Updated: Jul 15, 2019
The NosVRtu virtual reality game project started off well and I was hopeful about getting a good product at the end but unfortunately things didn't turn out exactly how Justin and I wanted. The tracking process of the Foley sounds, the sound design process, and the musical composition all went really well and we managed to get some high quality sounds but the implementation of these sounds into Unreal Engine was a different story. We had all these great ideas to make the level feel immersive and fill it with audio but there was multiple problems with working out how to actually get the audio clips to play inside of the level.
Before I get into that when we had our first implementation session with the animators before we could actually start the implementation process we had to first get FMOD to work with and connect to the Unreal engine. The version of FMOD that was already installed on the computer was a few of generations out of date and it wasn't supported by the latest version of Unreal which we only found out after researching a multitude of reasons why these 2 software's were not talking to each other. After we worked out the problem, downloaded the newest version of FMOD, and booted up the software again it still wasn't working, at this point Justin and I were probably a couple of hours into the session and we hadn't even started to get audio game. We did some more google searching and looking at some of the different FMOD preferences inside of Unreal and Justin worked out that the naming convention that was already preset inside of Unreal for our FMOD master bank was wrong, and all we had to do was change it from Master to Master_Bank, one small naming convention took up another hour of our implementation time. There were a couple of other things and naming conventions we messed around with to get it to work but that whole session is a bit of a blur and I can't remember exactly what else we had to do to get it to work.
Another problem was our lack of experience working with Unreal and not knowing how to set up the specific blueprints so audio clips can be triggered, this includes when the character interacts with specific objects and dialogue lines are played and when objects interact with each other like a book falling and fitting the ground. The animation students even though they had worked with Unreal they didn't know how to set up these blueprints and I think they assumed that we would know and we assumed that they would know which resulted in a situation where no one knew anything. We could have done some research but without having a copy of the Unreal session or a good understanding of how to use it any research leading up to this point wouldn't have been particularly useful. I think I did mention to one of the animators to work out how to get the audio to be triggered but that must not have happened. I think to try and prevent this from happening again I really need to make sure that the people I'm working with on a project like this have some understanding of how to do the audio implementation. I should also download some free Unreal levels and practice putting audio into it, but in saying that we were working on a VR game which potentially has different ways/needs of doing this compared to a traditional game level.
Another disappointing thing was that when the animators were trying to build the level as a playable desktop file the latest version they had with our latest audio session wasn't actually able to be built and the animators had to regress to an older session with older audio resulting in a build that wasn't as good as it could be. I would ideally have liked to have had another session with the animators going over the audio and adjusting the levels of everything a lot more more they were still working on it up until there deadline and there wasn't any time for us to do anything more. I did consider trying to work out some of the implementation at home with there final session but because I don't have a VR device there is no way for me to test the sounds, if this was a normal game level I definitely could have done this and potentially do the implementation without the assistance of the animators. Even though I did mostly enjoy the project and I think it looks and plays really well I don't think I'll be rushing to work on another VR game anytime soon unless I can have more time to work out how to do the implementation or there is a games student working with me who does.
Justin and I had a very small time frame to do this project in which resulted in not much time at the end for fine tuning. We did stick to the timeline the animators had given us but the problems that I have mentioned above basically destroyed the schedule as we spent more time problem solving than creatively working with audio inside of the game environment. I think we had some really great creative ideas in mind to really craft the audio inside of the amazing space the animators had created but with the time we had and the problems we encountered that was never going to happen. I believe if we had another 3 weeks to work out how to implement the audio with the animators or get a games student in to assist us we would have produced a much better product.
All up I can't help but feel a little disappointed with how the audio turned but I have now learnt that working with the Unreal engine and VR is really not an easy thing to do and there can be multiple unforeseen issues that can derail the creative process. I feel like I enjoy the creative process of actually recording and designing the sounds more than the implementing of them into the game so maybe I should try and focus more down the Foley/Sound Designer path that games audio itself.