Released to theaters this past summer, the movie “Kingsglaive: Final Fantasy XV” created quite a stir. The movie is done entirely in computer graphics (CG) and is set in the same world and time and uses the same characters as the video game of the same name. The latest, eagerly anticipated installment of the Final Fantasy video game series, “Final Fantasy XV,” was launched yesterday, November 29, 2016.
Sound production was primarily created using multiple Pro Tools | HDX systems and the latest Pro Tools 12 software. Creation of sound effects, foley work, music recording, pre-dubbing work that was done in Los Angeles and the final mixing at Toho Studios in Japan were all done using Pro Tools.
ICON interviewed two key sound production staff from the production: Supervising Sound Editor Atsushi Suganuma from Square Enix and Hiroshi Kasamatsu from Digital Circus, in charge of foley creation and final mixing, and asked them about the concept and workflow for the sounds throughout the film.
“Kingsglaive: Final Fantasy XV” – a side story for “Final Fantasy XV” created in full CG
ICON: “Kingsglaive: Final Fantasy XV” is an entirely CG movie based on the smash hit “Final Fantasy” game series, but to what extent is the story in the movie connected to the game?
Kasamatsu: The latest game, “Final Fantasy XV”, will be released simultaneously worldwide on November 29th, and the stories are completely connected. To put it simply, “Kingsglaive: Final Fantasy XV” tells the story of events that are happening in the same world and at the same time as “Final Fantasy XV”, but in a different location. The story in “Final Fantasy XV” is told from the perspective of the main character who is a prince, while “Kingsglaive: Final Fantasy XV” is told from the perspective of King Regis, the prince’s father, so the two stories are linked together like father and child. Of course, the movie can be enjoyed on its own, but I think that people who play the game will be able to more fully enjoy the world of “Final Fantasy.”
“Final Fantasy” is a series that’s been ongoing since 1987, and while there’s lots of stalwart fans around the world, we also wanted people who haven’t played the games to learn more about this setting too. And that was the thinking behind the creation of “Kingsglaive: Final Fantasy XV”. By creating it in full CG, we wanted to let people learn about the world as it exists in the game, as much as possible.
ICON: Mr. Suganuma and Mr. Kasamatsu, can you tell us about your roles in the production of this film?
Suganuma: I worked as the Supervising Sound Editor, overseeing the total direction for all sound. In terms of the “Final Fantasy” series, I first started on “Final Fantasy XIII” in 2009, and I also worked on “Final Fantasy XIII-2” in 2011 and “Lightning Returns Final Fantasy XIII” in 2013. And, of course, I also worked on the latest version, “Final Fantasy XV”.
Kasamatsu: I supervised all foley recording and the final mixing. This is the first “Final Fantasy” production I’ve been directly involved with, and I did not work on the “Final Fantasy XV” game that will be coming out soon but our studio, Digital Circus, did the foley recording for the previous movie project “Final Fantasy VII: Advent Children”.
ICON: Could you tell us a little more about the job of Supervising Sound Editor, please?
Suganuma: That’s a title that’s often used in Hollywood, but it is the director for all of the non-music sounds. It is pretty much the same role as the Sound Director in game production. Working with the director, you decide on the direction for the non-music sounds, such as sound effects and dialogue, then carry out the work to create that. In regards to the music, although I’m not involved with the details of the musical pieces themselves, I may also work on the mixing, to avoid any acoustic overlap or interference.
#1: Creation of sound effects — Layering multiple sound elements to create unreal sounds
ICON: When it was time to create the sound effects, what kind of requests did the director, Takeshi Nozue, have for you?
Suganuma: He said that he wanted us to make it identical to the game. He said that since the story takes place in the same world as the game, he wanted us to use the exact same sound elements, and create the identical world-view. In addition, since it had been decided that this film would be distributed not only in Japan, but overseas as well, world-class sound design was needed. In other words, it was necessary to set the bar even higher than normal. This was a big challenge for me.
