Pre-production

Documentation Portfolio

Pre-Production:

Week 1: Come up with an initial idea for your project

I decided to spent this week considering an idea I would like to pursue for my project. I have a strong interest in visuals and games, and wanted to combine these two interests into a project, so that I had a very strong passion and creative interest for the project. 

Games will typically propose a 'boss fight' to challenge the player and provide interesting mechanics, but the scene and story attached is very lacking and linear, as the fight must involve mechanics the player can realistically use. Some game trailers will adapt scenes such as these to give it much more character and interest for the viewer, when in reality the fight in the game might be much less interesting. 

As a result, I would like to take a typical 'boss fight' scene from a game, and recreate this scene in my own style and interpretation. The scene would reflect a scene you might encounter in a game, but shall be a unique piece of my own work, and heavily exaggerated and much more engaging. I would like to create a few characters and a boss character. For the boss character, I would like to create a dragon, as this is a challenging fantasy-based creature which is interesting for me personally, and would offer the inclusion of interesting VFX work such as the fire from the dragon. 

I would like for the characters to be two opposing groups, one group after the other. The events would lead both groups into the lair of this dragon, where they must put aside their differences and defeat this dragon together, using their strengths to defeat it. This would be representative of boss fights in games where you have different class types to defeat bosses effectively in groups. The end result would be characters who survive the fight, but not all characters will survive, adding the element of risk to the characters in the story. The initial events of the story could help allow the viewers to connect with the characters to make their deaths and potential survival more meaningful too. 

Week 2: Evaluating and refining initial ideas

Whilst I am very attached to my initial idea, I realise that this may not be very feasible considering my limited skills and time frame to produce the project. Additionally, past assignments have been flawed as I have tried to achieve a very high outcome, but the scope has been too great of the reality of the assignments, causing myself lots of 'crunch' and having to cut a few minor elements or time to refine the result sometimes. Because of these doubts, I decided to shelf the initial idea I had, and consider a new one more feasible. 

During our lectures, we were shown some examples of projects created in the years before. An idea which caught my eye was a render of a Japanese samurai warrior, a highly detailed model and environment to compliment each other, and was very representative of reality. Creating a project which looks realistic is something I am very passionate about, and this project gave me an idea of what I could produce myself. It also helped to have a clearer scope of what we needed to produce in the time frame, and the expectations of the project. 

I would like to propose the idea of creating a very realistic 3D medieval character. The purpose would be to design something which looks very realistic, which would require elements such as the reflections of the metal on the armour and the lighting to be carefully considered. The piece would include just one scene of the character, holding a weapon and standing/sitting in a suitable environment. The main goal would be to show off the visual result/design of the character as best as possible. The environment could enhance this, such as reflections on the armour from things like candle lights or the sky perhaps. 

In regards to the armour, I would like to consider the fantasy armour and recreational armour used in movies or other media, and how they compare to the armour used in reality/history. Fantasy armour is usually very exaggerated and includes ominous and/or bold features to make the armour more visually striking, but it usually serves no functional purpose. This is an area I feel strongly about, as it can hinder my immersion in a piece when I can see something has been included or removed which takes away from its functionality. As a result, I want to compare how medieval armour is represented in different pieces, comparing fantasy armours to realistic armours. What I determine to work best visually and functionally will be the direction of my design. 

During this week we were tasked to create mood boards, but my mood boards are unfinished, so I will detail them in the following week. 

Week 3: Mood boards

I spent time browsing images on the internet of images of armour, weaponry and suitable environments to create 3 separate mood boards to get a feel for the project. 

Armour

Weaponary

Environments

I tried to vary the mood boards as much as possible. Looking at the armour mood board, I have tried to include both realistic and fantasy styles of armour. I also did this for the weaponry, showing off different types of weapon such as maces and axes. The environment is the least varied in hindsight, as I am currently unsure what type of environment to include. I currently have the idea of a medieval knight sitting outside of a castle at night, but this may change based on my research. 

Week 4: Initial sketches and designs

After creating my mood boards, I feel that I have gained a good sense of direction for my project. I decided to create some initial sketches to consider what my project could look like, in regards to the character and environment. 

Sketch 1: Character

Here is an initial design of the medieval character I am thinking to create. I sketched this whilst looking at my mood board for inspiration of the sword and armour, especially for knowing different parts of the armour. I am relatively happy with the design, though my main complaint are the knee plates as these aren't drawn well due to my lack of ability to draw, and knowledge on how they should be implemented. The armour is quite plated and have lots of layers, something which helps to give it more depth and realism, and is something I would like to maintain for future designs. Also, I created the pose without much thought, but I am very happy with the pose, with the character looking directly at the audience in a more 'battle-ready' pose, with the sword being brandished, and would like to look at similar poses like this for future. 

Sketch 2: Environment #1

My initial idea for the environment included the character sitting by a castle. However, I felt that the character might be better placed inside a medieval courtyard or entrance instead, as a close shot of the character would not include much of the castle in the background unless it was in the distance, and would be wasted design. The fact the character is inside the castle also signifies importance to the viewer. The design of the inner castle should be more detailed, but I did not include this as of yet as I was not confident on the environment choice, and decided to explore some more options. It is worth noting that the design of the armour in these environments are very rough and not as detailed as I am inconsistent with drawing and wanted to solely focus on the environment design in these sketches. 

