Doom Eternal - Archville Glory Kill Sound Re-Design

I played through Doom Eternal's Ancient Gods Part 2 DLC recently, and had a smashing time - that sentinel hammer is such a satisfying weapon to bonk demons on the head with! Firing the game up again inspired me to go back and add some more details to an Archville glory kill re-design I made towards the end of last year.

Here’s the re-designed clip itself, and a video breakdown of the different sound layers that went into it:

I’m fairly happy with how this clip turned out, but I did struggle a bit with managing the mix. As there's so many big and hyperrealistic elements all fighting for the listener's attention at once (creature roar, fire magic, combat impacts, gore etc.) I found it tricky to get the balance right between these exciting but competing parts of the soundscape.⠀
⠀
In the end, I decided to downplay the Archville's flame sounds in favour of focusing on its roar, as that felt like the most visually important element of the clip. I'm not 100% happy with the roar sound, as I don’t think I quite managed to get the individual parts (metal scrapes, alligator growls, tiger, lion & elephant roars) working together as one singular & cohesive voice as I'd intended. Nonetheless, I think the roar sound I ended up with still works with the visuals, and it was an enjoyable part of the design to iterate upon.⠀
⠀
I then used Waves Factory’s Trackspacer plug-in to duck down the volume of the Archville layers when Doomguy performs the grab and stab moves, which helped keep those impacts feeling beefy and punchy on top of a loud and dense soundscape.⠀
⠀
Overall though, I'm pleased with the end result of this re-design. I had a lot of daft fun chipping away at it, and the process of putting it together has been helpful in identifying areas in my mixing workflow that I need to continue to work on.⠀
⠀
KAR EN TUK!

The Spooky Jar - Kirsten McDowell

Boo!

I recently finished up some sound design work for my good friend and environment artist Kirsten McDowell on her Spooky Jar project, so here’s a quick post now that the project’s wrapped.

The Spooky Jar is an environment art scene that Kirsten made for her final year honours project at Futureworks. The bar’s design is inspired by the The Whiskey Jar, a whiskey bar (who’d have guessed?) in Manchester’s Northern Quarter. Unlike the Whiskey Jar however, The Spooky Jar is home to some cheeky ghosts and ghoulies who have come out to play, and presumably neck a few drinks along the way!

You can see the scene in full detail and get more info on the design process by heading over to Kirsten’s ArtStation page to check out the bar and its 3D-modelled props in all their glory. Plus, while you’re in a link-clicking mood, you should also follow her on Twitter too so you can keep up with the rest of her environment and game design work as well!

Here’s a flythrough video showcase of the scene:

I worked on the sound design for this video showcase, which was a lot of fun! The brief was to create a spooky yet cosy vibe for the piece through area (non-positional) and source (positional) ambience sounds.

Most of the sounds I designed were sourced from my sound effects library, but I did get chance to do some field recording for a few choice elements in the scene. For example, to record the ouija board glass movement sounds, I grabbed my trusty (but hissy) H4n recorder, a wooden chopping board and a nice, weighty glass tumbler and had a good ol’ scrape about! After chucking the recordings into Reaper and applying a bit of EQ and Brusfri to remove top-end hiss and room tone, I had this:

I also recorded this creaky, crackly book which I intended to try and use for some of the dusty paper menus and posters in the scene.

However, although these crunchy book sounds were cool, I felt they didn’t work too well with my audio direction for the scene, and that they sounded a bit too over the top and out of place for the I overall mood I was aiming for, so I took them out. Not to worry though - I’ll no doubt go back to this recording for the next project that needs some extra crunchy ice sound design I think!

As I was mixing my sounds to a linear piece of media, and not implementing them directly into the UE4 project, I decided to replicate the feel of the audio being played back by the game engine by keeping a lot of the positional source sounds (door creaks, neon sign buzzes etc.) mono. This way, I could pan them around the environment according to the position of the camera, to give the feel the player is actively moving around the scene in engine in real-time.

View fullsize

Here’s a shot from my Reaper session where I’ve used track panning to position diegetic source sounds in the scene in relation to the position of the camera. In this sequence of shots (02:15 - 02:28) the camera gradually moves from left to right through the environment, so I positioned the creaks and closing sounds of the door in this shot according to where an in-game listener on the camera would likely pick them up.

Anyway, it’s coming up to closing time on this post, so thanks for having a look and listen to the sounds of The Spooky Jar - it was a short but sweet project that was great to work on! Stay safe out there folks! :)

May 2018 Sound Design Showreel

Made my first sound design reel! Picked a few sections from the various projects I worked on during the past year of my studies that show my implementation and sound design skills. It's rough and needs a lot of work, but hey for my first reel I'm actually quite pleased with how it turned out.

Sound Design MU2411 - Week 14 (12/12/17)

Assignment 1 Submission

““...and there I was, in the final hour.””

— Abe (Oddworld: New 'n' Tasty)

So, the big day finally arrived. I submitted my assignment by 10:21pm, making use of the online submission process to buy myself a few more precious hours to mix in the D-Control 5.1 studio.

Objective complete!

Those final few hours in the D-Control made all the difference...I hope!

Stemming Down

I used a Pro Tools HD iLok to make my final stems without having to delete tracks from my already maxed-out track count.

Final Export and Delivery

To complete my cinematic audio redesign, I bounced out the audio from my master mix session as a stereo audio file, and then added it into a new Pro Tools session containing the video file. This way, I was able to massively reduce the strain on my computer's CPU during the audio and video bounce process as all the main audio processing had been taken care of in the first audio-only bounce. Then, I bounced the audio and video to a QuickTime file, which I subsequently loaded up in MPEG Streamclip to convert the video codec from Avid DNxHD to H.264.

Limiting Issues

As I had to do the final audio bounce on my laptop (in order to make use of the some specific Kilohearts plug-ins), I couldn't use of Avid's Pro Limiter to do my final audio bounce. As an alternative, I used the Maxim plug-in on the Master output with the ceiling value set to -2.0dB.

On previous projects, I've usually set the Maxim ceiling value much higher (e.g. -0.3dB/-0.4dB), but after learning about inter-sample peaks, I decided to give myself more room for error and set it to -2.0dB as Maxim is not a true-peak limiter. 

Listening back however (and looking at the final waveform) the sound in he really loud sections of the trailer gets chopped off when it hits the limiter. Ideally, I would have set the ceiling higher to prevent the 'squashing' effect, but I thought it would be a lesser evil than having big transient spikes going through a higher ceiling that could potentially clip the digital 0dB ceiling.

LUFS Issues

Final LUFS reading I took in the D-Control 5.1 studio. I subsequently dropped global clip gain by aprox 3-4dB after taking this reading, so hopefully my submitted video falls within the EB -23dB acceptable range.

DEM Stem Problems

It was my intention to include separate DEM stems alongside the main bounce as part and parcel of a professional delivery package. However, despite setting up and assigning the necessary buses and routing in Pro Tools, I ran out of time to get this done to a satisfactory level that I was happy with submitting. The main problem was simply an issue of volume - the stems were being routed properly, but they were far too loud and massively clipping during the printing process. In hindsight, I probably should have just stuck a hard limiter on each individual stem to stop them going into the red, but as this was just before the deadline I was in a bit of a panic, so I decided to focus on just getting the essentials done for the submission.

For future projects however, I'll definitely start thinking about stemming earlier on the process, as it would have streamlined my overall mixing process a great deal, and would possibly have bought me more time to mix to a higher standard in the final few weeks of the project.

During the final mix, I set up dedicated individual stems for DEM and VO stems, but I ran out of time to get these done to a standard I was happy submitting before the looking deadline. Next time!