When Your Minimum Viable Product Isn’t Minimal Enough
So my last Studio Diary entry was 3 months ago, and a lot has happened.
I switched from Unity to Unreal as my primary platform (for several reasons, which I outlined previously) — this was both easier and faster than I expected. Having a clear understanding of many foundational principles mainly meant having to translate the Unity paradigms to the Unreal equivalents, and then build upon them. I had been trying to recreate a Mace Windu Clone Wars 2D short in 3D, and I was able to get to the same (somewhat basic) level of progress in Unreal relatively quickly.
After faffing around for several weeks with YouTube tutorials and Googling for all my questions, I bit the bullet and decided to pay for a professional training course to speed up my Unreal learning. After some research I settled on CGSpectrum’s Unreal Connectors Virtual Production 12 week intensive, which gives you several hours of on-demand training, as well as an hour every day with either a TA or a highly experienced mentor for a self-driven Q&A session.
Being able to ask directed questions and problem-solve in real-time is a huge time-saver, and having frequent access to virtual filmmaking gurus like Deepak Chetty and William Faucher is a complete game-changer. The normal version of this course is spread over 12 months, with once-a-week check-ins — this is compressed into a quarter of that time with daily check-ins. It is super intense, but I love it and highly recommend it if you are in a similar situation.
The third big change was a result of this sped-up learning and access to experts: the realization that I had clearly bitten off more than I could chew (for now) with the Mace short. I thought I had whittled my technical proof-of-concept down to a pretty minimal vertical slice, but the more I learned, the more I realized it was still pretty huge.
Part of it was my stage of learning — so many basic technical things to figure out meant that concentrating on the artistic choices was basically impossible. Another part of it was how much intense custom character animation is required in the Mace short — I’d known this was the case, but I hadn’t realized just how complex this particular rabbit hole would be. And finally, I just wasn’t feeling a sense of progress — a week would pass and I’d have spent a lot of time figuring out technical glitches and learning, but did not feel like the overall project was moving.
So I decided to put Mace on hold (I still intend to go back and finish it!) and work on something simpler that I could “ship” in a shorter timeframe. I settled on the idea of re-making the iconic opening of the original Top Gun, except using a space setting and X-Wings. Just trying to make up for that Rogue Squadron movie that never happened.
Watching the original for reference, it was fascinating to see the layers of narrative embedded in what might just look like flashy action cuts on the surface. The tight close shots were probably a result of practical shooting restrictions on the deck of an aircraft carrier, but those crops are used to create really beautiful and impressionistic compositions of humans dwarfed by machines. The sequencing is a symphony of equipment and competence porn: huge, powerful machines being operated by highly trained and coordinated teams working in perfect harmony. The cut from one iconic ’80s track (Harold Faltemeyer’s Top Gun Anthem) to another (Kenny Loggins’ Danger Zone) never fails to get the blood pumping. No wonder they recreated it almost shot-for-shot in the new movie.
Never one to learn from my mistakes, I figured I’d be able to get the whole thing done in 3 weeks. It’s now been 4 weeks working on it, and I’ve gotten about 10 shots out of 66 done. To be fair, it’s not a linear process. I spent 2.5 weeks just setting up the assets and environments, and figuring out a workflow. I’ve only spent about a week and a half “shooting” — but naturally have to stop and figure out technical stuff every now and then too in between shots. But I am feeling that all-important sense of progress and momentum which keeps the excitement going.
There’s a lot to talk about here — from the importance of reference (what’s in your head is not what it actually looks like) to the necessity of labeling and organization (which I only started taking seriously when trying to sort FloorPlate_03 from FloorPlate_96) to the inevitability of pivots and changes (I’d originally planned to use the Disney Infinity X-Wing pilot model, but it was much too cartoony for the environments and more realistic X-Wing models that I wanted to use — so I had to repurpose the characters from the BigMediumSmall Mech Squad kit).
I’ll get into all this in more detail in the next instalment. Until then I’m going to be trying to crank out 2–3 shots per day and trying to get this done in the next 3–4 weeks. If you want to follow my progress more closely, I’m posting dailies pretty regularly to @locodrome on Instagram.
