3D Animation 2 - Walk Cycles and Jump

Basic Walk

What worked on this animation was that I was able to get the basic key frames for the hips and legs and slowly adding my own style to the walk. Even though the walk looks good, its not a normal walk and looks more happy than basic. If i were to go back and improve on the animation, I would move the hip translate too high or too low. 

Personality Walk

When doing my personality walk, the animating portion of the walk was pretty simple with the flow of animating the hips and legs first then working my way up to the chest, then arms then head. With the feedback I got, I only had to work on the speed of my walk and how the arms are animated and overlapping.

Hips and Legs

Spine

Arms

What didn’t work was that I wanted to add a more paranoid personality to my character. From my previous reference attempts, it was more like she was cold than paranoid which was why I ended up doing a sneaky walk instead. I also wanted to add head turning to make her look around the scene but since this is a cycle, it would look weird on the character.

What I want to improve on is my overlapping on the arms and hands so it more fluent and also work more on the speed so it does not stop when the walk is cycled. I would also like to add head turn so it looks like she is looking around the scene while she is sneaking across the hallway.

Some problems I had rendering out the animation was that Bonnie had a few texturing issues such as pale skin having no eyes and the hair being black as well as when viewing and animation in the maya inter face and Arnold render, its normal however, when doing a render sequence, her eyes and teeth clip out while the head stays back. To fix the textures, I replace the blinn textures to aiStandardSurface and change the colour and style of Bonnie completely. When it came to the render, I check Bonnie’s geometry and notice that she had XY translate animation on the eyeball and mouth geos, so I deleted them.

Final Personality Walk

Jump

The jump animation was initially supposed to be a simple jump similar to my reference video, but I wanted to make it more than just a simple jump on a box so, I found a few reference videos of people parkouring off roofs. What worked in this animation was I was able to give the character weight when she steps or jumps off the ledge of the building.

I wanted to add a running start to the animation but ended up cutting it so the jump is more focused than the running start. Also, I wanted to add a roll to the animation however, I was not supposed to add turning or rolling in this assignment.

If i were to improve on this animation, I would add my ideas mentioned before and add a bit more overlapping and anticipation before she jumps off.  

Final Jump

3D Animation 2 - Advanced Bouncing Ball

Planning

Originally, I had chosen a Hand Ball and a Golf Ball for the animation. When given feedback in order to start my animation, I was told that both the handball and golf ball bounce very simpler and was suggested using a bowling ball with video reference.

While coming up with an idea for my animation, I simply wanted to just do a simple bouncing ball animation with the robot arm I made in first year as Rose suggested, however, when Rose said something along the lines of ‘… you can’t stop a light ball mid bouncing…’ and then I thought ‘well you can… you just need something to squash it.’ and then I had my story and asked Rose if I can animate it. I also wanted to add animation to both the robot arm as well as the camera so it shows both the impact and the reaction of the robot to show character.

Animating the Hand Ball

The issues I had when animating the handball, was how it stopped mid-air and basically made its own gravity. I was trying to make the impression that when the ball bounces mid-air, then next frame would be the bowling ball dropping down on top of it. It solve this issue, I made the arch in the graph for the hand ball while, on the bowling ball, made it drop while the hand ball was still making the arch. Since the audience does not see the ball inside, I thought this was not really an issue.

Animating the Bowling Ball

Using the video reference provided, I was able to get a good looking bowling ball animation, but, since the reference had a cloth as a surface for the bowling ball to drop, this made my animation not correct with the surface I’m working with. I only need to make it so the bowling ball rolled a little longer and stopped slower and smoother. Instead of simply making the bowling ball roll off the shot, I decided to challenge myself and see if I can recreate the bowling ball tilting, this was surprisingly the most frustrating as it kept looking incorrect.

All in all, although the animation looks a bit cartoony, I was happy with the final result of my animation. 

Modelling - Journal 3 Sculpting Project

For this assignment, we were required to sculpt a muscular structure of the chosen animal in our group. My group went with bears which are part of the Ursidae family. For the sub-species of bear, we picked the Polar Bear, American Black Bear and Panda Bear which all have different features and body structures to help them survive in their habitat even though they are part of the same family.

Research

While doing research on the polar bear, I discovered that polar bears have an elongated, narrow oval shaped skull and a long neck to aid them in the water while warming the air that they breathe. Also, their large, flat, oar-like front feet helps them swim in the water making them excellent swimmers. Polar bear rib-cages are long and wide because they have a thick layer of body fat which requires a bigger rib-cage. They also have a water-repellent coat that protects them from the cold air and water.

