Mon 24th Feb 20:
Initially, the character began as a box model with a few concept drawings done by a freelance amateur artist who was employed in exchange for exposure of their artistic talents. This reference art aided to model the character realistically in 3D dimension.
The video consists of reference tutorials on YouTube of Image Plane set up and low poly-resolution character modelling. These aided in improving my expertise in overall box modelling techniques.
As an introductory experience with character development through these techniques, the overall process took an exorbitant amount of time and effort, but the finished results have vastly encouraged a new approach to design.
The image planes were augmented to the cameras from each angle to produce an orthographic presentation to easily align the drawings.
The Multi-Cut Tool edged the box to morph the shape into a defined polymorphic design that enhanced the formation of the character into a more suited and layered-out establishment.
An edge loop split the box into two separate polygons to easily mould the character as if it was digitised clay, which simplified the overall construction process.
The video consists of reference tutorials on YouTube of Image Plane set up and low poly-resolution character modelling. These aided in improving my expertise in overall box modelling techniques.
As an introductory experience with character development through these techniques, the overall process took an exorbitant amount of time and effort, but the finished results have vastly encouraged a new approach to design.
The image planes were augmented to the cameras from each angle to produce an orthographic presentation to easily align the drawings.
The Multi-Cut Tool edged the box to morph the shape into a defined polymorphic design that enhanced the formation of the character into a more suited and layered-out establishment.
An edge loop split the box into two separate polygons to easily mould the character as if it was digitised clay, which simplified the overall construction process.
At the conclusion of stage 1, yellow and red lines were placed on the box (for moulding) which were for an unknown purpose at this stage. When starting stage 2, research was conducted through YouTube to discover what the purpose of the coloured lines was. The conclusion was that soft selection was enabled which insinuated that the feature had to be disabled before proceeding with the modelling.
Afterwards, the image planes had to change positioning as the back and front images needed inverting. During arrangement, a query arose concerning how to remove the image planes' back-faces as they were obscuring the model in the centre. Researching on YouTube led to back-face culling, but culling didn't function on the Image planes - only operating on polygon planes. A new solution for the task had to be implored, therefore all four image planes were displayed on two different reference layers so they couldn't be selectable. These layers were made transparent when those reference images were not being utilised.
Modelling continued with the devil having the vertices & edges moved around. Also, edges and vertices were incorporated using the multi-cut edge loop tool to line up the box with the reference drawing. Next, move and scale the right side image plane to fit the model and the back image plan.
Afterwards, the image planes had to change positioning as the back and front images needed inverting. During arrangement, a query arose concerning how to remove the image planes' back-faces as they were obscuring the model in the centre. Researching on YouTube led to back-face culling, but culling didn't function on the Image planes - only operating on polygon planes. A new solution for the task had to be implored, therefore all four image planes were displayed on two different reference layers so they couldn't be selectable. These layers were made transparent when those reference images were not being utilised.
Modelling continued with the devil having the vertices & edges moved around. Also, edges and vertices were incorporated using the multi-cut edge loop tool to line up the box with the reference drawing. Next, move and scale the right side image plane to fit the model and the back image plan.
At stage 2 conclusion, modelling the outline of the body was complete. When stage 3 began, the process started with the reshaping of the vertices to line up with the reference drawings. After the selected polygon faces - on the model's sides - were extruded, this created the legs of the model. The legs' extrusion was attempted at the same time, but they would not extrude in the same direction.
To retopologise in the future, the 'Maya: Retopology for Animation' video on LinkedIn would be the best instructional example video to show how to make a model with fewer polygons with only four edges and vertices using the quad draw tool.
When extruding the faces out for the legs, they should have been extruded in before being pulled down to create a slimmer leg. It took extensive time to make the shape for just one leg precise.
A few problems with the polygons of the character arose as ridges in the object were created. This caused it to appear offset so adjustments needed to be made by pulling them out until the ridges were smoothed out, although the character still requires minute adjustments to completely tone the body's overall form.
Another character will need to be modelled at a separate time to build the modelling skills that are already present within the first model. Whilst LinkedIn at University is still accessible, a video on modelling will need to be found for future reference.
Whilst modelling the Right leg, a thought arose about how both legs could be made to look identical so research resumed on how to mirror half of the object through the use of reference on YouTube. A video called '6 ways to Mirror Polygons in Maya' was useful, as this video would help to move vertices on one-half corresponding to the other.
Sculpting half the model to make the legs, arms, and head would have been enough to construct a uniform character on both sides however, the drawings of the model weren't symmetrical. Modelling the character again, only half of the model would have been required as mirroring the polygons would allow for it to be symmetrical and then the different side could be reshaped to look realistic. A video called 'Maya 2014 tutorial: How to mirror objects while box modelling' was discovered, that could allow for the object to be mirrored from only half of the design's structure.
Both legs had to be similarly proportioned by selecting the polygons on the leg to be duplicated so that both were evenly matched on the character's frame. When duplicating, the leg wasn't the only structure copied and instead the whole body was duplicated. This caused an extra task of having to remove the unnecessary polygons that weren't being used. As the copy was set in place, it didn't match the intended design due to the lack of knowledge on how to combine the two objects together rather than only parenting them together.