Sketch 3: Environment #2

For this sketch, I placed the character inside a blacksmiths or forge perhaps, I am unfamiliar with the correct term as of now. The idea would be the character could be sharpening his blade on a whetstone, creating sparks which could give nice reflections on the armour. The furnace in the background could also provide some nice ambient lighting behind the character. I am not keen on the camera angle in the sketch, but this is mostly limited by my drawing ability, and future implementation would feature a closer shot with the furnace behind the character in the shot. The environment is nice because it shows direct links to the medieval genre and time period, but admittedly the environment is more dull and uninteresting to the viewer. I decided to consider something a bit more active or involving. 

Sketch 4: Environment #3

For my final sketch, I considered the ruins of a battlefield of a recently fought battle. Whilst the creation of this would be much more work, I immediately resonated with the design as it gave me lots of ideas. The main idea being the story behind the environment, how the battle was fought, how the character survived, etc. I liked this as it helped to create more meaning for the character besides just being a suitable environment. Also, I personally found the environment to be much more visually interesting, offering lots of visual elements such as smoke or fire to compliment the design. Moreover, I could include damage in the character's armour as well, such as arrows, dents or large scratches. Or perhaps blood stains. All in all, there are lots of elements to consider, but this environment design is the one I preferred the most out of the 3 I had sketched down. 

Photobashing: 

I decided to have a go with photobashing, a technique we briefly covered last year. I decided it might be helpful to better visualise my ideas, both for myself and for my peers during the presentation session next week. 

Initial character pose and proportions

Initial environment layout

I created a very basic sketch of the environment and character to outline the design. I used the environment design from my first sketch despite preferring the third design, due to the simplicity of the environment to help me complete the photobashing attempt and not get stuck on more complex areas.

Photobashing attempt - Version 1

Here is my first attempt of photobashing. I am under the impression that the technique would typically use a combination of photos and digital drawing/painting, but due to my lack of drawing skills as well as limited experience of the photobashing technique, I decided to focus on merging and blending imagery instead. 

I was able to find lots of suitable pieces of images online, to fill in the outlines I drew earlier. I have tried to blend the images together as best as possible, but you can still notice the inconsistencies especially in the foreground, where textures do not match. I tried to match these as best as possible using techniques such as curves, levels and colour overlaying with different filters to match them as close as possible. Lots of the scene are a collection of individual images merged together, which is especially true for the medieval character, who is made up of lots of separate images of armour layered over the top of each other. 

I am not happy with the sword of the character, as I struggled to get a sword and hand which would match the perspective of what I was looking for. I also disliked the foreground scene, the grass and ground textures were not very nice visually, nor retained a high quality. Additionally, the sky was difficult for me to get right, as images of cloudy skies did not look right to me. I used a flat colour instead, but this also does not look right as skies aren't a flat colour. I tried to address some of these complaints in the next image. 

Photobashing attempt - Version 2

I decided to modify the image to help refine the result. The most immediate result is the yellowish tint across the image, which looks better in my opinion, as it gives connotations of old and historical settings lit with natural light such as candles, despite not being present. I also added a blur to the foreground, and increased it in size, removing the sky from view which had previously bothered me. The ground of the scene still looks poor, but the blur helps to obscure the lack of quality in this area. Finally, I reduced the vignette effect as it has been too strong before. 

Overall, whilst the result is lacking quality and care in some areas, I believe this was a good photobashing attempt, especially considering it was essentially my first attempt to use the technique. Despite the inconsistencies, it helps to visualise the direction of my project better, especially considering that my drawing skills are limited and not necessarily representative of my full idea, so it might be more ideal to present this next week to help my peers easily recognise the direction of my idea for my final major project. 

On a side note, I included a shield for the character to wield in this photobashing attempt, but personally I am not keen on this addition as it blocks a significant portion of the armour from view. Based on how this looks, I am currently thinking for my character to not wield a shield, and just a weapon instead. 

Week 5: Proposal & Feedback

To summarise my initial ideas and research, I was tasked with creating a proposal for my project. This took the form of a written document, with the purpose of proposing my idea to get feedback. To present this to my peers and lecturers, I simplified it and added images to help others to absorb the information better. I also showed my sketches, photobashing attempt and mood boards to show my direction for the project. 

Here is a link to my proposal.

Here is a link to my presentation. 

After presenting, I received some useful feedback. Here are some of the comments I received:  

During and after my presentation, I received some feedback about what I had proposed so far. Overall, the feedback I received was positive, which suggests that I did a good job on both presenting and preparing my proposal into a way people could easily understand it. Though personally I was nervous during the presentation, so this may have hindered the clarity of how I presented. 

A point raised was that my character could use more context and background. From my proposal, I have only outlined that I want to create a medieval character, and that it should have a well-defined design based on my research. Though, there is no further detail here, and context could help to clarify exactly what I am going to design. For instance, is my character going to be from the 5th century or the 15th century, as the armour designs and techniques used to develop them were significantly different between the time periods. For the design of my armour to be the main focus, this is an area I should have paid more attention to. 

It was suggested that I should consider the technical aspects of my project, notably the reflections and shininess of the armour. This is something I have considered, but not fully outlined as of yet, so this is something to outline clearly. It was also suggested to consider the pose of the character, and how this could be done to best show off the armour. This is an excellent suggestion - one I will take on board - though since I am considering adding subtle animations of the armour at the moment, this may not be as significant. 

