I’ve been quiet the last few months for very good reason. I’ve been generating the media with which DGA will advertise their new product – Zipbuds. We’ve been keeping this under wraps for a good four months (beginning of July) and in that time we’ve had at least six separate photo shoots and have generated close to 100 (91) finished Photoshopped images to be used in ads for the product and on www.Zipbuds.com. In addition to the photos, we put a web commercial together; everything from conception to completion was our responsibility. We even built some of our own tools to get the job done.
Who’s We? Scroll to the end of the post for list of people involved in production (Credits).
For starters, if you haven’t read previous posts, my client is no stranger. I’ve worked with DGA for quite some time now. Since completion this past month, I will have done close to ten photo shoots and three web commercials for the company. Also, for those reading who aren’t affiliated with Valhalla, I currently volunteer/mentor students from Mike Skocko’s Mac Lab. This project was utilized as a part of that collaboration and students were involved in all stages of production.
I’m not sure how to cover the amount of information pertaining to these past four months so don’t hesitate to ask questions, there’s just too much for me to remember in one sitting like this, and I’m sure you’ll notice as you continue reading.
So for clarity’s sake, everything that was generated for Zipbuds was completed in mini projects spread out over a few months, not one continual effort. After receiving the initial request from DGA, each project began with tedious pre-production. This was made up of meetings with students for brainstorming, storyboarding, location scouting, and test shoots.
Part I: The Photos
Photo shoots with models took between one and two weeks for pre-production, one day for the actual shoot, and another week or two for post-production. Pre-production took quite a bit of effort mainly because each pair of models was only provided to us for one six-hour block on each shoot/mini project, meaning we had to make sure that we could hit five locations in one day without any snags or else we wouldn’t finish all the photos planned. Needless to say, there was no room for do-overs. We had to make sure that the driving distance between each location was manageable and plan beforehand how long we would spend taking pictures at each stop. Don’t forget scheduling for favorable weather, wardrobe changes/choices, and props.
On the technical side of things, the Canon 5D Mark II was used for everything. We shoot RAW but even with the flexibility of the format I wanted the students to get in the habit of leaving as little to chance and to Photoshop as possible. When we scouted we did our best to figure out optimal camera settings for each location – picture profile, temperature/white balance, optimal aperture for desired DOF, optimal shutter speed for moving/jogging subjects. We used all natural light and bounces for the shoots utilizing models and four fluorescent studio lights (wish they were daylight) and a white poster board for the product shoots.
As far as digital workflows go, I implemented a variation on what I’ve done in the past with students. The basic idea was to first divide and distribute images among all the students and myself (depending on what they could handle/wanted), then color correct and do basic touch-ups in Camera Raw. As each student felt they completed this step, the image was shown to me and other students for comparing to pictures I worked on beforehand and used as references. If the image needed more work, I would discuss what needed to be fixed with the student, if not I send the picture back to them for more complex work inside Photoshop (CS5). Again, the image is shown to me, we go over it together and make changes from there. So to recap:
Camera Raw > Compare > Fix > Photoshop > Compare > Fix
In the beginning, there was a lot of comparing and fixing going on as the students were still getting used to this sort of work. I ended up doing many sit-downs and Skype calls to discuss techniques and tools that should be used and things to take into consideration. Consistency in the colors of skin, wardrobe, the product, and backgrounds from picture to picture can be a problem without constant monitoring especially when work is divided among many. As we strolled along though, I had to do less talking and each student did a pretty amazing job of developing their own custom techniques.
Part II: The Commercial
After the photos, DGA asked for a video. We went through many many versions of the story before settling on what ended up making it to production. One of the main problems we kept running into was simplicity. At first it was the story that wasn’t simple enough, then after we got it to a manageable point it was the storyboard that wasn’t. We had way too many shots and weren’t doing a good job of taking advantage of the power of suggestion.
Finding our actors wasn’t too big of an issue however, we had some problems there too. Eight hours before the original shooting date, our actress emailed saying she had strep throat. That unfortunately set us back another couple weeks. We eventually found a replacement (Dylan Quigg) through Josh, one of our actors.
Back to the technical stuff: 5D Mark II again and ugly yellow fluorescent lights were our main tools. We knew we needed a dolly but didn’t know if it was possible to make one in time. After Philip Behnam and his dad made a few working track-based prototypes, Danny Owens went off and built a huge dolly complete with a platform for both camera and cameraman. Philip and I put a makeshift follow focus together to make focussing while recording less of hassle.
The day of the shoot was stressful to say the least, even after several test shoots and half-rehersals things didn’t feel right for the first three hours. It was easy to lose focus and become overwhelmed with what you thought was going wrong but at a certain point we stopped, reevaluated the situation, and burnt through the shot list with the remaining time.
We may put some behind the scenes footage and images together in the near future if anyone’s interested.
Post-Production was mainly my responsibility on this project just because we were running out of time to meet the delivery date. Arranging the video went pretty well but when it came to color correction I ended up having to teach myself how to use Apple’s Color (FCP Studio) since FCP’s 3-way color corrector wasn’t cutting it. Color is incredibly powerful, it’s really good at isolating certain tones to edit, for example skin. I wished I’d learned it a long time ago, it’s very very handy. The only downside is that for one minute of graded footage it generated 14 GB of files…
In addition to the portion of the video with actors, I was also responsible for the audio logo at the beginning of the video. This is intended for use with future videos, sort of a recognizable signature for DGA. I scored it in Propellerhead’s Reason 4.0. Brownie points to whoever can find the easter egg in the audio
You can view the video here.
Part III: Aftermath
This section of the post is to keep track of Zipbuds and all the media we generated for DGA, as well as anything that occurs as a direct result of our projects. This will likely be frequently updated.
Besides myself, all of the following contributed to production in some way shape or form:
Actors/Models: Josh Marble, Dylan Quigg, Kale Beever-Riordon, Garrett, Jessica, Tenesia, Victoria