Once Upon a Time on a Dark and Stormy Night

The video club of which I’m a part has been working lately on finishing some of our first-year projects. One of them, entitled Once Upon a Time on a Dark and Stormy Night, is worth mentioning simply because I’ve invested such a large amount of time in it recently.

In our first year, we adopted a certain operation for our club. Each month, a new member would step up and offer up their own idea for a project. That person would often take on the role of writer/director and the group would rally around them and offer whatever help they could on the day of the shoot. Other responsibilities, especially those that require work outside of shoot dates, would invariably arise, but essentially, participation for most people would amount to an evening planning meeting and a daylong production schedule.

Every project will also have the typical post-production editing, audio, and music related tasks associated with it. What makes Once Upon a Time on a Dark and Stormy Night different is the amount of time we’re putting into extra effects.

Amelia Jenkins is the proud owner of Once Upon a Time on a Dark and Stormy Night and from the beginning she visualized it as animated. We discussed doing stop-frame animation, claymation, and the like, but ended up discarding those ideas because we suspected that they would be too time consuming. I suggested that we plan a normal shoot and use a combination of digital effects in post-production to achieve the animated look-and-feel she was after. Although I haven’t seen the movie Waking Life, I did read up online about how the director supervised the post-processing of the film. I was confident that, after the shoot was behind us, we could find something that would work.

The script for Once Upon a Time on a Dark and Stormy Night (like every video project we’ve tackled so far) is delightfully unique. Clocking in at only three minutes or so, it’s a brief dialog between a barista and her customer… conducted solely in spoken clichés. Because it’s hard to take people who speak this way seriously, the project really does lend itself well to some style of animation.

Unfortunately, this was the only shoot that I was unable to attend. Well, that’s not exactly true. I briefly visited the set while the crew was still ramping up. I had just returned from Cuba, had only five or six hours of sleep in the last three days, and was suffering from severe jet lag. Since I felt it was far more likely that I would curl up in the corner and go to sleep rather than offer any constructive help, I opted to go home and sleep through the rest of the day. I was just too tired to feel more than a tiny twinge of guilt at leaving a skeleton crew to manage everything.

For this project, Amelia brought in two of her talented friends from outside our club. Edra and Becky were perfect choices for the roles in Once Upon a Time on a Dark and Stormy Night as they were able to take amazingly difficult lines and deliver them in such a way as to make them almost sound as if they were having a real conversation. When I finally saw the resulting rough cut of the footage, it was packed with wonderful deliveries of witty lines and truly encouraged multiple viewings.

Before Amelia had even finished editing the rough cut, however, I obtained some of the footage and began testing the animation. In Premiere I selected some representative scenes and exported them as .BMP image sequences. Once the video was rendered as individual pictures, I imported them into Photoshop and began experimenting with filters, adjustments, etc. Once I found a promising combination of effects, I’d create a Photoshop Action and run the whole batch of images through it. When finished, I’d have a new directory filled with altered images, which I could then import back into Premiere in order to combine them back into a video clip.

Amelia and I viewed about a dozen or so such variations (with the input of other video club members, naturally), but nothing jumped out at us as the obvious choice. Most of the images looked great as stills in Photoshop, but when played back as video they broke down. Some flickered too much. Others removed too much detail. Our excitement would climb when we discovered one that was fantastic on the background… only to fall back when we realized how horrible it looked on the actors. It was a real struggle trying to find a single effect that would work for every scene.

One evening, I was experimenting with a combination of Accented Edges and Glowing Lines. I accidentally rendered out a sample video with only the Glowing Lines filter applied and everyone commented on how much they liked that particular look. To be honest, I had to agree – it was one of the best choices we’d seen so far – but I’ve seen Photoshop’s Glowing Lines filter used far too often and I told Amelia that I thought its overuse has turned it into a cliché.

Cliché. Once Upon a Time on a Dark and Stormy Night is all about clichés! When we remembered that, we just had to use the Glowing Lines!

That night I separated the whole three minute video into its 5,400+ frames. While I slept, Photoshop ran a batch Action the whole lot and the next day I awoke to 4.86 Gigabytes of new images with glowing lines. I just had enough time before work to import them into a memory-starved Premiere and start the process of combining them back into a video. When I returned home, the process had finished, but it took me a full 45 minutes simply to close the program. My poor machine had valiantly managed to churn through almost five Gigabytes of data with only 512 Megabytes of RAM, but it sure let me know it didn’t appreciate doing it.

Later that week I previewed the new Glowing Lines animation to the same group of people. I believe we were all in agreement that some parts were phenomenal – the café background, Edra’s hair, and the credit card that exchanges hands, for instance. Unfortunately, the actors’ faces devolved into a complicated mess of lines and the results were distracting, to say the least. The project’s strongest point was the dialog, true, but without recognizable characters to deliver them, we thought that the attention might be taken away from where it was needed most.

Wanting to learn more about After Effects, I offered to use the following week for experimentation. I learned enough about mattes to get me started, then set up a composition with both the original video and the Glowing Lines version included. I set about animating a masked-out section of each actor’s face. After finishing the first five scenes, I realized that this was going to be far more work that I’d imagined. At the rate I was progressing, I calculated that it would take me at least 18 hours to get through all three minutes of the video. Still, I couldn’t come up with a better strategy that would retain the amazing, glowing line scenery, while at the same time giving back the “character” the same effect appeared to steal from our talent.

I previewed the five scenes and the results were… better. The only problem was that the women’s faces floating on top of their Glowing Line’d heads came across decidedly creepy. After a minute or two, we speculated, the audience would probably get used to it, but it was so distracting at the beginning that we worried no one would pay any attention to the all-important dialog being delivered.

Amelia and I sat down to discuss our options once again. We thought that an ambitious plan to mask out the actors’ entire bodies would give us the effect we were looking for, but there was no way to estimate how much more labor that would be. It was bound to be brainless, crank-the-music-and-think-about-other-things work, though, and I was willing to put in the time if it was the effect that Amelia really wanted. Finally, we agreed, and with her vow to lend a hand with the digital matting process over the next few weeks, we got to work.

And that’s what I’ve been doing for the past month. After work, on the weekends, despite the sunny weather, I’ve been putting in hour after hour on this project. Combined, I’d say Amelia and I have put in close to sixty hours on just this aspect of the post-production alone. It’s been difficult at times to convince myself (and her!) that it would all be worth it, but we persevered.

At last, this Tuesday morning, thirty-some masks and untold keyframe- and control point-manipulations behind us, I found myself completing the last scene. That evening, when our video club convened (ironically to begin planning for Amelia’s next, larger-scale project), I showed off the results of our labor.

Surprisingly, my 27-inch television was more forgiving with the mask edges than the tiny After Effects composition window I’d been using on my 19-inch computer monitor. Despite the lack of fine-tuning that still needs to be completed, we were quite satisfied with the results. It was very difficult for me to appreciate the effect any longer since I’d been staring at this video for so long, but it was obvious in the comments of the others: we’d come away with something great.

I’m still going to make a final pass through the project to clean up some of the mistakes we made, plus there are some background audio issues that need to be sorted out. But the bulk of the work is done and I feel like I can relax knowing that we’ll have another video that everyone involved can be proud of.

I can’t wait to see an audience’s reaction to Once Upon a Time on a Dark and Stormy Night, either at the upcoming summer JUMP Society festival or at the next Panhandle Picture Show!