ICON: There were also presentations of concrete concepts, weren’t there?
Suganuma: Director Nozue told us that he wanted the sounds to have as much “credibility” as possible. For example, there are things that aren’t actually real that appear in the story, like monsters or magic. But, he said that he wanted the sounds that those imaginary things make to seem like they might actually be real. As a result, we needed to have an approach to sound design that was totally different from anything before. For the sound effects, we weren’t able to use any of the materials that had been used in the previous “Final Fantasy” series. The “Final Fantasy” series has a long history, and although there was a massive library of assets, we had to create all of the sound effects for this film from scratch.
ICON: Where were the sound effects created?
Suganuma: Normally, we’d start the work in a private space in our offices, but because we needed such high quality, this time we worked in our company’s studio right from the start. The studio has a Pro Tools HDX system, and the software versions were 10 and 12. Sound effects are usually created as monaural, and then upgraded to 5.1ch in the mixing. The session format was 24bit/48kHz.
ICON: When you started work, how far along were the visuals?
Suganuma: Almost nothing had been completed so we started working based on some simple VFX images and some imagination. It took about 8 months to create the sound effects, and in that time, the visuals were gradually completed. Of course, there were plenty of sounds that needed to be reworked to match the visuals, and the work was pretty tough. For the video, we monitored it by reading it into the Pro Tools video track, and then outputting it via HDMI to a multi-display.
ICON: Please tell us the details of the sound effect creation workflow.
Suganuma: We started by gathering up a wide variety of sound materials, and then investigating what kind of sounds they would suit. For example, for the Thunder magic spell, which is well known in the Final Fantasy series, it was decided that in this movie it will come from the hands. We had Director Nozue explain the thinking behind why the Thunder comes from the hands, and then based on that, we investigated what would best express that. Obviously thunder doesn’t normally come from hands, so first we had to think about how you could make thunder with your hands. If your hands actually produced thunder, wouldn’t the sound be so loud, you’d be deafened? Wouldn’t there be an impact like a giant explosion? Wouldn’t the skin and nails on your hands rip apart due to the thunder clap? If the skin and nails are burned and blistered, what kind of sound would that be? The images keep gradually expanding like this. So you combine the sound elements that you’d recorded on your own to embody those images. You might mix in the sound of hair burning to the sound of real thunder; you create the sound by adding or subtracting various elements. You work to balance reality and fantasy to meet the director’s requests.
ICON: So you create the sounds by importing the sound elements into the Pro Tools audio track and then layering them?
Suganuma: That’s right. Layering is the basis of creating sound.
ICON: How many elements do you layer?
Suganuma: Naturally, it varies according to the sound. As an example, for the sound of the rocks and rubble falling that was used in the battle scene at the beginning, we wanted to create a sound that was bigger and louder than reality, so the sound design was aggressive. In concrete terms, the sound for one rock was the combination of 10 tracks of sound elements, and then we’d mix in other elements, such as the sound of actual rocks breaking and falling. That kind of sound creation uses a lot of tracks.
ICON: You were probably using a lot of plug-ins for the sound creation?
Suganuma: No. For me, I almost never do any processing with plug-ins. I will do some basic processing such as level changes or a time stretch, but I prefer to use the sound in the original element as much as possible. Particularly for this project, we were looking for sounds that could really exist, rather than created sounds, so we used elements that were processed as little as possible. There are a lot of people who use plug-ins when they’re creating sound effects, but personally, I really don’t like that approach. Even when using a plug-in is unavoidable, I avoid using third party ones, based on problems with inter-studio compatibility. When I did use a plug-in, it was just Audio Ease Altiverb.
ICON: For the elements you used, were they mostly ones that you had recorded yourself?
Suganuma: As much as possible, I used elements I had recorded myself, but for sounds that are difficult to record, such as something burning, then I would use a commercial library. I think it was a 50/50 split between what I had recorded myself and material for commercial libraries.
