When planning a show’s post production workflow, the most important thing to do is to begin with the end in mind. That means asking two important questions: 1) what do I need to deliver? 2) when do I need to deliver it by? The answers to these two questions will inform every other aspect of your post production workflow, schedule, and budget.
What do I need to deliver? is best answered by the people handling the client’s final deliverables. For a network show, the broadcast department should have a “tech specs” or “deliverable requirements”. For an industrial project, it’s best to look beyond the person who hired you. For example: if a corporation hired you to create a presentation video, find out who’s responsible for the AV projection. If you’ve been hired to create a series of web videos, talk to the IT department, perhaps they will have archiving specifications that differ from YouTube’s.
I like to get all of this down in writing. Networks will often have multi-page PDF documents that I markup while reading them beginning-to-end. Boring, but very necessary. If the deliverables are more informal, such as via email chain, I will actually create a document that outlines the final deliverables and send that to the client for approval. I can’t fully communicate how valuable the effort spent clarifying what is expected at the time of delivery is. Finding out that the final Master files are in the wrong codec at 7 pm on the day of delivery is a miserable experience. If the final deliverables are signed-off on in advance, then the 7 pm error becomes a billable overage for the client.
Answering the “when” question is a little more nuanced. One one hand, you might only need to delivery one file on one date. But having a more complex delivery schedule is often the norm. For example: when you work with a network, the public relations department may want a “press cut” to share with reviewers weeks before the final Master is delivered.
What’s also important is understanding the order of operations for the final deliverables. That is to say, some deliverables are logically locked into the completion of others. For example: if your master file is in a frame rate of 23.98, but you are required to deliver a subset of files at 29.97; then the frame rate conversion is a hard logic process that must be completed before other tasks can be completed.
The tool I use to navigate these tricky rapids is the Work Breakdown Structure. The WBS reveals the complexity of a project by showing you all of the activities required to create your final deliverables, and the order in which you’ll need to complete them. Recently, I’ve started to use the excellent OmniOutliner in order to create my WBS’s.
But you don’t need fancy software. A blank sheet of paper and a pencil are all that’s really required. I like unlined paper because it enables you to work horizontally and vertically. I also recommend using a pencil because you’ll probably end up needing to erase as you figure things out. Remember: creating a WBS IS the process.
Start by listing all of the final deliverables. Then start to work backwards and think about 1) any work that needs to be done in order to create the deliverable, or 2) any work that can’t happen until this particular deliverable is created.
In my handwritten version of the work breakdown you can see that the 23.98, xdcam, Texted and Textless files are in the center because the other 7 deliverables are created from these two master files. As mentioned earlier, the frame rate conversion creates a hard logic dependency, which means that some deliverables need to be created before others. This is important information in multiple scenarios. For example:
The Public Relations department wants to share a 29.97, h.264, Texted master with the press. If the PR department can’t wait until the 23.98, xdcam, Texted master is created, then you will need to perform additional work, just to satisfy the PR department’s deliverable requirement. Your production may incur additional expenses because the post house bills you for the additional labor.
One of the reasons I like OmniOutliner is because I can migrate my work into OmniPlan once the WBS is ready to be managed using traditional project management techniques. OmniPlan not only enables to you visual the work, but also create time estimates and run simulations that predict the likelihood of delays.
Beginning with the End in Mind is a mindset more than anything else. It’s a way of thinking that will enable you to plan ahead and deliver your projects on time with minimal hiccups. I’ve used this process on all of my shows and I recommend it as a foundational post production management technique. In future posts we’ll work backwards through Finishing and Editorial. Until next time, I’d love to hear about the challenges you’ve faced during Master delivery.
My first series as Post Supervisor was Food Network Star for CBS Eye Too. I have always taken pride in my technical savvy, but this show taught me that some problems are beyond one person’s ability to solve, and that a Post Supervisor’s primary role is to be the team’s best communicator.
The following is an account of the most tricky troubleshooting problem I’ve yet encountered in my post production career:
To: Post Supervisor
From: Night A.E.
Subject: 0122D07 Tape Damage
Shit. This is not an email notification you want to see when you first wake up in the morning. However, this Night A.E. is one of the best. I know the email will be a solid description of the problem he encountered, steps he took to solve the problem, and the results. Plus, it’s better to be informed of problems sooner than later. Slide to unlock my phone. The email reads:
Tonight, while rewinding tape 0122D07 prior to capture, I noted a deck error on deck “B”, alternating “E-52” and “AUTO OFF”. I turned the deck off, called the post house’s technician and looked up the error, which is listed as “WINDUP REEL NOT ROTATING”. The Tech turned on the deck and ejected the cassette, from which a small amount of crunched tape could be seen from the edge. Attached is an image of the tape after The Tech wound it slightly.
The deck seems to be working correctly, and is capturing subsequent tapes without problem. According to The Tech, there is no indication whether the error is related to the deck, the camera, or the tape itself. I will continue to closely monitor all tapes as they are loaded.
Solid problem description. One damaged tape is an outlier. Nothing to be done at the moment. I get dressed and head into the office. The day almost passes until the Day Loader runs to my cubicle to tell me that tape 0120B13 was chewed up. Now we have a problem!
When you troubleshoot a problem you looking for commonalities. However, this problem was different from the start. The tapes came from two different cameras “B” and “D” respectively and failed on two different decks. Therefore, I want to investigate the tapes. I ask our coordinator to collect the tape serial numbers and email our vendor. All analogue mediums are subject to a bathtub curve failure rate; perhaps there was a manufacturing issue?
