FMP - weeks 3 - 7 - proposal
- Anouk Dutrée
- 26 nov 2022
- 6 minuten om te lezen
Bijgewerkt op: 27 nov 2022
So at the beginning of this module I had the intention of writing a blog every week... Yeah...Whoops! But that's no reason to give up now of course. I'm just going to walk you through all the work I did in weeks 3 to 7 in this one big blogpost. With week 7 having been the proposal deadline week all the work was related to that, but for this post I'll just highlight the user research outcomes and the dragon prototype development.
User research
In weeks 3 and 4 Rob and I spent a ton of time on user research. With out primary research question being:
How do we address/cater to a target audience defined by their interest in dragons, as opposed to their love of a specific gameplay genre, through a PC game?
We sent out a survey to gain a better understanding of the market segment that consumes dragon related media, who they are, what they want and why they want that.
The main aim of this survey was to answer the following questions:
What is the core player/consumer fantasy is when interacting with dragon media?
Is the player/consumer fantasy distribution similar across different media, and is the currently dominating “Slaying dragons” fantasy indeed the most popular?
The survey was short and primarily build up of closed questions. We had open questions to ask for more details about their favourite fantasy though. We posted the survey in a bunch of subreddits and honestly, it blew up! Overnight we managed to get 400+ responses! It was a dream to be able to gather so much data so quickly.
I analyzed the data with the use of a custom python script as wall as built-in google form statistics. The data showed a clear preference for two player fantasies:
Being a dragon
Bonding with a dragon
So that answered our first question. We could also see in the responses that the fantasy preferences were quite similar across media, and that the slaying trope was definitely not the most popular one. This was great to find as it confirmed our hypothesis.
When digging into the qualitative data on why people loved their chosen fantasy so much, it became clear that our target audience is really passionate about the top two fantasies. The people who picked the slaying fantasy typically said something along these lines when asked why "I like skyrim", whereas the "being a dragon" people answered entire paragraphs about how much they love dragons. We also often found answers of people saying they simply "Didn't know any game since Spyro that succesfully does it". This gave us confidence that there truly is a gap in the market! It also allowed us to map our target audience to three user personas.
Dragon prototype
Our prototype approach was two-fold; a mock-up of the game loop and its associated controls, plus a dragon character prototype. Rob focussed on the first, while I focussed on the latter.
My dragon prototype consists of a dragon 3D model with the following animations:
Flying
Soaring
Landing
Idle
Walking
Firebreath while flying
Perched
An animation reel of the dragon prototype can be found here, or in the description of Figure 11.
The prototype served two purposes:
Getting a character ready that we could use game prototype development directly after the proposal hand-in
Finding out if the art style we would like is one that I can achieve with my current skillset
Both Rob and I were happy with the quality of the dragon model and the time investment that was necessary for it, so we will be proceeding in this style for the rest of the game. The dragon model can be seen in Figure 5.
4.1 Development process
The workflow I followed is a standard workflow for creating game ready characters, as presented by Abbitt, Price and Totten, among others (Totten 2012; Price 2017; Abbitt 2020). I picked this workflow since it allows for fairly high detail while not requiring a lot of computing power from a player’s PC.
The development of the dragon model started with a moodboard and a sketch of a dragon that combined my standard art style with the style of our moodboard (Figure 6). I used this sketch to make a low poly base model (Figure 7). The sketch was made before we had the survey results though, and I didn’t have a very clear idea of what our audience would like to see.
After we had analysed all survey responses and crafted our user personas, I made some changes to accommodate our findings. It became clear that the dragon needed to appear majestic and powerful, yet still approachable, so that it can convey a sense of power to the player while still inviting users to create an emotional connection to the dragon.
The most important change was that I simplified the design to achieve a more stylized look which would appeal more to “Mary” (mainly noticeable with the tail, wings, and head). I felt that my sketch was too “grizzled” to appeal to her. I also opted for larger eyes than in my sketch to give a hint of cuteness. Large eyes are known to invoke a baby effect that could potentially help in triggering the “bonding” process that appears to be important to our target audience (Glocker et al. 2009; Miesler et al. 2011).
When making the low poly model I had to keep the animations in mind that the dragon would need to have. For this reason, I made sure I added additional edge loops around key joints to allow for better freedom of motion later.
Once the low poly model was done, I proceeded to a high poly sculpt (Figure 8). This high poly sculpt was made to be baked onto the low poly model. This way I would be able to give the dragon more detail and character, while keeping the load on hardware of end users low, as Totten and Price explain (Totten 2012; Price 2017).
In the sculpting process I tried to enhance some features that personas “Peter” and “Mike” would like to see. I made the dragon appear more muscular, added some gritty details to the face (especially the eye area), and I made the “fingers” of the wings look “bonier” and a touch demonic. I made sure to double check the model from the key gameplay angles (primarily from the back and from above).
After I baked the high poly sculpt onto the low poly model, I moved on to texturing (Figure 9). With the overall design being quite stylized, I wanted to balance that out with a more realistic texture. I was aiming for a “ragged skin” look. To up the realism I added a crocodile skin bump map.
Considering that the dragon is only half of the player character, I wanted to be able to use the texturing to connect the dragon to its human counterpart. The human part of the character is never fully in view as its in FPV. However, the hands are in view and can be tweaked. Werewolves are typically displayed with extra hair on their hands to refer to their wolf counterpart (McMahon-Coleman and Weaver 2012). I intend to alter the skin texture to reflect the dragon, most likely by including the base colour/texture of the dragon in the fingertips of the human hands. Potentially I’ll also use a normal map that will give the illusion of scales on the knuckles. This way there is a visual link between dragon and human.
After texturing I also added breast plates and teeth. With the breastplates I wanted to give a nudge to Spyro’s design (Figure 9). This would hopefully help make the dragon more approachable for “Mary”, while still relating to “Peter”. With the teeth I went more realistic though, as opposed to the cartoony teeth from Spyro. The dragon overall needed to radiate strength to be able to give that power trip player fantasy captured in user persona “Peter”. Dragon representation throughout Western culture puts incredible emphasis on a dragon’s fangs and general mouth area to convey a dragon’s ferociousness, as Gilmore notes (Gilmore 2003). Therefore, I felt that the dragon would need proper fangs to fulfil the player fantasy of being a dragon.
Since we plan on having a large customization feature in our game, I needed to make sure that my textures were made in a way that allowed for that. For that reason, I split up the texture in 5 parts:
Base skin
Eyes
Horns and claws
Breastplates
Webbing of the wings
This would allow for a set-up in Unity where we can easily change each of these parts separately.
The last step in the development process was rigging and animation (Figure 11). I rigged the model with a modified horse-rig, as a horse has similar joint placement as the dragon. I made sure to add bones for the nostrils and eyes to allow for some facial expressivity in close-up shots.
For the animations I studied quadruped and bird locomotion to make sure I was moving the joints in the right way (MSC Art 2020; Zamarripa 2020). Due to lack of time, I kept the animations simple for now, and I plan on refining them later in the project.
The breast plates aren’t moving perfectly in the rig, which is noticeable in the flying animations. However, the dragon will never be visible in frontal view while flying so a player should never be able to see that.
Comentarios