When were the first three videos in the series made?
Robert Wunsch (Director): The very first video we shot in Bangkok, April 2019, at the end of this video we switched into the virtual world to introduce CG to the audience. The second, which was the first full CG video, came out in October 2019. The third video, “In Your Eye,” came out in January 2020, and the last video has been released in March 2021.
What are the main differences between the first three videos and the fourth one?
Robert Wunsch (Director): There are big differences between them. When we started the project over the following two years, a lot of things became possible for the fourth and final video. To start with, we did not have access to those technologies. The differences are very big because, during the project, we had a continuous learning process and things started to become possible like we could have never imagined.
For example, one very simple thing was when we started the project, we did everything with the first two videos using Epic’s Unreal Engine; at that time, it fit our needs and our production scope and requirements the best, and, therefore, we decided to produce the first two CG videos in Unreal Engine.
Especially, the production of the third one made us realize that there are a lot of things that we couldn’t achieve and make possible within that pipeline, so we decided for the last video to move out of Unreal and create the last video in a classic CG pipeline using (Autodesk’s) Maya, (Maxon’s) Cinema 4D, and (OTOY’s) OctaneRenderer, and all the things Mark probably already described. That’s why there are such big differences. Also, we cannot forget about the different time schedules we had. Everything is developing so fast these days, and two years in the CG world is a decade in our normal world.
What types of visual effects are used?
Mark Scott (Generalist): Almost all of the imagery is fully 3D-generated content and has only been enhanced in compositing. We avoided doing much post work on this project, as keeping it all within the 3D software gave us the most flexibility. In a few instances, we have relied on VFX techniques to add particle and dust effects, heat distortion, and atmospherics.
What other tools did you use in the pipeline?
Mark Scott (Generalist): Maxon Cinema 4D was the carrier software package, in which all assets and animations have been collected to be rendered out with Otoy OctaneRenderer. Our character animator used Maya to either fully craft the animation or to enhance (Adobe) Mixamo presets. As for texturing, most of the assets, including the Marvelous Designer clothing, we relied on the (Adobe) Substance Suite. After smoke and particles have either been simulated within (SideFX’s) Houdini or (Janga FX’s) EmberGen, the exports were composited in (Foundry’s) Nuke.
Was this a new pipeline for your studio?
Mark Scott (Generalist): We have built this pipeline from the ground up. It leverages each individual freelancer’s strength within their favorite software package and makes beautifully rendered exports possible. There are some limitations regarding particle and smoke simulation integration, but it works quite flawlessly and predictable.
What made you change from using Unreal Engine in the earlier projects?
Mark Scott (Generalist): The traditional route of using 3D animation packages instead of Unreal offered both stability, security, and high-quality renders. Hair and cloth simulations are not only a possibility now, but an essential. The only true cost of going this way are the expensive and time-consuming renders, which is offset easily by all the positives, though.
Robert Wunsch (Director): At one point we realized that all the benefits the Unreal Engine brought us, there were also many restraints, such as: instability, glitches, bugs. Hair sims and, clothing sims were not possible. The animation tools were not as polished and wouldn’t let us achieve so many things. So, we decided to take this different approach for the fourth one.
What were some other advantages?
Mark Scott (Generalist): The high-quality previz renders in Octane made us skip most of the look development. Within only a few hours, we could lock the look of a whole sequence and push through each sequence in a logically sound order.
Robert Wunsch (Director): The biggest advantage, of course, was that we created a much more polished look — the overall look was way higher end and way more realistic than what we were able to create in the Unreal Engine. Hair and clothing simulations were much better and easier to integrate, and only possible to integrate in this pipeline compared with Unreal.
Was there anything unique about the workflow?
Mark Scott (Generalist): Our hair and cloth simulations and the volumetrics were definitely unusual for a project with such a limited time frame. I guess it is also quite unusual to produce cinematic content with characters and storytelling using Cinema 4D and especially its internal hair tools, which are seen by most artists as lackluster. Hopefully I have proven this sentiment wrong!