The duplicate leg was removed and the polygon faces were selected to extrude the leg out. This would connect it to the body - if the knowledge was known at the time of processing. Pressing combine would have been all it took to join the two models together.
To retopologise in the future, the 'Maya: Retopology for Animation' video on LinkedIn would be the best instructional example video to show how to make a model with fewer polygons with only four edges and vertices using the quad draw tool.
When extruding the faces out for the legs, they should have been extruded in before being pulled down to create a slimmer leg. It took extensive time to make the shape for just one leg precise.
A few problems with the polygons of the character arose as ridges in the object were created. This caused it to appear offset so adjustments needed to be made by pulling them out until the ridges were smoothed out, although the character still requires minute adjustments to completely tone the body's overall form.
Another character will need to be modelled at a separate time to build the modelling skills that are already present within the first model. Whilst LinkedIn at University is still accessible, a video on modelling will need to be found for future reference.
Whilst modelling the Right leg, a thought arose about how both legs could be made to look identical so research resumed on how to mirror half of the object through the use of reference on YouTube. A video called '6 ways to Mirror Polygons in Maya' was useful, as this video would help to move vertices on one-half corresponding to the other.
Sculpting half the model to make the legs, arms, and head would have been enough to construct a uniform character on both sides however, the drawings of the model weren't symmetrical. Modelling the character again, only half of the model would have been required as mirroring the polygons would allow for it to be symmetrical and then the different side could be reshaped to look realistic. A video called 'Maya 2014 tutorial: How to mirror objects while box modelling' was discovered, that could allow for the object to be mirrored from only half of the design's structure.
Both legs had to be similarly proportioned by selecting the polygons on the leg to be duplicated so that both were evenly matched on the character's frame. When duplicating, the leg wasn't the only structure copied and instead the whole body was duplicated. This caused an extra task of having to remove the unnecessary polygons that weren't being used. As the copy was set in place, it didn't match the intended design due to the lack of knowledge on how to combine the two objects together rather than only parenting them together.
The duplicate leg was removed and the polygon faces were selected to extrude the leg out. This would connect it to the body - if the knowledge was known at the time of processing. Pressing combine would have been all it took to join the two models together.
After modelling the body and legs, a change of task was needed to keep motivation and so the environment was the next project that was assessed after the character. Modelling the environment was done by creating burning trees that would layer the background and add a point of focus in the dark colours of the Hell environment. First, research into creating fire allowed for a video of a 3D particle bonfire helped to realise the aspects of the graphical fire using the Maya software renderer.
A model of a tree for the environment was needed so a tutorial for tree modelling was found so that I could make the tree from scratch, although the video presented the realisation that the length to complete a satisfactory model would be too long. As a beginner to modelling, making a tree from scratch wouldn't have been achievable if the tree was to look professional and complete.
Going back to find the video that I previously used for the creative collaboration module was better to fully showcase the tree without making it look unrealistic and tacky. Previously creating the intro animation for the 'In the Woods' animation, the video showed how to draw out and create a tree using the generating tree brush which swirled out to create the tree model.
When creating the tree, the selection wasn't appealing. However, the 'In the Woods' tree from the creative collaboration Maya file had the perfect form, so a new file was saved called "Fire Tree". The three extra, unnecessary trees from the file were removed and the file was ready.
The same soft selection problem occurred on the tree's stump, so a recap over how to disable it was conducted. Later the tree's leaves had to be removed due to the fact that a burning tree would have no leaves, they would already have been burned (an example being the show Once Upon a Time when Maleficent made a never-ending burning tree).
After, the properties from the 3D bonfire video to create a realistic fire using a particle emitter were added to the tree. First rendering the fiery tree in the Arnold renderer, but this caused the fire to appear a grey colour which made it look like smoke rather than fire. Different properties were tested to get the grey fire to look red & orange until the renderer had to be changed to 'Maya Software' so that the colour wouldn't be changed to grey.
A model of a tree for the environment was needed so a tutorial for tree modelling was found so that I could make the tree from scratch, although the video presented the realisation that the length to complete a satisfactory model would be too long. As a beginner to modelling, making a tree from scratch wouldn't have been achievable if the tree was to look professional and complete.
Going back to find the video that I previously used for the creative collaboration module was better to fully showcase the tree without making it look unrealistic and tacky. Previously creating the intro animation for the 'In the Woods' animation, the video showed how to draw out and create a tree using the generating tree brush which swirled out to create the tree model.
When creating the tree, the selection wasn't appealing. However, the 'In the Woods' tree from the creative collaboration Maya file had the perfect form, so a new file was saved called "Fire Tree". The three extra, unnecessary trees from the file were removed and the file was ready.
The same soft selection problem occurred on the tree's stump, so a recap over how to disable it was conducted. Later the tree's leaves had to be removed due to the fact that a burning tree would have no leaves, they would already have been burned (an example being the show Once Upon a Time when Maleficent made a never-ending burning tree).
After, the properties from the 3D bonfire video to create a realistic fire using a particle emitter were added to the tree. First rendering the fiery tree in the Arnold renderer, but this caused the fire to appear a grey colour which made it look like smoke rather than fire. Different properties were tested to get the grey fire to look red & orange until the renderer had to be changed to 'Maya Software' so that the colour wouldn't be changed to grey.