ICON: You didn’t use software instruments or similar?
Suganuma: I did use Native Instruments Kontakt as a sampler, but beyond that I pretty much didn’t use one. With Kontakt, you can play back sound elements that you recorded on your own, not just the existing library. It is really useful because pitch and other factors can be controlled via MIDI, and you can use the keyboard when you to create layers for a sound.
ICON: How many sound effects in total were created for this project?
Suganuma: There’s too many to remember (laughs). We had to rework a lot of them, and I think maybe we made enough sounds for about 3 times the actual length of the movie!
ICON: What sound did you work particularly hard on?
Suganuma: I had to remake the cry of an octopus-like monster called Orthros several times, so that one really stands out in my mind. That one was rejected so many times – I really struggled with it. I was initially using a variety of animal sounds, and it took multiple retakes until I created a sound that matched what the director wanted. As well, there were lots of other sounds that were initially approved, but when the visuals were finished and we checked the overall flow, we had to do retakes.
#2: Foley production — Recording sound without interruption to capture the full atmosphere
ICON: How did the foley recording proceed?
Kasamatsu: We started the foley work in January of this year in our studio. You can’t start the foley work until after the visuals have been set, so work started after we waited for a while for the images to be fixed. When we talked with the director, we learned there was no need to do anything particularly unusual, so the recording proceeded basically as normal.
ICON: Did you ask a foley artist to do the recording?
Kasamatsu: No. For me, I do the acting myself. I asked one of the staff to handle the operation of Pro Tools, and we worked together as a pair. For the foley recording, I used a Pro Tools | HDX system and Pro Tools | HD MADI as the audio interface, and the AD converter built into the System-5.
ICON: Did it feel pretty elaborate when you were recording the sounds?
Kasamatsu: That’s right. For me, the sound is created to a certain degree by the combination of the mic and pre-amp. For this project, I used a Focusrite ISA 115 HD and a Manley Slam. I mainly used the ISA 115 HD, which I’ve accustomed to using over many years, but when I wanted to try slightly trickier stuff, I chose the Slam. On the ISA 115 HD, it is like the EQ sense is part of my muscle memory, and I just can’t let go of it. For the mic, I really love my Baby Bottle from Blue, which I’ve been using for the past several years. I like that the mic output is high, and the S/N is good, but it is comparatively cheap, so even if I drop it into water, the psychological shock is pretty minor, which is good. (laughs)
ICON: Doesn’t that hamper the dynamics outboard?
Kasamatsu: For the compressor, we used the UREI LA-22. This is a compressor that I saw an engineer using while recording percussion and I really like it, but since the attack is really fast, it is suited to live recording.
ICON: What kind of processing is done after the recording?
Kasamatsu: All I do is some fine corrections, and adding a little bit of ambiance. I rarely use a process that makes big changes to the sound. Even in terms of the levels, for me, I often make the decision when I’m recording.
ICON: Mr. Kasamatsu, what would you say is your focus in foley recording?
Kasamatsu: I think it is an issue of preference, but I really don’t like using sound elements that are intermittent. As an example, for a sequence where a person is walking on gravel and then stops, even when they’re stopped, when they shift their weight, it still makes a sound, doesn’t it? So I want to capture the whole atmosphere, including those kinds of parts, that’s the sort of thing I’ve been trying these past few years.
ICON: What would you say is the most impactful foley work in this project?
Kasamatsu: Earlier, Mr. Suganuma mentioned how it was necessary to make unreal sounds seem like they actually exist, and the foley work was created using this same concept. For example, when a warp occurs, we used particles of magic dropping to the ground to express that, and the sound was created with a stream of individual particles. Also, I think that the sound for Orthros, which Mr. Suganuma mentioned earlier, had a really interesting result. This sort of monster is the first to be created through a combination of foley work, sound effects and voice, and I think you could say it is a crystallization of everyone’s efforts.