An interesting suggestion was to tell "an untold story or anecdote" in my project. Whilst I personally disagree with the suggestion in its entirety, it gave me an idea when reading "an untold story". I would like to tell a story which is not told to the viewer, one that is viewed based on the visuals and the elements in the scene. For instance, my character could be the lone survivor of a battlefield, which could suggest the story of a battle taking place. This could suggest elements of the character, such as being skilled and courageous in order to be a survivor. I'd like to explore this idea of an 'untold story' embedded within my piece, told just visually. 

One of the most useful pieces of feedback I received was how I would be presenting the project to the viewer, as this was something I had completely overlooked. One of the suggestions was to present my project as a hologram, but this is something I ultimately dismissed because I wanted more control over the scene, with the ability to add camera motion or multiple shots, and include animations to the character if I decide this will be suitable for my project. Despite the suggestions, I would like to consider presenting this as a video format, which could either be set up and shown at an installation, or viewed and shared online (or both, if this is acceptable). Although, due to the uncertainties of COVID-19, my current aim is to have my project shared online. This is something which would require further discussion, but I would prefer this myself as I would like my project to be something which can be viewed on more occasions than one, and to be more accessible. 

Overall, presenting the proposal for my project yielded in some crucial feedback, helping to steer my project in the right direction, and pointing out elements which I had either yet to consider, or plan out in more detail. I have added some clarification to my proposal due to this feedback, mainly how I plan to present my project. From here, I will start thinking about some more detail of my project, and the direction I wish to take it. 

Week 6 & 7: Success Criteria and Experimentation

- Defining my Success Criteria + Feedback - Week 6

For Week 6, our lecture had us consider some criteria for success for our project. We were asked to think of 3-5, then received feedback on our current progress and aim for the project. I recorded most of the feedback I received, and have shared it below. 

Here is a condensed video of the majority of feedback I received on my current progress. The focus was to help develop my criteria into something to be used in my dissertation. 

One of the main points brought up was that my project seems to have a focus on 'Function over form', which is true. This could be a focus point of my dissertation, looking at how armours and swords are functional and why. 

Contextualising my research was mentioned during this feedback session. It was suggested to use other media which may not necessarily adhere to 'Function over form' to better contextualise my discussion. A game called 'Soul Calibur VI' was mentioned, a fantasy game with lots of exaggerated and fantasy designs and animations. Whilst not what I am aiming for in my project, comparisons to media such as this could benefit my research and clarify what I am trying to achieve. The mention to this sort of media could be brief or detailed, but with the aim of contextualising my research better. 

It was suggested that I could look at defining what is realism and fantasy in my criteria, as two separate criteria. This could help to focus my research, and could be used as a foundation for my third criteria. 

Whilst I was very grateful for this feedback, I personally still felt a bit overwhelmed; The task of developing success criteria for my project was an area which seemed daunting and confusing, as I am unused to writing as critically and deeply as the assignment is asking of me. I decided to try to take the feedback on board, but I will look for clarify in the following week in which there will be another one to one session. 

- Refining my Success Criteria + Feedback - Week 7

Here is a condensed video of the feedback I received this week. As I was still a bit confused from last week, I decided to get some more clarity on the success criteria I was to develop, what their purpose was and how I should go about it. 

It was suggested that over the next week, I should actually find a piece of armour that I liked or wanted to use as a base, and 'bring this in' for the next week. I could use this as the foundation of how my armour should look, making it easier to solidify what my armour will look like in my final major project. This was an excellent suggestion, because my current method of thinking was rather limiting. I thought that if I would first define how fantasy and realism compared, I could create something completely unique and effective. However, it would make creating the armour a difficult task, to actually consider every individual element which will vary across different armours. Having a strong starting point should make this process much easier. I will look into something suitable for next week. 

- Experimentation - Weeks 6 & 7

We were also tasked with creating something which could directly benefit our project down the line. This was to be completed across these two weeks. My goal was to first experiment with the nCloth physics we had learned a few weeks prior to test their possible implementation, if my character might wear a cape or some other clothing. I then decided that I would like to look into how you would make a metallic material, so that it would offer reflections and look shiny like real metal. I decided to do this in Maya. 

I started by creating a very basic chest plate, simply for the purpose of holding a cape above the shoulders. However, to hold the cape in place, I added some basic cape holders to keep it in place, both logically and physically inside Maya. I then added the nCloth physics to the plane designated as the cape, and the physics worked. However, the cape holders were not enough to keep it in place properly, so I had to use cloth constraints to keep the cape in place, which worked in the end. 

I then added some simple colours and materials to the models. I used flat colour for the cape, but shiny materials for the rest. However, it dawned on me that a reflective material needed something to reflect (obviously!), so I decided to use a HDRI. A HDRI is known as  'High-dynamic-range imaging', it is essentially an image of the environment which has been captured to wrap into a spherical shape. The shape is then used to calculate the lighting based on the lighting in the image, which offers very realistic lighting depending on the HDRI image used. I was able to source a free image online, which would work to provide reflections. I then added an Arnold 'AiStandardSurface' material which would offer a 'Metalness' option to reflect the scenery. However, I wanted to test this on a more complex surface, so I decided to have a go and make a very basic medieval-themed helmet in Z-brush. 

Here is the model I created in Z-brush. I opted to use Z-brush because it is the tool I feel most comfortable and able to produce models within. Using some loose inspiration from my mood board, this is the design I created, using separate layers for pieces of the helmet to give it more depth and realism. However, I struggled to get the helmet to have flat areas of geometry, which is especially noticeable about the visor where it looks very messy. Also, the helmet is too thick, something I did not think about until it was mentioned by one of my lecturers during my feedback session. Additionally, the details are low quality. These are things to keep in mind for future production. However, the result was good enough for my purposes, so I did not spend more time refining it. I did play around with some of the materials in Z-brush to get a feel for what it might look like when the correct material is applied. 

Back in my Maya project, I added the AiStandardSurface Arnold material to the chest, and upped the metalness value to 100%. Though it showed the area as black, so I needed to view the viewport via Arnold rendering, which gave the shiny result. I imported my helmet made in Z-brush and applied the same material to it. The proportions look a bit off, but I did not fix this as I wanted to look at the metallic options here. 

I then added roughness value to the material, which adjusts how reflective the material is. Since armour is usually not fully 'chromed' and purely reflective, I added some roughness to the material to make it look a bit more realistic. Though, looking at the result, the roughness value might have been a little too high. Despite its flaws, I was able to successfully achieve my goal of creating a metallic material, which reflected the environment (the HDRI image). My biggest problem however is that the result was quite noisy and grainy, especially noticeable in the background, so this is something I need to look at fixing if I am to render my project using Maya. 

I decided to take this experimenting further, and look at how other software might render the result, to see what might work best. I started with Unreal Engine 4, and imported my models here. Though the material was not compatible, and didn't work, so I had to create a custom material in Unreal Engine 4. Additionally, the HDRI image did not match, as I was unable to figure out how to change it here without possibly breaking the project file. As a result, the renders are not fully comparable as they don't have the same factors involved. Though, it was interesting to play around with the lighting and see what it looked like here. One thing I noticed was that Unreal did not give a noisy result like Maya did, so that was an immediate upside. However, the shadows of the model were not very consistent, as the inside of the helmet did not show dark as it should, especially noticeable in the first image. This could be a problem with how I set up the project file though. 

Finally, I decided to look at Blender to see how the program would render my model. Again, I imported the models and materials from Maya as I had done before for Unreal. Like Unreal, Blender couldn't recognise the Arnold material used in Maya, so I had to create another one from scratch. Blender comes with a default HDRI option when rendering, which was convenient for me since I don't know how to use Blender very well at the current time. Though - like Unreal - I could not figure out where to use the HDRI I had used in Maya, so the results were not very comparable. Like Maya, the result was good, but was quite noisy. I attempted to play with some of the denoise options Blender offers, but unfortunately the program crashed and I lost my work, as I didn't expect to need to save this project on this occasion, a mistake on my part. Hopefully this option could help clear up some of the noise in the result. 

From the 3 programs I looked at, all 3 offered good quality in the renders. Maya and Blender both showed lots of quality, but there was lots of noise and grain in the results, something I have yet to understand how best to reduce/remove it, and hope to in future. Unreal Engine showed promise, but the shadows were not very realistic, and did not appear properly inside the helmet. I have yet to understand why, and whether this is a fault in me or Unreal, but hopefully this can be fixed. I am still unsure which program to use when rendering, especially with the flaws I encountered. Though at the current time, due to my bigger knowledge and comfort with Maya, I am opting to use Maya as of now. This may be subject to change. 

I feel confident that I was able to cover my area of focus to a significant degree. In addition, I found time to explore some of the rendering possibilities and compare the results, benefits and disadvantages of each that I had used. Though there is still lots to improve on, most notably the design of the helmet. I was not keen on the bumpy and rough look, despite it not being a focus of my experimentation. I would like to figure out how to create a smooth and flat finish in Z-brush, as this will be especially useful in my production. I would also like to improve the metallic look of the armour, showing imperfections and a more realistic look to it.  In regards to my experimentation of nCloth in Maya, this could be useful if I decide to introduce any fabrics to my character, such as a cloak or possibly chain mail. This will require further discussion. 

I presented this work on the 7th week when it was due. The feedback was positive, commenting on how the work does appear to be 2 weeks worth of work, and that my areas of focus were relevant and interesting. One comment made was on the thickness of the helmet I had produced, in that it was unrealistic and should be much thinner. The odd proportions were also mentioned. Whilst I was aware of these flaws due to my focus being on the materials used, I will keep these points in mind for future production. 

Week 8: Beginning my Dissertation

During this week, I was unfortunately very busy with other assignments, so little work was done this week. In our lecture, we discussed the idea of our methodology in our project, how we were going to understand our project from a sense of 'practice-led research'. However, the explanation for this term did not resonate with me, so hopefully I can get some clarity on this in upcoming sessions. 

For the next week, I decided to come up with some details of what the elements of my project should include, use and look like. For instance, I discussed the type of armour I would use in my project, and how it should be functional and effective as armour. However, I explained some possible visual decisions too, such as to not include a shield so that more of the armour would be displayed. I will have this ready for next week when receiving feedback, so hopefully my lecturers can assist me in my success criteria for the project. 

Week 9 - 12: Dissertation & Further Experimenting 

- Success Criteria - Week 9

In this week (Week 9), we were tasked with making a start on our dissertation, primarily by adapting our current proposal into the format of the dissertation document, which is where I started. However, I felt like my research was starting to take a different turn from what I had originally pitched in my proposal, as I had started to focus on more of a realistic approach than a combination of fantasy and realism. This is something my lecturers seemed to be picking up upon, as they had asked me to look at an example of an armour to use as a starting point. 

15th century Italian armour style (Wasson, 2017)

After looking through lots of different designs, I settled on this one. This is a 15th century Italian style armour, made for a knight. The armour is fully plated, and highly functional. I liked this as a starting point as it is a recreation of real armour used in Medieval times, so that suggests it will be highly functional. My idea was to take this as a starting point, then to modify certain elements based on what I felt would benefit from it. For instance, I was thinking to add some ornamental details to the armour to make it look more unique and be more visually appealing for the viewer, the armour in the image looks pretty plain in my opinion. 

From this starting point, I received feedback giving some crucial starting points for my dissertation. The most valuable was in terms of what success criteria I should use to evaluate the success of my project. 

They were: 

1. Armour
2. Weapon
3. Environment
4. Visual Storytelling

It was suggested to use 4 success criteria, as this would more accurately help evaluate the success of my project more thoroughly. The armour and weapon used are important criteria, as they will be the main visual element of the project. The environment will compliment the character, benefiting it in terms of reflections and perhaps other lighting effects. 

For my fourth criteria, it was suggested to consider the aspect of 'Visual Storytelling', because I had mentioned the idea of perhaps telling a story about the character based on the environment and character design. I had mentioned prior that the environment of a recently fought battlefield could suggest a lot of story around the character to the viewer, and this was something I felt could have a lot of potential. The idea of telling a story without the use of dialogue is considered 'Visual Storytelling', and is something I will need to investigate further to better understand it. This would tie in with my other 3 criteria, as the 3 elements would help to create a visual story, and the 4th criteria can be used to evaluate how successful this has been implemented. 

Now that I have my 4 success criteria laid out, I will begin progress on my dissertation over the last few weeks of this term, aiming to have a first draft complete as soon as possible to receive much-needed feedback and guidance. 

We were also tasked with further experimenting this week, which is due to be presented on the last week of term. I will have a think about what I would like to try to achieve and produce in this time, and make sure it is suitable in terms of the scope of work done over a 4 week period. 

- Disseration - Weeks 10-12

I now have my 4 success criteria, so I can start making progress on my dissertation. I started my dissertation by setting up the document, and putting it into the correct format. I was about to start blocking in parts of my proposal, and merging it with the dissertation document, but I felt that a lot of what I had written in my proposal was either no longer relevant or useful other than my ideas for the practical work. As a result, I decided to make a start on my success criteria. 

Shoulder blades example

For my armour success criteria, I wanted to give a basic outline to the specifications of the armour, and any modifications I am going to make. So, I stated I will be looking at a 15th century Italian style armour and using this as a starting point. From there, I will be adding certain modifications to improve the visual quality without hindering the functionality of the armour, such as ornamental designs, gilded trims or other elements. One suggestion I had was for the shoulder blades of the armour, which are typically quite plain and simplistic. During my research I learned that some medieval armour would have large blades protruding from the shoulder plates, with the functional purpose of preventing a blade from sliding across the shoulder to the neck of the wearer. This can be seen in the image above of a different style of armour. There were present in the 15th century Italian design, but were very small, so I wanted to modify these to be larger and more effective for the wearer. 

Elements of a sword

I then moved on to the success criteria for the weapon of the wielder. I had similar discussion, considering the starting point and then moving on to possible modifications to improve the visual quality without hindering its effectiveness. I specified my weapon would be a 'hand and a half' sword. A topic I brought up was in regards to the hilt used in a sword. A sword's hilt is made up of several parts, but is essentially all parts except for the blade of a sword. This hilt would take the form of a 'Cruciform' which is in the shape of a religious cross, to be a religious symbol for the wielder. There is a lot of research around the Medieval period, and there was lots to discuss about the significance of religion around this time. I determined that this could be a very important design to include in the sword. 

Environment example - Battlefield (The Witcher 3 by CD Projekt Red)

My third criterion discussed the environment of the scene. From my sketches and initial ideas, I decided to go ahead with my previous decision to use the environment of a recently fought battlefield. My idea was quite similar in regards to the image shown above, but I am currently debating on the specifics of the scene, such as whether it should be set at sunset or midnight. I think I would like the scene to be set at midnight for better reflections and a darker more gloomy atmosphere. I went into more detail about these elements in the environment, and how they would reinforce my previous 2 criteria (armour and weapon) by offering lots of reflections and enhancing their visual appeal. 

Finally, I discussed my fourth criteria: Visual Storytelling. This was considerably my weakest area, and one I need to research more in depth. However, I was able to gather that visual storytelling is a technique in which you can visually tell a story, which would be highly relevant to my project as it will feature a single scene with no dialogue or further story other than the scene provided to the audience. I looked into some relevant theory associated with visual storytelling, such as 'Show, don't tell' (also known as The Golden Rule), a writing technique which sells the idea of showing an action rather than telling, as the action sells the story in a much more powerful way than simply telling it. This theory will be highly applicable to my scene; For example - My character could be slightly limping on one leg, but not giving up and remaining undefeated, which would demonstrate a lot about the character and his/her traits to the audience without telling them. I went into more detail about this and some examples, but this is an area I still need to cover in some more depth. 

For the remainder of my time, I started blocking in other sections of the dissertation document to get a better bearing of my work and structure it more clearly in my head. It helped to get more things down so that I would have less to fill in. Most notably was the section on my ideas for the practical work, which I used my proposal to help fill in this section, with some improvements based on my feedback from past sessions with lecturers and peers. Though some areas I struggled to fill in myself, as I got confused, such as the Methodology section which I still was not fully understanding. I decided to fill in the sections as best as I could, before receiving feedback from our one-to-one session next week. 

