting in a much lower vertex count for the final mesh in the end.

As you can see in the screenshots (and in the .blend files provided with this cookbook), starting to sculpt the character with the Multiresolution modifier or the Dynamic topology is quite different. In the end, the process of sculpting is basically the same, but in the first case, you have an already smoothed-looking mesh where you must add or carve features; in the second case, the low resolution base mesh doesn't change its raw look at all until a part gets sculpted and therefore subdivided and modified, that is, all the corners and edges must first be softened, in order to round an otherwise harsh shape.

In this recipe, we'll use the Dynamic topology method. If you haven't followed the instructions of the previous recipe, just follow the steps from 12 to 17; otherwise, just open the Gidiosaurus_Dynatopo.blend file that is provided.

As usual, it's a good habit to save the file with the proper name as the first thing; in this case, save it as Gidiosaurus_Dynatopo_Sculpt.blend.

If you are going to use a graphic tablet to sculpt, remember to enable the tablet pressure sensitivity for both size and strength; in any case, it is better to set the respective sliders to values lower than 100 percent; I usually set the size slider around 30/35 and the strength slider to 0.500, but this is subjective:

The tablet pressure sensitivity buttons for the size and the strength

Just a quick note: we don't actually need to go out of Sculpt Mode to save the file, it's possible to save it periodically (press Ctrl + S or Ctrl + W to save the file over itself, and Ctrl + Shift + S to save as) without needing to exit the sculpting session each time.

The completed sculpted mesh compared with the reference templates

So, here we are; the character's sculpting is basically done. We can work on it a lot more, tweaking the shapes further and adding details such as scales, wrinkles, and veins, but for this exercise's sake (and for this recipe), this is enough.

A nice aspect of the Dynamic topology feature is the possibility to actually join different objects into a single mesh; for example, with our Gidiosaurus, we can join the teeth and the talons to the sculpted base mesh and then keep on sculpting the resulting object as a whole.

Actually, there are two ways to do this: simply by joining the objects and by the Boolean modifier.

To join the objects in the usual way, we can do the following:

This is the way you join objects in Blender in general, and it can actually work quite well. There is only one problem: there will always be a visible seam between the different objects, and although in the case of teeth or talons this will not be a problem, in other cases it should be avoided. Let's say you are working on a separated head and later you want to join it to a body; in this case, you don't want a visible seam between the head and neck, obviously!

So, the option is to use the Boolean modifier:

Unlike the previous method, with Booleans, it will be possible to sculpt and smooth the joining of the different objects without leaving visible seams.

In this recipe, we'll use the Dynamic topology method. If you haven't followed the instructions of the previous recipe, just follow the steps from 12 to 17; otherwise, just open the Gidiosaurus_Dynatopo.blend file that is provided.

As usual, it's a good habit to save the file with the proper name as the first thing; in this case, save it as Gidiosaurus_Dynatopo_Sculpt.blend.

If you are going to use a graphic tablet to sculpt, remember to enable the tablet pressure sensitivity for both size and strength; in any case, it is better to set the respective sliders to values lower than 100 percent; I usually set the size slider around 30/35 and the strength slider to 0.500, but this is subjective:

The tablet pressure sensitivity buttons for the size and the strength

Just a quick note: we don't actually need to go out of Sculpt Mode to save the file, it's possible to save it periodically (press Ctrl + S or Ctrl + W to save the file over itself, and Ctrl + Shift + S to save as) without needing to exit the sculpting session each time.

The completed sculpted mesh compared with the reference templates

So, here we are; the character's sculpting is basically done. We can work on it a lot more, tweaking the shapes further and adding details such as scales, wrinkles, and veins, but for this exercise's sake (and for this recipe), this is enough.

A nice aspect of the Dynamic topology feature is the possibility to actually join different objects into a single mesh; for example, with our Gidiosaurus, we can join the teeth and the talons to the sculpted base mesh and then keep on sculpting the resulting object as a whole.

Actually, there are two ways to do this: simply by joining the objects and by the Boolean modifier.

To join the objects in the usual way, we can do the following:

This is the way you join objects in Blender in general, and it can actually work quite well. There is only one problem: there will always be a visible seam between the different objects, and although in the case of teeth or talons this will not be a problem, in other cases it should be avoided. Let's say you are working on a separated head and later you want to join it to a body; in this case, you don't want a visible seam between the head and neck, obviously!

So, the option is to use the Boolean modifier:

Unlike the previous method, with Booleans, it will be possible to sculpt and smooth the joining of the different objects without leaving visible seams.

As usual, it's a good habit to save the file with the proper name as the first thing; in this case, save it as Gidiosaurus_Dynatopo_Sculpt.blend.

If you are going to use a graphic tablet to sculpt, remember to enable the tablet pressure sensitivity for both size and strength; in any case, it is better to set the respective sliders to values lower than 100 percent; I usually set the size slider around 30/35 and the strength slider to 0.500, but this is subjective:

The tablet pressure sensitivity buttons for the size and the strength

Just a quick note: we don't actually need to go out of Sculpt Mode to save the file, it's possible to save it periodically (press Ctrl + S or Ctrl + W to save the file over itself, and Ctrl + Shift + S to save as) without needing to exit the sculpting session each time.

The completed sculpted mesh compared with the reference templates

So, here we are; the character's sculpting is basically done. We can work on it a lot more, tweaking the shapes further and adding details such as scales, wrinkles, and veins, but for this exercise's sake (and for this recipe), this is enough.

A nice aspect of the Dynamic topology feature is the possibility to actually join different objects into a single mesh; for example, with our Gidiosaurus, we can join the teeth and the talons to the sculpted base mesh and then keep on sculpting the resulting object as a whole.

Actually, there are two ways to do this: simply by joining the objects and by the Boolean modifier.

To join the objects in the usual way, we can do the following:

This is the way you join objects in Blender in general, and it can actually work quite well. There is only one problem: there will always be a visible seam between the different objects, and although in the case of teeth or talons this will not be a problem, in other cases it should be avoided. Let's say you are working on a separated head and later you want to join it to a body; in this case, you don't want a visible seam between the head and neck, obviously!

So, the option is to use the Boolean modifier:

Unlike the previous method, with Booleans, it will be possible to sculpt and smooth the joining of the different objects without leaving visible seams.

A nice aspect of the Dynamic topology feature is the possibility to actually join different objects into a single mesh; for example, with our Gidiosaurus, we can join the teeth and the talons to the sculpted base mesh and then keep on sculpting the resulting object as a whole.

Actually, there are two ways to do this: simply by joining the objects and by the Boolean modifier.

To join the objects in the usual way, we can do the following:

This is the way you join objects in Blender in general, and it can actually work quite well. There is only one problem: there will always be a visible seam between the different objects, and although in the case of teeth or talons this will not be a problem, in other cases it should be avoided. Let's say you are working on a separated head and later you want to join it to a body; in this case, you don't want a visible seam between the head and neck, obviously!

So, the option is to use the Boolean modifier:

Unlike the previous method, with Booleans, it will be possible to sculpt and smooth the joining of the different objects without leaving visible seams.

First, we are going to look for the LoopTools add-on, an incredibly useful script by Bartius Crouch that extends the Blender modeling capabilities (and that also has other functionalities, as we'll see in the next chapter about retopology); this add-on is provided with the official Blender release, but still must be enabled. To do this, follow these steps:

Now, we can start with the scene setup: