top of page
  • Writer's pictureGabriel Fassenfelt

Thursday 22/4/21 - The first scene

Updated: May 6, 2021

I decided to start on the very first scene as it was probably the simplest to create and would actually technically be used twice in the film. For this I wanted to create a scene that used a lot of thunder and rain for atmosphere and drama so I looked for tutorials on how to create this before actually placing the objects and creating / importing a set.


I started with something more basic, trying to make the night sky and moon that the camera would pan up to. For this I followed these tutorials:



The moon tutorial was simple and very straight to the point, it was actually pretty easy to create as they provided the textures for the moon. The downside was how quickly they spoke and didn't provide enough details so I often had to fill in the gaps. Despite this the outcome was great, I now had a moon to use in the scene and for the entire film.



Next up was making the night sky, this tutorial was a lot easier to follow as they explained each action clearly and with enough details. The outcome was a night sky with clouds that looked really good for something that took just under 10 minutes to create. However, once I placed the moon into this I noticed that it looked really out of place with the clouds as realisitcally it would be behind the clouds. This meant I had to remove them from the background, I decided to not make my own clouds for the time being as the night sky looked fine without it.


Next I decided to import the helmet model and the body for the scene as I wanted to get an idea as to how the scene would look before implementing rain and thunder. I also was considering making the set alongside this.

I extracted and downloaded the model helmet as a separate model alongside the body without a helmet, I used a default head underneath as I wasn't planning on showing the head of this character. Placing the helmet was simple but the body would be difficult as I needed to pose it. I placed a plane underneath and gave it a apshalt texture using an image I found for free online. This was the result (I also turned on bloom, ambient occulsion and other effects that I used in my experiments to give it a more cinematic and film-like appearance.





Next was the body. The issue I had was it needed to be rigged, which essentially means giving it a skeleton underneath for all the joints so you can actually pose the models. This is something I haven't done before and would take extra time in a process that's already very time-consuming. So with the help of a friend I was recommended a website called Mixamo where users can send models which will automatically be rigged in a few minutes. Not only this but it also provides tons of free animations that would help me a lot when creating the film. This is where I ran into a problem, whilst it rigged the model perfectly, it wouldn't have any of the textures. I found no help online and my friend was also unable to figure out the issue. This meant either I had to rig the model myself or find another solution. For now I decided that I would just leave the model out of this scene and just feature the helmet for the time being.


Next was going to be getting the set and adding in the rain. Based on the work so far I was also thinking ahead to what I would do to create the animation. The website not exporting with the model's textures meant that I was limited with options, I decided that I would first attempt to rig the models using tutorials online and then work based on the outcome. If rigging proved too difficult I'd have to reconsider my options for the film, one possibility being making the entire thing made from stills and transitions like my plan for the extra scenes. The other would be using the game itself to record the scenes, this is a practice used by a lot of people already as the game has it's own built in theater mode that allows players to move a camera around footage of them playing. This was my least preferred option as it would really limit my choices for the film and produce a low quality outcome. I also spoke to my friend to see if he could look into why the website wasn't working for me.


For the set I decided not to create the full set at this moment and instead keep things simple by adding a black wall in the background. This scene is meant to show limited visibility so I wanted essentially only the ground and the helmet to be seen. I also found a solution where I could keep the body in the scene. I coloured it to a very dark grey, the low visibility meant you could see the silhouette and I didn't have to worry about the body's textures. I also felt I should attempt to implement fog to add to the atmosphere of this scene, though I first decided to work on the rain as that would be really important.





Creating the rain required following a tutorial, in the end it proved to be a lot more difficult than anticipated as it took an entire day to work on. The tutorial felt very rushed as the person behind it kept jumping between parts without much explanation. This meant that the outcome wasn't as great as it could have been, I felt that the rain looked really weird compared to real life but unfortunately the tutorial didn't explain how to fix this.


This was the tutorial I followed


These are two screenshots of how the rain looked whilst making it



This was my first draft for the firsts scene, I plan on adding a thunder light effect and attempting to adjust the rain so it looks more realistic. For my first attempt at making the scene it was okay however I felt the rain was too distracting. It needed motion blur and changing the size. I also felt that whilst the static camera angle worked, experimenting could find a better result that helps affect the atmosphere of the scene.

In my attempt to improve the scene I first made the rain particles smaller and added more motion blur to the droplets, this was pretty simple though the settings for motion blur was relatively confusing as I used terms I haven't heard before. I also decided to add to the theme of the calm before the storm, this time in a more literal sense. Since this scene is technically at the end of the short film's narrative I used keyframes to make a flickering light effect to simulate thunder. This also added to the suspense of the scene as it left it in near darkness aside from short moments, giving the audience a bigger sense of mystery.


This is where I began to experiment with camera angles as I was happy with the other changes I made to the scene. The first I tried was having a dolly shot moving away from the helmet, I wanted to experiment with the idea of having the audience leaving this mysterious helmet in the dark. To do this I set keyframes at the starting point for the camera as well as the ending point since it was only going to be a simple dolly shot.

When watching this however I felt like it didn't quite capture a feeling of suspense as you started on the subject, I wanted the scene to have the viewers question what they were seeing. Having the audience start on the subject meant they could get a better look pretty early on. I wanted to try a dolly shot getting closer to the subject but first I wanted to try something a bit different. At this point in time the shots were always from the same angle, I felt that a different angle might prove to be beneficial to the shot.


For this one I wanted to start the camera looking slightly from the side towards the subject from a distance away, instead of from the end of the street like before. This also meant the red light would be in shot the entire time, something I liked as it could be used to create a silhouette. The ending shot was just going to essentially be a dolly towards the helmet.





I liked this better than the dolly moving away although I didn't feel this was right, it left the body barely in shot so even less noticable than I had originally intended. It also felt like the shot started too close to the helmet meaning the camera moved in too quickly to make the scene as suspenseful. One aspect I sort of liked was that this time the helmet was looking almost directly at the camera which could create a sense of discomfort but I also felt it didn't quite look right in the shot. It looked like it had been placed there instead of falling on the ground more naturally. I wasn't against making it face the audience but for now I prefered it being at an angle. Though I did want to play around with having the red light in shot, I felt if I could create a silhouette that could create a really interesting shot and play into my plan of keeping the helmet in the darkness. I also felt I could experiment with a panning shot here instead and look completely from the side instead of a more diagonal angle like before.




I really liked the outcome of this experiment, the silhouette looked really good and helped with my idea of trying to keep little visibility of the subject for atmosphere and suspense. However, because it was a panning shot this felt more like a transition between shots instead of the first actual scene. I decided to not used this because of that but I really wanted to reuse the idea of having the character silhouetted on a red light, very similar to the Starboy photoshoot from my moodsheet and experiment.


At this point I wanted to see how a simple dolly shot would work moving towards the subject much like my initial experiment. After seeing the results I was sold because whilst it was a simple shot, it worked best to create suspense in this one scene alone due to the pacing and angle. At this point I felt like I was ready to move on from the first scene.



6 views0 comments

Recent Posts

See All

Sunday 16/5/21 - Final Evaluation

I had one major goal for this project, trying to push myself outside of my comfort zone more than I had done for my previous pieces of work. I wanted to not only use skills I was more experienced with

Friday 14/5/21 - Editing and sound design

I wanted to think carefully about editing the clips together, I felt that just putting them together in chronological order and moving on to the sound design wouldn't be a good approach. The main focu

bottom of page