Feedback

I was able to receive lots of valuable feedback on my dissertation draft. I was able to record two important clips of feedback I received in my one-to-one session. 

Feedback video #1: Tidy up paragraphs, use images to help explain your research. 

In this video, I was critiqued on some of my writing, namely where I have 'overcited' sources of information. It was suggested prior in the lecture to only cite bits of information that are key to making a point or explaining a term. This was then followed up in my feedback session where I have used lots of sources unnecessarily which means I have all the Harvard referencing of the sources across my paragraphs, also using up precious word count too. This makes it hard to read, which was mentioned in the video. 

It was suggested to use images to help with this and explain my point, instead of needing to source numerous times by comparison. I conveniently had an image in my notes document which was a perfect example of the kind of imagery which could really help to demonstrate the points I want to make, in a clearer and more visually friendly way. This is something I will take on board. 

Feedback video #2: Advice for my methodology

Diagram of 'Arts Praxis' to help with our methodology

Some well-needed advice was given on the topic of my methodology. I have linked the image above as it is relevant and was mentioned during the video. I had filled in 2 paragraphs of what I had thought to include for my methodology by summarising my experimentation, with the intention of presenting it as practice-based research, but this was incorrect. Looking at the triangle in the image above, you have 'Know-how', 'Know-that' and 'Know-what' which are the 3 elements which I need to cover in my methodology. This only needs to be brief (~200 words) but my lecturers gave me an excellent and helpful example of the weight of the armour, and how I can adapt this into my methodology. I may not necessarily use the example of the weight of the armour, but the example really helped me to understand more clearly what was being asked of me, and I will now be able to better respond to the methodology section of my dissertation.

Other feedback

Some other useful feedback was mentioned as well, such as my research area. I had started to draft ideas of what to start to research here such as the Medieval period, and some context in this area as the structure from presented slides in lectures had outlined, but this was critiqued as generalising my area of research, since it did not really correlate with my area of focus for my dissertation. This area of focus should have been on the armour or the weaponry for instance, and not a broad description of the Medieval period as I had started to work on. Whilst I feel some brief context will be useful as it will tie in with my dissertation later, this was a good point and something I will keep in mind. 

Conclusion

I was left with a lot of highly useful feedback from this particular one-to-one session, and am grateful for the help I was given. It shows that whilst I was on the right track for things like my success criteria, I needed to improve my focus and clarity in what I was writing in other sections of my dissertation. I could also benefit from the inclusion of images (or other media) to help illustrate my points easily and clearly. From here, I think I will be rewriting a lot of my dissertation which was critiqued since it was only blocked in basic note form, and use the feedback I received to this time write it better and clearer. I also intend to look back at my success criteria paragraphs and tweak some of them to be clearer and use images in some sections. An instance that springs to mind is my overview of the elements of a sword, which I had tried to detail in a paragraph; I will replace a lot of this with an image instead.

- Further Experimenting - Weeks 9-12

For my further experimenting, I have decided I would like to investigate the program Substance Painter, an industry-standard texturing software. Because my project is to look as realistic and detailed as possible, I felt that learning this program could be invaluable, as I have heard of the potential of the software. Thus, my aim for these weeks was to create a helmet and try to texture it in Substance Painter as well as possible, and learn the basics of the program. From this experimenting, I will be able to determine whether or not it will be something I would like to use in my final major project or not. To increase the scope of work, I will look at making a few helmet designs, and see which design best works for me. At the moment I am looking at 15th century Italian armour as a starting point, but I don't know the extent of how much I would like to modify/change of this starting point, so this might be an interesting way to gauge this. 

15th Century Italian 'Avant' Armet (helmet) - Reference image

Here is the 15th century Italian-style helmet, which I was using as reference to create my helmets in Z-Brush. I decided to start with something closely resembling the reference image, and become more creative from there. 

Z-brush - Screenshots of 4 helmet variations

Here are the Z-brush helmets I managed to create. Despite the order of images, the order is bottom to top, since every time I created a helmet I would move it down and create a new one, so that the symmetry option would maintain. Each design offers a visor which opens. The bottom design most closely resembles the reference, where as the top image is very loosely based on the reference to give some variety. Despite my efforts, I was most keen on designs #1 and #2 here (the bottom 2 helmets) since they looked the most elegant whilst also retaining a sense of high functionality to me. Because I had added a leather strap around the back of the first helmet, I decided to use this for texturing, so I could add some variety with a leather texture. I thought it might be interesting to play around with mixing metallic textures with non-shiny ones. 

I exported the bottom helmet into Maya, and retopologised it since the geometry of the original was quite messy, using triangles in some areas and overall being too high poly. The process was simple, and I retained each layer as being separate layers.

 I then UV mapped the helmet onto a single space, in the most optimal way. In the screenshot you can see I have made the UV shells as optimal as possible in the texture space to give as much quality as I can, a good practice I would like to continue for my actual project. 

After subdividing the model for a smoother appearance, I imported the model - now UV mapped - into Substance Painter. I watched some tutorials to give me a basic introduction to the program, then had a go at texturing my model. 

Substance Painter - Rendered

This is what I was able to produce in Substance Painter. The result is very high quality due to my UV mapping, but this made me more aware the quality might be much lower for the full model, if using a single texture space. This will be something to think about, but I am currently thinking to use 3 texture spaces, one for the helmet, one for the sword and some extra details, and another for the main portion of the body. 

