Post Mortem - Character Project

Post Mortem - Character Project

Over the past 7 weeks, we were tasked to concept and produce a character with our own likeness that will fit within the constraints of the village style guide. The character would have to be around 7.5 heads tall that conveys a light-hearted and positive mood. In addition, our character may have some stylisation to its anatomy and clothing such as large chunky shapes. All textures must be hand painted and the Character must be rigged with a bi-ped using no less than 28 and no more than 60 bones. In terms of budget, we were allowed up to 4000 Tris for the characters body and up to 500 Tris for any props or accessories.  Moreover, one of our expected learning outcomes was to produce clean and efficient topology for our characters, which will allow for clean deformation during the rigging and animation stage.

Notes from the drapery session

The project began on week 15. The first week back and we were introduced to an anatomy boot camp, which consisted of a proportions checklist lesson both on body and portrait as well as a brief lesson on drapery. The anatomy boot camp lasted roughly around 3 weeks and by week 18, I started my concepts, which lasted around 2 weeks until I finally delved into modelling texturing and rigging for the rest of the project.

Five things that worked well

Tartan texture

Textures of plaid/tartan clothing - After spending so much time doing research and reiterating upon reiterating I feel as If I managed to create an efficient tartan texture to convey my characters Scottish background. What worked well in particular was the fact that firstly the texture was tile able meaning it didn't create any noticeable seems and Photoshop as a digital medium allowed to me to manipulate the colours via hue and saturation with a clean outlook.

Colours varied in saturation

Colours - The colours of my final model I feel were varied and used natural colours, which convey a friendly and warm look to it as well as the balance of vibrancy and saturation, which allowed colour to not be so aggressive to one's eyes. To some extent, I would argue that it somewhat fits with the style guide, however certain areas and harsh brush strokes need to be refined to fully fit within the guide.

Research and mood boards - I felt like my research was very extensive and detailed and it lead me to find out about cultures, which aren't particular represented within the mainstream media. In addition, from this, by producing these I got to learn about historical cultures and job roles which have now peaked my interest.

Tri count - Normally budgeting Tris has always been something I have struggled with however with this project I made the conscious decision to always cut down on unnecessary geometry after finalising each stage of my model. This allowed me to allocate Tris to my accessories, which I had planned to really sell the role of my character.

Similarity to final concepts - Compared to my previous projects I do feel as if this is one of the first time I managed to get it as close as I could to my original concept. This may be because after spending all this time on the course I have accumulated skills, which have now allowed to obtain a look similar to the scope of my ideas. During the course of modelling, I felt as if in the most part, I was in control and I didn't run into too many technical issues that I didn't know how to solve.

Five things that didn't go so well

Stylisation and anatomy - One of the challenges I had to face when modelling my character was maintaining a stylised look without ending up with broken or odd anatomy. I believe this did not go so well as I had hoped. There features that I wanted to exaggerate were the hands and feet of my character and model. These possibly weren't the best of features to choose from considering they are the two features I struggle drawing traditionally. This as a result led to the feet of my model to look like Duck like feet in addition to my hands looking like mittens. Looking back I should've really studied what I wanted to exaggerate in more detail so that I have the ability to manipulate them without them look too broken.

Too much details on such simple geometry

Texturing on low poly props and accessories - When texturing my low poly props I think I went into too much detail in painting the textures of the models, more specifically the pouches. I rendered them with such detail including folds and distress however; the model itself is a block square shape. This creates a distracting and cheap looking contrast, which aesthetically does not work very well. When making the texture of my model I had initially planned for the model to have a higher ploy count however due to budgeting reasons I had to take it down which lead to the box look of the final prop. If I had more time I would have adjusted the texture of the pouch to accommodate this.

My rig around the kilt - To the best of my ability I tried to weight and rig the kilt of my model to achieve a clean look however as much as I tried this wasn't the case as there is still some clipping around the legs. If I had more time and practise with rigging I would like to remedy this problem, however I have been told by some of the older years rigging with baggy clothing is quite a difficult task so I shouldn't be so hard on myself.