American Black Bears are the smallest compared to the polar bear and brown bear. They have a flat pointy skull as well as short, non-retractable claws that help them climb trees. The rib-cage is smaller and sharp-angled. Also, their spine is lower compared to the other bears due to the second layer of protection between their skin and the fur. Even with that extra layer, their body is designed to move extremely fast and climb trees.

Panda Bears or Giant Pandas are the smallest mammals in the group. It has a similar skull shape contrast to the brown bear and American black bear but the tip on the back of the skull is smaller and has a hook instead of a round point. The giant pandas have an extra digit on their paw that functions as a thumb to help them tear the bamboo off the ground. In addition, they have strong broad jaws that help crush the bamboo easily and their teeth are adapted to crush tough plants stems.  

All in all, each bear all have different features and body structures in order to help them survive in their environment. Polar Bears have a narrow oval shaped skull and a long neck to help them swim in the water as well as a thick water-repellent coat that protects them from the cold air and water. American Black Bears have non-retractable claws that help them climb trees and even though they have an extra layer of protection between their skin, their body is built to move extremely fast. Finally, Giant Pandas have an extra digit on their paw that functions as a thumb to help them tear bamboo and strong jaws and teeth that helps them crush the bamboo easily.

Process

While doing research on the polar bear, I originally thought of doing a stand up sculpt pose because it looks more dynamic and shows the full length of the polar bear but when I trying to draw out my plan, I did not know how to work with the bottom half of the reference; so, I ended up doing a walking pose of my sculpt since it more simple and dynamic. For the drawing process of my sculpt, I only wanted to draw shapes that are simple so I can easily model them in Z-Brush. I also, used the reference image as a guide to help me make a more accurate rough sculpt.

Once I had my finished drawing planned out, I started getting a Z-Sphere in Z-brush and started making the skeleton of my sculpt. I then used spheres and cylinders to add mass to the skeleton and define the bear geometry more using simple shapes. For the shoulders and pelvis, I used a sphere and the move tool to make it look more like a shoulder blade and pelvis so, at the point where I need to Dyna-mesh the shapes, I don’t have to brush the parts in. While doing this process, the goal for me was to use simple shapes to define the animal and then once it is merged, I don’t have to create lines that tell where body parts like the arms or legs are.

As I was about to create the muscles anatomy using the image references as a guide, I got some feedback. One of them was that we are modelling an animal with skin not the anatomy of the said animal which I did not know about. The other feedback I got was that the animal my group has selected does not show a lot of muscle due to the heavy coat bears have. The suggestion was to make a separate layer; one for the animal to have the coat and the other layer for the bear to have no coat, just the skin.

Once I’m happy with the sculpt, I now needed to make a separate sculpt of the bear with the coat. I duplicated the model so I don’t have to remodel the polar bear and rename the first model as ‘PolarBearNoCoat’ so I can tell the difference and not get confused. I then used the same technique that I used when making the mass of the bear which was using spheres and cylinders, but since I only want to make the polar bear’s body to have a heavy coat, I only used spheres. I then used the brush to add more mass to the arms, legs and neck as well as, removing all the muscle detail I did on my previous sculpt. I initially wanted to make fur on the model. So, I did an experiment to see how it would looked and it did not look anything like fur. It looked more like bark on a tree than fur so I scraped that idea and kept it simple.

The feedback I received was that my anatomy was very poor especially for both the muscle as well as the legs which I agree with. I was recommended to look at an anatomy book called ‘Science of Creature Design – Understanding Animal Anatomy’ by Terryl Whitlatch and while going through this book, I found a very good images of the muscle anatomy of a bear. So, instead of spending hours changing the sculpt, I thought I should start over.

So with the anatomy book with me, I started making a new skeleton for my sculpt. I also changed the scale of the image grid from 3 to 5 because, I work better on a bigger scale and also, I can add more detail to the muscle if I wanted. I then started adding a head to the skeleton. Instead of making a sphere head like I did on my previous attempt, I made an over shaped head because on one of my reference images with the polar bear’s head close up, its more wider and has an oval like shape. Furthermore, instead of sculpting the ears, I made two cylinders with symmetry on and then started moving the head to match my image reference. For the limbs, the goal was to make the muscle structure using simple shapes like cylinder and then filling and brushing over the shapes to create the actual muscle. This so I don’t have to spend my time making indent from a certain part of the anatomy such as making an indent from the arm to the shoulder. 