Robert Wunsch (Director): Due to the extreme tight schedule we had and the deadline for the last video, we had to do a lot of steps in parallel, whereas you normally do them sequentially. We had to have a very solid pipeline, and we had to be very much on point with the steps we did, because there was absolutely no room for mistakes. Had we made a mistake in one step, which was made parallel to the other, it could have affected all the other steps as well.
What were the most challenging graphics to create and why?
Mark Scott (Generalist): The mirror sequence was the most complex for sure. As the reflections of Toni Garrn are not her, but a manipulated double doing its own thing, I had to split many shots into layers to be rendered separately. It does not seem as complex — but having to manage many different shots and their unique shenanigans proved to be quite a challenge.
How many artists worked on it?
Mark Scott (Generalist): The following artists proved to be centerpieces in our pipeline. Without them, it would not have been possible to push this much content: Robert Wunsch, director; Mark Scott, lead generalist; Daniel Poschinger, generalist ; Max Stöhr, lead character animation; Kiyan Forootan, motion design; Batuhan Perker and Henry Medhurst, simulation; Tom Prendergast, composition; Robert Wunsch, writer; Toni Garrn, Robin Schulz, and Felix Jaehn key cast.
How did the use of CG elevate this project? What was the objective of using CG?
Robert Wunsch (Director): The objective of using CG was that the artist himself doesn’t like to be so much in front of the camera, he doesn’t like to shoot music videos, like any type of things in front of the camera for a very long time. Therefore, the idea was to create a digital avatar for him so we could shoot the music.
Mark Scott (Generalist): As the story takes place in a digital world, disconnected from reality, it was essential to move the production into the virtual space. Using VFX to enhance real footage would have proven difficult, especially considering that many shots are just physically impossible to produce with a camera.
How long did it take to create?
Mark Scott (Generalist): We only had about six weeks to produce the whole thing. I was just finishing up my bachelor’s thesis at the end of February and immediately started working on the video, the same day as my graduation. All the clothing were pretty much completed by then. Due to some technical difficulties with rendering the heavy geometry of the robots in the last scene, we extended the work for a few days.
Robert Wunsch (Director): The whole project as described before started in 2019. After we had the go from the label for the last video, it took us from start to finish, such as the first conceptional phase over preproduction to the finish product, two and a half months, and the production on itself took us six weeks.
What was the secret to doing all this on a limited time frame and budget?
Mark Scott (Generalist): Planning the whole pipeline in advance and removing room for error was essential. Even long before production began, I made sure that all needed assets would translate into Cinema 4D perfectly. A good example is the hair: Only having to set up the groom once per character, I could simply move it onto new character meshes and simulate the shots. If this was not possible, I cannot image how much time would be lost.
Did A Current State do all the work or was some outsourced?
Robert Wunsch (Director): The A Current State and a team of freelancers did all the work. (The team of freelancers is listed above.)
How does this project differ from some of the other work that A Current State has done?
Robert Wunsch (Director): I mean, it’s a CG video, we are not a CG studio. Alot of our projects have some kind of VFX or CG integration, but shooting a whole three-minute music video completely in CG is obviously something way different for us because normally those kinds of projects are only done by big game studios or CG production studios, and is not something a smaller creative studio like us is going or even able to do.
We are a creative studio working in advertisement, the music industry, and we do film, photos and CG. We do design and often use these different elements and bring them together to create a product, but doing a whole CG video and focusing just on creating was nothing we have done before — it’s different from filming or shooting that involves real-life elements. We never did something like this before, and hopefully we will do it again in the future because it was a lot of fun. I think that’s an interesting aspect because before this project , when it came to integrating CG or CGI elements in the stills or film projects, we did mostly simple motion, like 2D motion design, 2D effects, and some simple kind of like 3D animations integrated in our films, but this project was a huge step up for us.