#3: Pre-dubbing — Specialists for each field working simultaneously maximized time efficiency
ICON: For this project, Greg P. Russell was credited for the re-recording and mixing, so where was the dubbing carried out?
Kasamatsu: It was at Technicolor in Los Angeles. The materials prepared in Japan were sent to America, and Greg performed the pre-dubbing for the sound effects. Fundamentally, the pre-dubbing was all done in America. That data was taken back to Japan for the final mix, so I performed the final mix myself.
Suganuma: Because Director Kozue had the intention of creating a world class product, after a great deal of consideration, we chose Greg, the top engineer in LA, who has worked on over 200 Hollywood productions. There were a variety of possibilities for the actual studio, but the projects that had been created at Technicolor in the past matched the sound design direction we wanted. We really felt like it had to be there, or not at all.
ICON: Mr. Suganuma, we understand you joined in on the pre-dubbing work at Technicolor. Is that right?
Suganuma: Yes, I did. Due to their schedules, other people weren’t able to go, so I went to America on my own, and I worked with Greg from the middle of April to early May. I guess the work period was about 25 days. This time, we didn’t tell Greg what the work method was, but fundamentally the pre-dubbing proceeded according to his methodology. This was our first time working together, and initially we were sort of flailing around, but the work went really smoothly. We’re from different countries, but we had sound creation in common, and there was absolutely no feeling of being uncomfortable. On the other hand, for the music, the recording was done at the same time at Ocean Way, Nashville. So while we were doing the pre-dubbing work, we’d receive a string of these newly recorded tracks. And for the sound effects, they weren’t all completed when we started the pre-dubbing, so we remade some of them based on the director’s instructions.
ICON: Tell us about the work environment at Technicolor.
Suganuma: The console is a System-5 but because we had to take everything back to Japan to perform the final mix, Greg delivered everything in a format that could be completed in a Pro Tools session. In concrete terms, it was as if the sounds that were mixed on the System-5, were each recorded as individual sessions. Greg didn’t normally have to deliver files that would be completed in Pro Tools, and because it seems like it will become necessary in the future, he said he’d have to study up on that.
ICON: What kind of plug-ins did Greg use for the mixing?
Suganuma: Mainly the Avid Pro series plug-ins, like the Pro Multiband Dynamics and Pro Subharmonic. The use of the Exponential Audio Phoenix Verb plug-in during the mixing really left an impression.
ICON: Was there anything unique about Greg’s mixing?
Suganuma: More than anything technical, the team work was great. Specialists in each field were all doing their own work at the same time. As a result, we were able to make the most efficient use of the time. Greg, his assistant, the dialog mixer, the mix technician who managed the consoles and session layouts, and the sound editor who did the edits on-site when any sound editing was required; these five people were normally working simultaneously to move the work forward.
#4: Final mix — After creating the sound with just the dialog and sound effects, music is added and minute corrections are made
ICON: Tell us about the process after the pre-dubbing was completed and you returned from America.
Kasamatsu: After the pre-dubbing was completed, all of the sound elements were gathered together and the final dubbing was carried out. The final dubbing was carried out on the Toho Studios dubbing stage at Seijo, Tokyo but prior to that, we worked on everything we could here at Studio 2001 in Akebonobashi, Tokyo, and then we took it to Toho Studios.
Suganuma: It was a huge amount, and if we did everything at Toho Studios, it would have taken a lot of work days. By doing the preparation here, we were able to complete the final dubbing at Toho Studios in 4 days.
ICON: Tell us about the system used during the final dubbing.
Kasamatsu: We ran three separate Pro Tools | HDX systems, one for the sound effects, one for the dialog and one for the music, and we used a System-5 as the mixing console. We used Pro Tools | HD MADI for a MADI connection between the Pro Tools | HDX systems and the System-5.
ICON: How were the sound effects, dialog and music organized within Pro Tools?