At this point I DynaMesh the model to 80 so the resolution isn’t too big and then used the clay build up brush to make the muscle creases. I also added indents for the eyes that I am going to clean up later. I then started following the anatomy book recommended for me so I can add muscle detail and change how the front limbs look like. I had to change how I made my front feet because before it looked too big and human like similar to my last attempt model. I did the same thing for the back legs and then smooth the muscles because the brief asked for a sculpt of a muscular animal structure but with the skin. After a little more cleaning up as well as fixing up the legs so the connecting tissue from the leg to the body is present and connected, I started adding more detail such as showing the ribcage, adding detail to the head and adding nails to the paws of the model.

Now was time to add the coat to the model. I first duplicated the model and used the quadfill brush to fill in more mass to the sculpt because the polar bear has a heavy coat. The problem with using this method is that the model starts to become to rectangular and no longer has its round shape so, I had to both smooth and round the model until it looks like my image reference and also, looks in proportion in all sides of the sculpt.

The challenging thing about sculpting was that the only reference I had on bear muscles were drawings. So, interpreting drawing into a 3D sculpted proved to be difficult especially the legs and head. But, with the book recommended to me, this helped out a lot when sculpting the limbs. Furthermore, the reference photo I’m using has the polar bear looking down, but it is hard to determine if the head is either looking straight or angled slightly. Throughout majority of the sculpting process was me looking at the reference and visualising how to looks in a 3D space and seeing if it looks right or not.

All in all, this has helped me improve on my understand of anatomy as well as determine the right proportion on a model and using my knowledge of anatomy to visualise and interpret a 2D image into a 3D object.

Modelling - Journal 2 Task 2 (King Spear Tip)

Attempt 1

When I first saw the model, I thought maybe if I just used a plane, I could not only extrude out the plane faces, I can also have control on the amount of edges I use. The shape looks accurate comparing it to the spear tip model and so I extruded out the plane to make it 3-Dimensional. I then deleted the faces that are bevelled inward because I thought I could extrude and change the edges position to make a nice slope.

When I was about to extrude though, I realised that while the front and side view looks the same, the model is actually a little round and not cubed. The top view is completely different from the spear tip top view. This would require me moving the vertices of the model and mess the model completely. What would be better is if I used a cylinder instead of a plane so I can keep the spear rounded on the top and bottom. Also, the plane does not provided depth which the spear tip model has. All in all, while this was able to maintain its shape from the front and the side, it does not from the top as well as not have any depth present.

Attempt 2

For this attempt, I thought I could try using the same method on my first attempt but make an outline of the side view. The first attempt was too flat however, if I started on the side and work my way inward, I can get the 3D aspect of the spear tip. I got a cylinder and rotated it and begin with the very tip of the model because the spear tip has a curved tip which looks like its curved over. I then deleted all but four faces just to get only the outline of the tip of the spear. This is so later on, I can just make the outline of the spear tip without worrying about making it 3-Dimensional unlike my last attempt, I only focused on getting the shape of the tip accurate. Once I got the outline of the spear tip, I started extruding inward. Furthermore, I extruded down the middle so I can later on use the multi-cut tool and make edges. The only issue with is technique is that the topology in my opinion does not look right. If I want to add an extra edge around an edge border, it will go around the whole model and make it too sharp compared to the spear tip.

I thought later on I can fix it later since I want to mainly focus on getting the shape of the model correct. So, I then used the bridge command to bridge two edges one by one while going down the model. As I was bridging, I realised there are too many edges on certain points which make the whole shape of the model. If I removed an edge loop, then the shape will be distorted. This attempt while it got the 3D aspect of the spear tip, the topology and methods I used were not good. While it was quick just to make an outline and extrude in, the topology needed for other elements such as the rims and the grooves were not presented well. In addition, on the side view, the spear does not have any bevelling or depth compared to the spear tip model. For my next attempt, I will use the cylinder again but not to make the outline. The reason why I am using a cylinder again is because a cylinder is used as the base of the spear tip.   