The result I have created looks solid, using lots of different layers to build the result. The model uses a diffuse (base colour), roughness, metallic, normal and height (displacement) map. Substance Painter made these exceptionally easy to create, and in high detail too. I found the software to be a vast improvement over the alternative of Photoshop, which I struggled to use before. I will definitely be using this in my final major project. 

I was able to use all sorts of maps to create this helmet. Normal mapping is something I am not very familiar with, especially since they are difficult to create from scratch. However, I was able to do this with relative ease in Substance, as you can paint normal maps on quite easily, even using preset options of primitive shapes as well. In addition, I was able to use existing materials in Substance and delete everything except the normal map, to add it onto my design, which helped to produce more realistic lighting results without having to create a normal map from scratch myself. 

There are some flaws with the result, most notably the inconsistency of the ornamental designs. I did this quite freely to generally experiment with the sort of techniques I could use in final designs. I was quite happy with the simple golden trim across the helmet, but I think I would want to research more accurate designs before future implementation. I tried to add a large pattern on the side using a height map to give it some depth too. The result looks interesting, but messy in my opinion, especially due to the inconsistency. I had already used a simple stripe across the edges of the helmet to displace it, which I think worked quite well, but I think elevating the gilded patterns across the edge of the helmet might look better. I will work with references more closely when implementing this in future. 

I also had some difficulties when creating this. My main problem was that sometimes painting on one part of the helmet would cause it to appear on another side. It seemed to be areas which were very close in geometry. For instance, painting on the leather strap would often paint on areas of the helmet near the strap. I don't know how to fix this issue but it definitely caused an element of struggle during the creation of the textures. Also, I ran into some minor issues such as not being able to mirror some details, or use certain effects. One effect I wanted to use was a worn metal edges generator, which would show the paint/material only on the edges of the model. I had planned to use this to create the gold trim you see on the design. However, the effect did not work as the effect used a 'Curvature map' to generate it properly. I have no idea what a curvature map is, I had never come across the term before now. This effect would provide a big element of control when texturing in future, so I would like to investigate this further and try to add a curvature map to my model if possible so that this effect would be usable. 

When I rendered my result in Substance Painter, the visual result was really effective and of high quality. Though, I quickly realised that this would not benefit me since I wish to include some subtle animations and other details in a scene, but Substance Painter only captures a still. I therefore decided to have a look at other rendering options and test their rendering capabilities. To do this, I simply exported the texture maps out of Substance Painter so that I could import them elsewhere. 

Blender - Rendered

Here is the result in Blender, using the Cycles renderer. I was able to connect all textures except for the displacement map, which I had trouble adding to the material correctly. The result is very close of what Substance captured in its native renderer, and would be suitable quality for my project. I was able to fix the noise issue I was having in my previous experimentation by altering the de-noise options. I tried to replicate the lighting as best as possible, but unfortunately the HDRI used in Substance was not compatible in Blender, so I had to source one with a similar lighting setup/result. Overall I am happy with the result, but it could be improved with some better post-processing such as perhaps an increase in contrast, something which seems lacking when compared to the render done in Substance. 

Unreal Engine - Screenshotted

Here is the result in Unreal Engine, using the same HDRI as what was used in Blender. In terms of texture quality, the result is quite close to what was shown in Blender. Unreal Engine would also be suitable for rendering if I chose, since I could import the scene and animations into Unreal from another package. However, there are a few problems with the result. The biggest problem is the same as before in terms of the inconsistent shadows. This is especially apparent in comparison to the renders done in Blender, as you can see how the shadows should be appearing around the visor and inside of the helmet. Unfortunately this issue has persisted across both experimentation sessions in Unreal, as you can see in the comparison image (which also shows the vast improvement of quality). I don't know the cause of this, but if I am unable to find a solution Unreal Engine will be unsuitable for use in my final project.

Additionally, the helmet shows some transparency in some areas leading to gaps in the model. This is a fault on my part by using planes to construct the model to save resources and texture space. However, Unreal Engine does not appear to render these (perhaps as an optimisation method) so if I were to use Unreal Engine, I would need to modify my project to accommodate this problem.  All in all, Unreal has a good result, but its flaws mean I will probably be ruling out Unreal in the use of my final project. 

Maya

Unfortunately, upon trying to import my helmet in Maya, the result just didn't work, I honestly couldn't get it to import correctly into Maya using the Arnold renderer. It either didn't work or gave me a result which was not the same as what I had created or got in the other 2 renderers. I hope to find a solution to the problem, as it would be unusual if it were not possible, but it will take some figuring out on my part to get it to work. If I cannot find a solution, I will have no choice but to rule out Maya for the use in my project. 

Feedback

I presented my work from my further experimenting over the past few weeks this Friday. Overall the feedback was quite positive, complimenting my level of detail achieved in the model. It was suggested 

that I could look into particle effects in Substance, for things like damage to the model or procedural effects like blood or rust perhaps. I think this is a great solution, something I had seen in videos about Substance but not one I had investigated myself. I would like to look at the potential of this when working on my final major project, especially since I will already be using Substance in my project. Finally, a comment was made about the metal 'stripe' across the model, saying that it was effective and really helped to elevate the model and boost the visual result. Due to the positive comment, I will be including this in some form in my final project. Unfortunately I was not able to receive feedback on my problems other than a suggestion for Unreal Engine which I ultimately dismissed to not being a solution. 

