Categories
Journals

Working 9 to 15 August 2015, what a way to make a livin’.

It was a busy week at the dayjob with not a lot of energy left over for movie making. A motion-capture rig arrived on Thursday so I lost a night to getting that set up and working… then recording myself flipping the bird and importing that into Blender. Then slowly extricating myself from the mocap rig again.

Anyway. Movie stuff. The feedback I got from readers (thanks!) indicated the process stuff was interesting and it was preferable to reveal snippets of the story rather than dump it all out there at once.

Scene 5 is almost in the bag. Almost. There’s one specific character reaction that’s holding me up. It’s a pretty key character reaction and it feels distractingly contrived right now. On the day I drew it I wasn’t fussed, I was happy to have something. Now that I look back at it, it doesn’t ring true. The reaction works to move the story along but it doesn’t work in the sense that it breaks the illusion of the character’s emotional consistency.

It’s an important moment so it’s worth spending the time to get it absolutely right – if it clangs, the spell will be weakened if not broken and the audience will get distracted. I do not want this.

Getting this moment right means spending time getting it wrong in different ways and then finally figuring it out, only to wonder if I’ll look back at it a couple of days later and think “nah that doesn’t work either”. Far from being a triumphant Hollywood moment of hitting “save” with a grin on my face, it’s more a case of trepidation on the day and relief a couple of days later when I look back at the work with a clear head and see that it does the job.

This is completely normal.

I worked on Scene 6 a little bit too – just sketching it out in rough on paper. The development story of Scene 6

is an entire blog entry in and of itself, so I’ll leave that for next week. 🙂

See you then!

Categories
Journals

Here you can see Gronky and Pointy standing on a screencap of the actual honest-to-goodness AMITS beat sheet. Darker yellow lines have been storyboarded, lighter yellow lines have been partly storyboarded, white lines have not been storyboarded yet. With that, here’s the project report for 10 to 16 November.

tl;dr

  • No more storyboards this week, just rewriting and investigating C64 stylings.
  • Story development is overall taking longer than I thought it would but a big part of the task is finding a good story-making workflow.
  • Nobody seems to talk about story workflow in sufficient detail!
  • An abandoned animatic is now available for viewingincluding translations into French and “Frisian”. It’s a bit shit.
  • Last Tumblr update for a while, at least until this round of story dev is finished

It’s strange to find a public knowledge gap around this part of the process. No book I’ve read and no website I’ve come across really goes into the actual nuts and bolts methodology of how an animation director can take an animated film from script to storyboards to ready-for-production. Nothing I’ve come across seems to go into detail about building and working with animatics. The literature I’ve amassed so far is super keen on the 12 principles, character design, pitching, etc – the nice-to-look-at bits. What to expect when refining the story? Not so much, and definitely not in the kind of detail I’d like to know about.

How’s the story process even done, then? Here’s what I’ve pieced together.

I started with a synopsis. Then I threw it away and wrote another one without as much stuff in it. I liked that one better, so I cut the synopsis up into “beats” and put those beats into a spreadsheet, one per line. I drew the beats in the spreadsheet as seen through a camera (storyboards), then put them all together in an animatic. It was a start, but definitely not as good as I wanted it to be.

I bounced back and forth a few times between spreadsheet and storyboard in between other production responsibilities and gradually built up another animatic. I liked it, even.

After showing the animatic to Mathieu at Blender Conf’s open day and getting some advice, I knew I had to change it drastically. It wasn’t just the material I was producing that was falling short, the methodology I was producing it with had been (rightly) questioned as well. This got me thinking: OK, how can I organise these storyboard images in such a way that allows me to alter it. Mathieu’s advice to me about not locking anything to music too early also had an implication

We’ve all seen storyboard walls. They make a kind of sense. Read from left to right then top to bottom. OK. Cool. But that’s a snapshot of the process. You can put things on the wall or take them off. That’s how you iterate.

When it comes to naming files, which is one of the things I’ve been struggling with of late as a relatively minor production concern, it’s less simple than it first appears.

Let’s use the current Blender Open Movie as an example. The production is open enough that Blender Cloud subscribers can directly browse the movie’s repository online and pull down storyboard images, animatic files, concept art, etc. It’s nice to see but it also gives curious filmmakers a look into the brass tacks workflow of a production.

From Blender Cloud, you can find Gooseberry storyboard images in repo/pre/artwork/storyboard/ and you’ll see the boards separated into scenes.. sc01-opening, etc. There’s numbers in the actual filenames.. some of them.. and they’re definitely broken down by scene.. but some are all numbers (13-04), some have letters (C_01), some have letters and numbers (M_01-04), numbered letters (A2-02), some are marked with bis, French for again (06-bis-02), some are marked RTK (RTK02-04)… it’s hard to know what actually goes where based on the filename alone. At the macro level, It’s a mess.

Maybe there’s more clues in the animatic where the images are put together. And yes, it turns out there’s a vague sense of sequentiality from image to image on a very small level. 27-03, 27-04, 28-01, 27-03, 27-04.. 50-04, 50-01, 52-01, 51-03, 51-04, 52-02, 52-03, 52-04.. what? What happened to everything between 28 and 50?

Who knows. This is but a snapshot of an on-going process. It may have been neatly organised at one point with respect to file naming but the needs of story reiteration outweighed that.

It goes to illustrate something that my production designer friend Grant said to me yesterday: story is a chaotic process and necessarily so. It’s a mad science lab where the raw material of the project is constantly being re-examined and rewritten and re-tested and rejigged and reiterated.

That said, what happens in story stays in story. Once the animatic or 3D previz is completely sexy and awesome and amazing and ready for launch, the animatic/previz is treated as a blueprint for the production pipeline and resegmented appropriately into sequence prefix, scene number, shot number. It’s a hard transition point. Every other hacky non-convention used during pre-pro is necessarily discarded. It’s no longer needed.

So, my takeaways for the week:

  • Story development takes a while.
  • The chaos of story development is something to be embraced: avoid anything overly organisational that disrupts the flow of creativity.
  • Any given storyboard is not guaranteed to be permanent, so numbering schemes should be considered even less so.
  • Brass tacks: storyboard without the expectation that the boarded “shots” will be sequenced in the same order that they’re numbered. Got 44 shots? The next shot, even it’s an insert between two existing shots at the beginning of the piece, is shot 45. It is emphatically not shot 2A. Do not waste time pissing around with clever yet superfluous naming conventions. Record where the shot fits on a master spreadsheet for future convenience and get it into place as part of the animatic. Then do the next shot.
  • Find people and talk to them about their workflow. They know stuff!
  • To reiterate, story development takes a while.  But it should.

Since I’ve made enough progress to put it behind me, I’m publically posting the broken old animatic with rough score. This is what I showed to Mathieu. There’s transcriptions of all my messy handwriting and a couple of translations – one in French, and one to follow the customary Blender tradition of hiding comedy Australian subtitles in the “Frisian” track a la Sintel.

And now to business – these updates and other process opening work take a very long time. In future they may come less frequently since there’s not that much to say from week to week while I’m doing story development. Since I’m not seeing much traffic (but big ups to the two three people who are visibly and regularly showing interest in what I’m up to), I’ll hold big round-ups back for when there’s something meatier to talk about. You can always get in touch on Twitter as quollism​ if you’re curious about where the movie’s at, and I’ll drop a few things on this tumblr if they’re quick enough to drop.

Until then, thanks for reading!