Colour studies look muddy - When iterating colour variations for the costume of my character I simply blocked colour on top of my chosen value pattern and set the blending mode to overlay. Because I chose very dark midtones to work with this meant that the colours I placed ontop created this really muddy and ugly look to them which caused problems in the future when trying to choose a set of colours to work with.

How it conveys its job role- I feel like if I were to show someone my model and ask them what type of character my model is I feel as if most people will assume that he is a merchant type character due to the pouches of the character. This may be due to the obscurity of the chosen role I selected for my character. Although interesting it isn't a stereotype of medieval history people would associate with.  This all goes back to the idea that the game industry works on visual stereotypes. If I really wanted to emphasise the job role of my character I should've spent more time allocating Tris to my accessories to really sell the job role. On the other hand, if the character was with a barber building, I think that is where my character would work contextually but on his own it suffers from lack of readability.

What project changes happened?

My initial reference for my character

Sketch I produced of my ref 

Before the character project had even officially begun, I had planned to make a blacksmith character based on a pedestrian who I had run into around Leicester. This was of course before we had the official brief and we were tasked with taking pictures of people around Leicester to build a visual library of interesting faces. Initially I had thought that the role of our characters had to be based on the buildings we had produced for the village building project to which mine was the blacksmith.

Ac Brotherhood blacksmith

Visually the outlook of the blacksmith character that I thought at the time was going to be producing would have been similar  to the Blacksmith character from AC Brotherhood. Looking back It was a good thing the brief changed as I would have produced a copy of an already existing character.

Initial concepts

After receiving the final brief and doing some research into medieval job roles the character that I concepted originally was planned to be a Moorish barber surgeon with the intention of reflecting some of my African decent not just in skin tone but in clothing and textures as well in my character.

However, after an initial review I was told I was not meeting the style guide as it turned out. Although not clearly stated in the documents our tutors specifically wanted a European medieval look to the character. Initially I was quite annoyed as I felt I would have to sacrifice the whole "likeness" tone to my character in order to fit the style guide. On the other hand as my tutor explained it "if I were to go in industry and I was tasked with designing a medieval character and I handed them something out of the Arabian knights even if he is historically accurate they would ask me to scrap that character and do it again." This bit of verbal feedback allowed me to contextually understand what It is I am making as well as to understand that one could argue that the game industry works on visual stereotypes to direct and allow users to understand what is what.

This lead me to design a barber surgeon highlander to which looking back I feel is much more visually interesting than my original concept which relied on very mainstream middle eastern stereotypes.

What would you do different

One of the first things I would do differently would be to have better housekeeping of files. On the hand in day, I spent a good couple of minutes trying to sort and arrange files out to make sure my design process made sense when I could've been using this time to adjust rig of my kilt. Moreover I think although the job role I chose was very different I should've chosen something that is quite well known among the public like a medieval doctor, this way avoiding the issue of my character not reflecting their job role so well.

Another thing that I would be mindful of is how anatomy affects stylisation. To elaborate I should've really taken in consideration how ones knowledge of anatomy can separate what makes good stylisation from bad. I decided to stylise areas of the human body, which I am weakest at instead stylising areas, which I feel more confident in. This as a result caused me to seek shortcuts when modelling these areas of the body, which in turn degraded the quality of these places.

In addition, I would change how I would approach making colour variations of a costume for iteration purposes. I am now aware that value can affect the colour which is placed on top and I will be more careful when making these colour iterations so that in the future it's much easier to choose a colour palette to work from.

Finally, one thing I cannot stress enough would be that In the future I should pay more attention to the brief that is set. On the last day of hand in I found out my model was smaller than the brief requirements in terms of CM. I spent the whole day stress and Google searching how to re size a model without it damaging the rig and bi-ped. If it wasn't for the tutors helping me to re scale the scene my character probably would've been the size of an ant when placed in engine.

Conclusion

Overall, this has been one of my favourite and most beneficial projects of the course so far. I have learnt so much in terms of 2d and have refined my skills in 3d as well. In terms of the character itself, I would say it has been a success considering it has been my first attempt at character modelling however there is a lot of room for improvement. Something that I will seek in the coming weeks.