Attempt 3

So as previously stated from my last attempt, I will try using a cylinder to make the model since the spear tip has a cylinder like base down up. I started by using the measurement tools and implementing my measurements into the workspace. This is so I know the height and distance between certain parts of the model and where they transit from one part to the next in a nice slope. I got a cylinder with a sub-division of 12 and started extruding to match the image reference. Why I choose this technique is so I can make the starting point on when the shape changes and extrude out. There are 3 shapes I can already see, meaning 3 segments I need to extrude. I then delete all faces leaving a quarter of the model so I can focus on one side instead of trying to make all sides symmetrical. I then extrude the middle part of the component and tried to keep in mind of the amount of edges in the model so it’s does not end up causing pinching or small dents in the model. I start off with the middle because it the biggest piece out of the 3 and also, it’s the point where the spear tip starts to change.

I then repeat the same thing on the bottom curve of the spear tip since this technique worked well from the middle part of the spear tip. Also, this makes the topology of the model have a nice flow. I go back to the top of the component and extrude the faces again to have a nice topology on the model. The then extruded inward but only to make the groove around the top of the spear tip. I was considering to use Boolean just like the bishop model but I thought it was not necessary since this is not a difficult shape to model and while the groove has a little depth, the shape is just a triangle. I finally duplicated the model on the X scale and combine and merge vertices and then repeat for the z scale. Since I wanted that take a look at the finished mode, I need to duplicate all sides. The reason why I remove all by a quarter of the model is so I don’t have to remove the steps and waste time on the other 3 sides.

For the base, I used a cylinder with a radius of 0.885 because when I measured the base radius, it was 0.885 and started extruding down. I used the circle curve to set my measurements so I can C-snap on the curve similar to when I did the base for the bishop model.  

This attempt while the shape is nice, is bottom of the spear tip is has too many edges also, its pinching and is too sharp compared to the spear tip where it is more smoother and no bevelling. The outer rim from the middle and the bottom are too thin and sharp. There is no gradual slope and it’s too sharp when there should be some roundness. The top also has no gradual slope from the top of the spear tip to the middle. All in all, the general shape is ok, but the rim and the groove do not look accurate to the spear tip. What I should have done was break the model into 3 segments and using different polygons for each segment. Also, instead of working in the middle, I should have worked from the top down since the top requires more edges then the middle and the bottom because the groove is sharp around the outer edge.

Attempt 4

As I said on my previous attempt, I will separate the model into 4 segments because all look and flow differently. To start with the top, I choose a cube because on the side of the model, its square a little bit and the top looks like its sloping off, not rounded. I start extruding down up to the point where its start to slope inward to the middle. I wanted to focus on getting the shape first and worry about the groove later because, the groove while it requires more edges, the shape of the model is important. If the shape does not look like the spear tip, then I would need to change some edges on both the model as well as the groove which may cause the groove to distort.

For the middle part, I though instead of getting another polygon, I can just continue extruding down because if I did use a cylinder, I would have to bridge both component when I could have just extruded down and change the position of the vertices as I go. This however does not work on the bottom as I keep extruding because the spear tip on the bottom flows different to the middle. So, I had to get a cylinder and rotated it so the edges are angled and started extruding. I rotated the cylinder so the topology looks better and also, so I don’t have a lot of edges around the base of the spear tip. I then had to combine and bridge the two components and fix the topology so it accurate to the spear tip model.

Out of all 3 attempts, the last one is more accurate. The only issue is that the rims are too soft and not sharp enough. Also, on the side view of the model, the middle and bottom does not have a small bevel compared to the spear tip. Furthermore, the top and middle do not have the nice slope that slopes down and back up again in order to make the rim. Similar to my last attempt, while the shape looks accurate, the detail like the grooves and the rim need more work. I can still use the same method, but I will start from the bottom and work my way up. I can just use this model and make a few changes, but it would be much faster if I started on a new model instead of wasting time trying to fix a small part of the model.

In conclusion, while my first attempt was able to make the shape on the front and the side using just a plane, it does not show any depth or dimensional compared to my other 2 attempts and the spear tip. My second attempt on the other hand was able to keep the 3D aspect of the model, but living the other elements such as the rim, grooves and the topology of the model does not look good. My third attempt similar to my second attempt it kept the 3-Dimensional shape of the spear tip, but when it can to making the groove and the rims, it was either too soft or too sharp. There were no slopes presents on the rims and no bevelling between the middle and bottom of the model spear tip. My last attempt however, while it uses a different method, had a similar result to my second attempt. The rims this time where too soft and the slope that starts from the top and works its way down the middle isn’t presented very effectively. 

