Independent Project - Evaluation

My main goal for this project was a VR or 360 piece. In my prospectus, I explored the theme of dreams and how I would incorporate these types of elements into my piece. For example, I looked at dream core as an example as well as some 360 video pieces like a video of the "Tripp" Vr experience and "Dreams of Dali". I looked at what parts of these pieces I liked and thought about how I could incorporate them into my piece. The things I picked up on were the use of sound and lighting as well as the calming imagery used. I don't think I explored these aspects as far as I would have liked, due to having troubles with Unreal Engine (which is what I mainly used to create my project). However considering this was my first time using Unreal, I think this was a good experience for learning the basics. 

I chose a poem by Edgar Allen Poe to use as a base for the imagery of my piece to be based on. I recorded myself reading the poem and used that for the audio. Although when exporting the 360 video from Unreal this was done through images for each frame so the audio was not exported with it. I think the imagery I created is quite a symbolic take on the poem, especially the use of flowers. However, I do think I could have developed this concept further and done a better job at introducing the different elements. 

I stuck quite well to the synopsis/treatment and storyboard I created however I wish that I had done some more research into using light as a guiding element in Unreal since the light in my scene didn't really seem to do much when I moved it. If I had more time to work on the project this would definitely be something that I would try to look into. I also would have liked to add more nighttime/space elements like what I included in the last section of the storyboard. One reason I didn't include them was because of the timing of the poem, the poem reading lasted around 55 seconds once edited to include breaks and I wanted the flower to bloom around the middle of the piece. Therefore I had to fit the butterfly flying in before this and the introduction of the clock and watering can objects to be after. And since I didn't want the objects to come into the scene too quickly or to be shown for only a few seconds they ended up taking up most of the time after the flower bloomed. Even with this amount of time, I think I could have improved the animations of these objects since they were a bit sudden in appearing, also it would have been nice if I had been able to fade them in similar to how you can fade images in using animate or premiere pro. 

I also thought it would have been nice to have the watering can work, but I wasn't too sure how to implement that. I also looked into the VR grab component and I thought I could use this to allow the user to pick up these objects however since I ended up exporting it as a 360 video I removed the component since it messed up the animations.  Another thing I would have liked to incorporate if I had gone all in on the idea of using VR would have been to have the different parts of the project play ( via the sequencer) as the user walks through the scene. I knew this would have been possible since I found information about it in a tutorial I watched, however, this might have been a bit too complex for a Unreal/VR beginner to use. This would have also required lots of testing to make sure it works. 

Another thing I would have liked to look into was the camera movement. The project was set up as a VR one so theoretically if used in VR the user would be able to control the camera. However, when testing and exporting the 360 video the camera stays still since there is no vr user to control it and I didn't know how I would change this without changing the whole set-up of the project or figuring out how cameras work in Unreal. Also as I mentioned before I had lots of problems with Unreal and I didn't want to end up breaking everything trying to get the camera to move. Therefore instead I positioned things around the camera. I also tried exporting in both 4k and 8k but both of these didn't really clearly show the models in the scene unless they were super close to where the camera was positioned therefore this would be something else I would have liked to look into, but since I wanted something to be able to show at the presentation I had to leave it as it was and also the rendering took more than half an hour so it might have taken quite a while to figure out a fix. 

Another part of the project which I think I could have improved was the clouds. I watched lots of different tutorials but I couldn't really find anything that was a good balance between realistic and not being something that would ruin the performance of the experience. The particle clouds were probably the most realistic however lots of particles are probably not going to be very good and also they were not very substantial (Which is in a way realistic since clouds are not really something you can touch, however, I wanted to clouds to seem like ones that you can touch) I used the model and material clouds alongside the particles however where I made them transparent this made inorganic shapes to show where different parts overlapped and I couldn't really find a solution for this without having to use a super complex material blueprint that someone else made. Although I did use tutorials for the clouds I did try to differ slightly and for the 3D cloud material I had to figure it out myself partially due to the tutorial being old and some things not existing anymore. It was also my first time working with blueprints and creating materials with them, it was somewhat similar to the nodes on Blender however I haven't really used those before either. I did learn quite a lot from this though.

I could have also worked on a UI for the project since there isn't really any introduction aside from the fade-in from black, I think this would have taken the project a step further, however, I don't really know how I would have made that work and I would have had to work out how UI works in Unreal for VR since UI would not be interactive in a video. I could have also incorporated a game-like element to the project similar to the idea of the user progressing the events by walking through the scene, they could go to the next scene by picking up a certain object, I think I would know how to do this in Unity however I'm not sure how I would do it in Unreal. 

Overall I think I stuck to my project proposal quite well and despite the problems that I faced I think I managed to overcome or find solutions for the majority of them. There are lots of things that I think I could have improved on for this project however I think I have learnt a lot and had a good attempt at trying Unreal for the first time and trying to create something different from what I have made before. I will take what I have learnt and the things I would like to improve to my future projects and hopefully if I ever have to use Unreal again I now have some basic knowledge about how it works as well as some more knowledge about Blender too.