Conclusion

From my further experimenting session, I have been able to create a few varied helmet designs, and learn the basics of Substance Painter to add a high level of texture work to it, then was able to render the result effectively in a few programs. The experimenting has helped me to learn the huge benefit to using Substance Painter, so I will be using this in my final major project. I was also able to identify the sort of design I was looking for (at least for the helmet): Realistic and closely resembling the reference, though not entirely identical. 

This experimentation also brought my attention to a few issues. The main issue being the program used to render the result, as all software (Blender, Unreal Engine, Maya) had some underlying issue when it came to getting all the texture maps to work properly. This is something I will need to investigate and hopefully fix for a minimum of 1 rendering option. Right now Blender has the least problems, but Maya is the program I am most comfortable in, so I would be glad to get the problems fixed for Maya if possible. Though I recall Maya had a grainy result - which can also be fixed - but might be a difficult fix, so Blender seems like the ideal choice as of right now. 

Furthermore, I have decided to model my helmet with no inside. I don't believe I will want my helmet to open up since there will be no actual person inside of the armour, so it is a waste to texture and maintain the inside of the armour, especially when it caused some issues like in Unreal Engine with parts not showing properly. This is also how one of the models which I was using as reference had been created, so it seems to be suitable practice. I will give this a go and see how successful it is. 

Moreover, I am now aware of the potential problem of UV space, which would be alleviated but not solved if the inside of the armour was no longer considered. This could be optimised largely, but ultimately I think the use of multiple UV maps will be necessary to retain a good sense of quality. I would like to look into potentially UV tiling or simply using separate UV maps for different sections of the armour. This is something I will need to investigate and test myself to decide. However, I learned the benefit of using a 'UV checkermap' when UV mapping (you can see areas stretched or low on quality) so hopefully this will help me to diagnose and better optimise the layout for the highest quality result. 

Finally, I would like to investigate some of the problems I encountered in Substance. Most frustrating of all was when painting on one UV shell would paint across to another, but  I have a hunch that this will be an easy solution, so I will look into this. In addition, I want to look into curvature maps and how I could implement one, so that I can easily use the generative effects in Substance for more creative control and freedom over the model. 

    -    Christmas Break:

During the Christmas Break, I would like to make a start on my model, whilst making sure to complete my dissertation. I'd also like to investigate some of the issues I had encountered. 

Finishing my dissertation:

After a couple weeks of break, I made it my top priority to complete my dissertation. In summary, I went through and completely rewrote some paragraphs such as the section on the success criteria #2 (the character's weapon). I did this because not only was it agreeably hard to read, but I also decided to change the sword type to a two-handed sword, which I felt would fit better for the visuals despite hand-and-a-half swords being common weapons of knights. Although this was a slight dent in the historical accuracy, I am focussing on the realism of the visuals and the functionality of what is there, as opposed to how it should or could be. I believed I had a solid reason to change this aspect of my criteria, and did so in a more consise and well written manner. 

Furthermore, I went through and added images throughout my dissertation to help illustrate and visualise my points made in my work. This was espeically useful when describing some of the details I wanted to include or look for as reference, as opposed to describing them in monotnous detail. This certainly helped the readability of my document as well. Moreover, I was able to trim out some of the unnecessary citations I had made, since I had included an unecessary abundance of sources to try and reinforce my points, despite the fact it mostly just made it hard to read. It was advised to use sources to quote key information instead, and I aimed to achieve this. 

I was able to go through and fix my methodology thanks to the help of my tutors' feedback. Their explinations and suggestions made this much easier to understand and write down. Despite it only counting for a small portion of the document, I feel that I covered this area to a good standard. 

Adding to this, I went to my preliminary research section and rewrote this completely, as a lot of it was irrelevant. I had originally kept some historical context here to tie into my weapons criteria section as it explained some relevance of a Cruciform hilt, but this idea was ultimately dismissed along with the hand-and-a-half sword. This gave space for more meaningful research which was more significant to the focus of my FMP being about 'Function over Form', which became my title of the document; This title felt much more fitting and relevant to the overall theme of my dissertation. 

Moreover, I eventually finished my 4th criteria: Visual Storytelling. This was a section I had been struggling to complete due to its heavier dependance on theory and technique, that which I was not too familliar. However, upon suggestion on earlier feedback sessions, I decided to look into potential camerawork and their meaning. This was useful research which is something I will be able to implement and also explore during my production, but helped me to outline a detail I had not yet fully considered despite its importance. 

Finally, I completed the section regarding my plans for the production, trying to express as much detail as I could without elaborating too far and using up precious words contributing to my total word count of the document (3000 +/- 10%). I feel that I was able to get across a clear picture for the direction of my FMP without elaborating on the absolute specifics, which were probably unnecessary for this document and will be detailed in future. 

After reaching this point, I was nearly ready to hand in my work. However I waited for the feedback session during the 8th of January to hear what my tutors had to say and any advice they might have. Whilst suggestions were made and discussion was raised, the feedback was quite minimal which suggested that I had done good work thus far. The main comment raised was simply the inclusion of links between paragraphs, which was described as something really easy to do, but really helps to increase the level of the work submitted. This was referred to mostly for my conclusion, and I implemented this as suggested.

This work concluded my work on the dissertation document, which I proceeded to submit for marking. I am currently awaiting results. I feel that I have produced a substantial body of work, and hope this is reflected in my grade.