Modelling - Journal 1 Task 1 (Bishop Spear Tip)

Attempt 1

When I starting modelling the model, I thought I follow what Jacqui did with the class last year which was using a curve to make the dome of the model and then using Boolean to make the grooves on the model. As I was making the curve, revolving that curve and making it into a polygon, the model has a lot of N-gons revolved around it. While this was a fast way to make the shape, I do not have control on the amount of edges and sub-divisions necessary for the model. It would have been fast and more effective if I had created either a pipe or a cylinder, that way I can simply change the sub-division in the input channel box instead of using that time to delete and remove edges.

Attempt 2

My next attempt was to experiment with the polygons provided for me. So, I thought maybe a sphere would work since it has a smooth curve similar to the dome base of the model. I got a sphere, set its sub-division to 8 because I want to start off small instead of sticking to the sphere’s default setting which is 20 since 20 is too dense for this model and also, if I need more edges, I can add more using the multi-cut tool. If 8 sub-division does not work, I can get another sphere and change it to a higher value like 12 (trial and error). Next, I change the position of the edge loops around the sphere to match my measurement from class. I did not change the transform constraint from off to edge slide because I did not want to alter the shape. The sphere has a smooth curve which makes the shape have a nice slope.

For the grooves I wanted to try using a plane and change it shape to match the grooves on the model. This way I can just extrude to make it 3-Dimensional and maintain the position of edges and vertices and I also don’t have to terminate any edges. I then used Boolean so I combine shapes that would be difficult to model. This left a hole in the model so I had to fill the hole using Fill Hole and use the multi-cut to merge and create edges. I later made to make all the N-gons around the groove into quads using the multi-cut tool.

This attempt while simple was not very effective. The edges around the groove were making the model shape instead of curved. Also, the grooves from the top to the bottom is causing this weird pinching and is also too soft. Even if I added another edge loop around it, it would make the groove have too many edges. While the sphere made a nice curve around the dome bottom, it might be more better if I used a cylinder for the rest of the dome and maybe try using a torus as well for an experiment.   

Attempt 3

For this attempt I thought maybe I try using multiple polygons instead of using one polygon to make the whole model. I took some photos of the actual bishop/spear tip and divided it into 3 segments because I want to simplify the shapes from a 3D model to 2D shapes. I then determine the polygons I can already see so, for the bottom for example, I can see a cylinder because from the top-view it’s a circle and in the from view its straight and for the mid, I see the inside of a torus since while Jacqui demonstrated this model, she used a torus and deleted all the faces except the middle and reversing the normal. I wanted to initial use a cylinder, but I also wanted to experiment with use of a torus and see the different between the two. For sub-division, from my last attempt I used a sub-division of 8 which in my opinion was too low for this model. I don’t want to make the model too d

ense since the model is too small. I ended up starting with 16 because if I started with 12 I cannot add another edge without ruining the shape of the model.

I used a circle curves to make a structure and get my radius for each measurement. If I had just used measure tools to add my radius measurements, then I would not be able to get an understand of measurement. So, I used curves for my radius and the distance measure tool for the height. I start to build from the bottom up by creating a cylinder with a sub-division of 16 as previously mentioned, with a radius of 1.5 (my measurement from the bishop was 3cm in diameter) and removed the top and bottom faces leaving the sides because I don’t need them currently and I can easy Fill Hole and used the multi-cut tool to add the edges back but instead of triangles, I make then into quads. I also, scaled the height of the cylinder to match the measurements of my reference. I did not extrude because I want to focus on getting the shape of the model first then later on add edges or bevel the edge borders.

For the mid -section, I create a torus with a sub-division of 16 as mentioned before, to experiment if the torus works on this attempt and has a radius of 1.3. Once happy with the shape, I combined and bridge both component as well as bevelled the bridge edges to make the bump and bevel the edge borders from the cylinder. By combining and bridging the components together, I wanted to see the final shape of the base before starting with the dome. It ended up looking nicer than my last 2 attempts because from my first attempt, the curve does not give me much control and for my second attempt with the sphere, the base is too rounded and soft. So, if I were to do another attempt if this does not work, I will be using a cylinder again for the bottom and try to use another cylinder for the mid curve. Although the torus works nice, I just think it using too many edges just for that one space. 