Suganuma: For the sound effects, they were classified into BG, foley, hard effects, magic and monsters, and the 5.1ch stem was split into 12 groups for output. The music was output as 2 sets of 5.1ch mix composed of 8 stems, but because the stems were split finely, making adjustments later was extremely easy. The dialog was output as 4 groups, with voices processed as principal dialog, crowd noise, non-principal, etc.
Finally, the mix was done on the System-5, but in reality, it was only used as a simple summing mixer. As a result, we could reproduce the same conditions either here, at Studio 2001, or at Toho Studios just by starting up a Pro Tools session. As a result, even for the work at Toho Studios, the console was only used as a summing mixer.
ICON: Would you tell us the details of the mixing process flow, please?
Kasamatsu: For me, after the sound was created first from the dialog and sound effects, once the music was added, we made minute adjustments. We only used standard plug-ins such as multi-band comp and de-esser.
ICON: Were there detailed instructions from the director?
Suganuma: In the overall flow, he would ask “I want the music louder here”, that sort of thing. But, because the director couldn’t come to LA for the pre-dubbing, we did the mix by splitting the stems as finely as possible, so if we had to do a retake, we could do it with a minimum of work. If a change did become necessary, we could make the correction by returning just that stem to pre-dubbing.
ICON: Was there anything you had a hard time with for the final dubbing?
Suganuma: More than having a hard time with anything, I think that there were a lot of things I learned. For this project, the materials were prepared by Japanese people, and the final mix was done by Japanese people, but in the end, we produced something that had a slightly Western feel to it. While on the one hand it was great that the project ended up as a world class movie, just like the director wanted, it was also a little frustrating. (laughs) Whether that was due to the technique of Greg who worked on the pre-dubbing, or some kind of Technicolor studio ‘magic’, I’ve tried to analyze it a lot, but I really can’t figure out the reason. (laughs)
ICON: For this project, you used Pro Tools from beginning to end, was there a feature that was particularly useful?
Kasamatsu: For me, having track commit and track freeze was a big one. When using a plug-in with latency, you’re able to easily output it as audio with the parameter settings still in place, which was really useful.
Suganuma: For this project, work for the dubbing stage was carried out in a variety of locations, at our studio, the Technicolor studio, Mr. Kasamatsu’s studio and Toho Studios, and I think that without Pro Tools it couldn’t have happened. Every studio in the world has Pro Tools, and even if you can’t communicate via words, you can communicate through Pro Tools. When we finished this project, I really felt that if you’re creating a product for the world, then you need Pro Tools.
ICON: Is there a scene that you’d like people who will see this movie to listen particularly closely to?
Suganuma: I’d really like them to watch the 20 minute scene at the beginning. I think it is incredibly immersive. And, they’re using the same elements as the game versions, but I’d like people to really pay attention to the sound effects for the firearms. We used sounds that were recorded from real guns in California. It cost a fair bit to create them – we called in a gun-recording specialist, and I think the sounds are really good.
Kasamatsu: For me, rather than just an individual scene, I want people to listen to the overall quality. I think that it is a movie that was made by Japanese people that also has the feel of a Western film. The music is superb as well.
ICON: Finally, do you have anything to say to young people who might be interested in sound design and mixing for movies?
Suganuma: I think you can say the same thing about making music too, but making sounds that you like, and making sounds for a client’s product as work require completely different skills. This might be the biggest struggle for people who aspire to sound design. And I think that the only way is that you have to build up experience.
Kasamatsu: I think that if you only study the technical aspects, you’re only going to get so far. Beyond that, the feeling and senses that everyone has becomes important. This might be harsh, but just because you studied really hard in school doesn’t mean that you’ll be able to make good sounds. My advice is that getting involved with as many different types of products as possible is the best. And if you want to be a sound designer, don’t just concentrate on sound effects, it is important to listen to all the sounds, like the dialog and music too. In other words, I think you should have your antennae pointing in every direction.