Sound Replacement Post Mortem

My first major project for AUD211 was a sound replacement task for the trailer of the Netflix series, Castlevania. The project has finally come to an end as we’ve sent out our deliverables after a making some last minute changes and additions - and are beginning to move onto our next projects. It’s an important process for me to reflect on the production as a whole, including what went well and what didn’t, and how I can improve on this type of project next time.

I formed a group with Tom Meo, Justin Dando and myself and after some back and forwards between what media clip we would choose, we ended up picking a trailer for the second season of the animated TV Series, Castlevania.

A couple of important things happened throughout the course of the project, but not many unexpected things happened. An significant event in the project would have to be the first recording session. This was kind of planned last minute as a place to start, we only expected to get a couple of the basic foley sounds such as footsteps and other general things. This wasn't the case as after we figured out the workflow of the C24 desk, we ended up smashing out about 75% of the foley assets, or at least the base sounds of more complicated ones. What was originally a 4 hour studio session turned into 6 hours, however Justin had to leave after 4 hours so Tom and I ended up filling out some more sounds. We then booked out another studio session for later that week in which we finished off some more foley sounds, and created some more abstract sounds to use as a base for some of the sound effects.

Another important change was the change of roles, instead of me doing the mixing and tom doing the musical elements it kind of switched around the opposite way. I didn’t really have that much part in mixing the final foley assets and was primarily on the soundtrack and atmospheric elements, whereas Tom focussed on creating sound effects and mixing the foley.

The project as a whole was filled with many positives and very few negatives. I’d say the overall turnout of the project was as good as I had hoped, and I am very happy with the end result. I believe that we all worked well in our areas and were efficient with the time we had. I was originally expecting the project to be a little flawed in some areas due to the large, detailed scale of it, but I am pleasantly surprised at the outcome and how well it replicates the original, and in some areas even betters it.

That being said, of course there are usually some negative aspects to be found somewhere in a project. I personally found that perhaps the communication between the group could have been a lot more clear in some areas, for example knowing and confirming who was doing what and when, other information like that. For example there were some instances where it was unclear who was doing what part in the workload, in one case I had started mixing the track during our spare session one day, but somebody else ended up started mixing the session too so the work that I did wasn’t actually used and I could have spent the time being efficient in another area of the project. For the most part this wasn’t a huge deal, but it created some unnecessary back and forwards of things that were wasting time. I also felt sometimes that that I might not have been doing as as much work as the others in the group, but when asking if there is anything I can do there wasn’t really any tasks allocated for me, so perhaps the work distribution could be more evened out next time.

Recording wise, I would still stick with the microphones we used in the recording, mostly in the use of the RODE NTG2 shotgun microphone. From my experiences I would definitely stick to this mic or experiment with similar shotgun mics for foley recording. We also used the C414 for some other sounds that were a little more broad, and I found this microphone to work very poorly in regards to getting any type of clean recording. I’m unsure if this is because maybe we were using this microphone incorrectly, but it seemed to have a really loud noise floor and just wasn’t clean sounding at all. I don’t really see a purpose for this in my productions as even when recording the ADR, it still picked up a lot of room noise, and I prefer to usually opt out for a clean isolated recording with effects applied in post.

Creating the plan in the beginning of the project consisted of sorting out each sound, time-stamping them and placing them into different categories that may use the same type of mic placement, for instance all of our footsteps were put into the same category so that we could get them all done first without having to move the microphone in between each take. This worked very well as we were able to list what sounds we had to do and what had been done, and recognise sounds that would require a bit more work to create.

I’d say that for this project most of my creative input was into the creation of the soundtrack. I wanted to create something similar to the soundtrack of the original trailer, which was a mix of orchestral and synth elements. We found early on in the production that a lot of the trailer was filled by musical cues, which meant it was important that the soundtrack I created was timed correctly with the trailer. For some areas this was easy as once I had figured out the tempo of the original score, I found that it didn’t sway off time and was on the grid. What was more of an issue however was the cutaways in the trailer where a character would speak or something else would happen and the music would stop, these were off the grid and sporadic. This was an inconvenience to work with as I would have to time these properly, and step off the fixed tempo.

I don’t have access to Pro Tools from my own personal computer, which is where the backing track needed to be created due to the use of my third party samples and plugins. This meant that I was using Ableton Live 9 to create my track - Ableton Live has a built in video playback window, but it doesn’t work very well, at all really. It is extremely laggy and makes it so that it’s pretty much useless. I worked around this by using the audio of the video, and having a copy of it open to see and hear where the cutaways are and used this to line up my shots. There were many cases where I would line up the beginning of my music to a transient on the video and try and hit play at the same time to see if it was still lining up correctly. It’s a pretty bad practice, I would never use it for any small and important timing details, but it worked in this scenario.

Towards the end of the soundtrack creation I decided to take the route of including some heavy metal elements to the track to portray an evil, vampiric vibe in the style of the show. I also wanted something that was a bit different than the original trailer. I spent some time trying to make the metal, orchestral and synth elements work cohesively with one another, and think that I pulled this off relatively well in a way where it's not too jarring but still an interesting change up. One of the main ways I did this was by making or manipulating samples of relatively electro sounds, and running them through the Guitar Rig 5 plugin by Native Instruments. I will right a blog soon being more in depth with the creative processes of the soundtrack production.

I’d say the soundtrack worked well with the trailer in the end, however I would like to have spent more time mixing it if I had the time to. It’s something that's more in the background of the trailer so intricate mix details are likely to go unnoticed, however it would have still been good to have more time spent in the mix down process. It would also be worth the shot looking more into mixing tips for film soundtracks, to see if I could apply any of the processes to my own mix.

Overall, I believe that this project was successful and I’m proud of the work we’ve created. Foley recording, and more particularly composition for film is something I have found to be of the more interesting topics so far at SAE. I can definitely see this as a career path and something that I might continue to do in the future. Without further ado here is the final product and how it turned out,

Thanks for reading!