For the top-section/dome, I used another cylinder and scaled the cylinder because from my previous attempts, while a curve can make a more accurate dome, I does not give me a choice on how my sub-divisions and edges I want. As for the cylinder, I have the option on the amount of sub-division I need and by using the circle curves, I can C-snap the edge loop to the curves. I forgot to delete the top and bottom faces of the cylinder because they were tri faces and I wanted them to be quads which I can do later. Since Boolean did not work as effective from my last attempt, I thought I try just deleting the faces that make the groove and extrude inward. This did not work as well. This made the grooves even softer than using a Boolean. The top and bottom are way too soft compared to the bishop and my last attempt.

From this attempt, the shape looks much better so I will be using a cylinder for the dome. Also, as stated before, for the base, I will be using cylinder, one for the bottom, and one for the mid curved. For the mid-section, I can get a cylinder and C-snap the cylinder to the curve and extrude in instead of getting a torus, deleting but the inside and reversing the normals.

Attempt 4

Since my last attempt worked better than my first two, I will be doing a similar method with a few changes.   I first divide the object reference but instead of 3, its 4 because the dome has a curved bottom which would work better with a curve object like a torus or a sphere instead of a cylinder. I will begin with the dome first instead of the bottom since it requires more edges compared to the rest. The sub-divisions will stay the same because it worked effective on my last attempt. If I add more sub-divisions, I would use more edges then necessary.

For the dome bottom, I create a torus with a radius of 1.55 and delete all but the outer middle ring faces. This way I can keep the curve and not have to change the position of the edges. I then I created a cylinder and scaled to the match measurements and finally combine the two components. Although, what I could have done was extruded the edges from the torus upward instead of using most that time bridging and combing the components. Furthermore, as mentioned before, from my previous attempt with the cylinder, it was accuracy with the bishop’s shape and thus I am using the cylinder again.

For the groove however, I had to use Boolean again since it helps make difficult shapes easier to model and also, from my last attempt, removing faces and extruding in do not work very well compared to using Boolean. So, I just a created a cylinder and scale the edge borders to match the bishop model grooves and placed the cylinder just outside the model. The reason I used a cylinder is because the bishop model has some thin oval grooves which is curved inward just a tiny bit and also, the top is smaller compared to both the middle and the bottom which look similar but are slightly different, which is why I had to scale the cylinder a bit. I had to however, delete a few edges and faces in order to fix the topology and edge flow of the object. I also had to terminate the edges by making an edge ring around the groove so it does not ruin the mesh and flow of the model.

For the bottom of the base, I used a cylinder as previously mentioned with a sub-division of 18 instead of 16 because when I check the amount of edges necessary to bridge, the poly count of edges said 18 instead of 16. For the mid-section, I wanted to try using a cylinder instead of a torus to see if there is a difference between using a cylinder and using a torus. Although the torus from my previous method works, I had to delete and reverse normals. For the cylinder however, I simply added an edge using edge ring and split and scale the edge to the curves’ measurement making this more simpler and more effective than my previous method. When comparing the torus and the cylinder, I honestly don’t see a difference. The torus has more edges but, if I deleted them and scaled it a bit more, it would look just like the cylinder. So, using a cylinder works just as much as using a torus.

After that, I combined and bridged the two objects and bevelled the bridges edges making a small indent. Since I was happy with how to turned out, I started adding edges and edge loops to sharpen edge borders and make small details like the small bevelled indent on the base. Once finished, I deleted all but the groove side because the bishop has a total of 6 grooves around the model and I need to duplicated 2 more models. If I duplicated 5 instead of 2, it would take longer to combine and merge 5 times. So, I duplicated 2 models and combined and merge vertices on the 3 components and repeated for the other side by duplicating and set the X-scale to -1.

Out of all 4 attempts I did, this attempt was more effective and more accurate to the bishop model compared to my other 3 attempts. I just need to make a few changes to the groove and make the topology have a better flow as well as fix the grooves to make it sharper on the sides and a tiny bit softer on the top and bottom.

 In conclusion, while my first attempt was about to make an accurate shape of the bishop model, I was not able to have control on the amount of sub-division I want in a model. In my second attempt, while using a sphere was able to make the slop of the dome smoother, it would be more better if I used more than one polygon for the model as well as maybe trying using a cylinder for the base and part of the dome. For my third attempt, using a cylinder for the base and the dome looked accurate to the model, deleting a few faces and extruding in does not work as well as using Boolean. Finally, my last attempt, it matches the shape however, I still need to edit a bit on the topology as well as fix the grooves’ size and depth.