While I wait to hear back from the tape vendor I update the team:
To: Post Production Team
From: Post Supervisor
Subject: IMPORTANT Tape 0120B13 Lost
Another tape was just chewed up: 0120B13. This was a different deck and a different camera than 0120B13. This is not a camera problem, and it is not a deck problem. I am looking into having the tapes spliced and dubbed, but can not guarantee this will work. Is there anyway we can start implementing safety recording in the field as soon as possible?
My phone’s notification: “so TWO tapes are destroyed!?”
My phone’s notification: “Holy SHIT stop all tapes NOW”
My phone’s notification: “This is a REDACTED problem! Did you call them?”
In retrospect I can’t believe I sent so careless an email. I can’t believe I sent an email!?
Lesson: when dealing with potential crisis, use the phone!
Identify and inform key stakeholders of the problem by phone. Tell them the facts, tell them what’s being done, and if they freakout, don’t take it personally. Repeat until all of the key players are aware of the situation.
When presented with an information gap, people make stuff up to fill it in. Email is a low bandwidth form of communication and people make stuff up to fill in all of their questions. A phone call allows immediate question asking and answering. In addition, the tone of your voice communicates so much more than your words. If you sound confident and “on it” things will go better for you.
Years later I read the following in Vanity Fair about World War Z’s troubled production and Producer Ian Bryce’s effort to fix it:
Movie sets, in his experience, function best when people talk to one another, he said. So before Bryce crawled into bed, he got a list of department heads and senior crew members and sent an email of his own. “If anyone is up for a new way of doing things, here’s my phone number,” Bryce said he wrote. “Here is my assistant’s phone number. Call me. I can talk faster than I can type.” It didn’t take long for people to respond. “It got some immediate hit backs of ‘Hooray!’ ” he said.
Our tape vendor got back to us and said that the tapes were manufactured at different times. But the the Panasonic representative the vendor spoke to was very concerned and asked if it be ok for them to reach out? “Of course,” I said. Having someone from Panasonic help troubleshoot could only be an asset, right? Well kinda. I’d soon learn that once an equipment manufacturer gets involved, so does everyone else.
I tell the EIC about Panasonic’s desire to help and immediately I’m instructed to setup a conference call will all parties. ALL parties. The conference call dial-in included the EIC and show engineer, the post house engineer, the camera rental company’s engineer, all of their attendants, and the Panasonic representative.
After the endless round of introductions the representative announced that they had pulled the service records of all the tape decks. And that there were none. The post house immediately became defensive and said that their decks were maintained internally by certified technicians. When Panasonic rep repeated the same trick on the camera rental company the result was exactly the same. By this time every person on the call was trying to contribute their two cents, but the only effect was an increase in volume and a decrease in the quality of conversation.
All the while, production had kept shooting because the camera bodies were most likely not causing the problem. This continued to add to the post department’s backlog. We were falling behind loading and episode prep fast. We needed to figure this out soon, or post was going to have to postpone the editors and story producers.
With no easy answers, and everyone feeling defensive, the only thing to be done was to suggest an end-to-end run of the entire workflow with engineers checking the tapes at every point in the process. It went something like this:
The cameramen would roll tape like normal. After every load the camera engineer would check each tape and hand them off to the Panasonic rep who would confirm the integrity of each tape. Then the representative would hand deliver the tapes to the post team who would load the tapes under the post house engineer’s supervision. We couldn’t recreate the problem.
Therefore, we started loading the backlog of tapes and almost immediately, this:
It was during the time when the tapes were being hand delivered from the set to post, that the culprit was discovered. While the show engineer was chatting with the camera rental engineer, the show engineer saw an assistant cameraman and field producer reviewing a tape on the QC deck in the corner of the set.
Turns out that the camera department kept a small QC deck on set so cameramen and field producers could watch footage before sending the tapes to post.
The QC deck, which had been out of sight, went unnoticed during the end-to-end test runs. When the engineers checked the QC deck they discovered an unreported servo error that caused the tapes to slightly unspool during playback. When the tape was played back again during loading, the loose tension of the tapes was causing the E-52 errors during capture.
From: The Post Supervisor
Subject: Tape Troubles Solved
Good News: When you open the top of a cassette, the black tape across the top should be taut. The video tape across the bad tapes are very loose. By opening up the tapes we are able to identify which tapes have problems and then fix them by manually rewinding the tapes. After the tapes have been manually tightened the tapes work just fine.
Followed by the Night A.E.’s inevitable shift report:
To: Post Supervisor
From: Night A.E.
Subject: Shift Report
Tonight I checked the cassette of every uncaptured tape available for tape slack.
Of the 700 or so tapes I checked, about 2 dozen needed tightening, with about 8 being bad enough that I would expected that any attempt to load them without tightening would have failed. E Cam tapes represented about half of the tapes I had to tighten, with F, G, H, J and K cams also having at least 2 slack tapes.
Overall, there seems to be some improvement, I checked all 139 tapes from tonight’s drop and only 3 needed to be tightened. Compare that to the 1/20 tapes, where I had to tighten 10 tapes, with E Cam being the biggest culprit.
Including tonight’s drop, all uncaptured tapes have been checked for slack and corrected if necessary. Tonight’s tapes can be found under the desk in the main A.E. room.
Checking 700 tapes. Now is that an A.E. assignment if ever were one!
Some problems are beyond one person’s ability to solve. In this situation, with 5 or 6 engineers involved, I learned that my role as Post Supervisor was to be the team’s best communicator. I also learned that sometimes troubleshooting involves a little bit of luck!
Our library of 3,000 DVCProHD tapes is divided by episode, with two additional sections for b-roll and interviews.
Tapes are labelled date_camera_load. For example: Tape 0210B13 is from February 10, Camera B, thirteen tape of the day. This system works for us. I’ve never been happy about it, but I can’t think of a better system, and I’ve spoken with many colleagues about alternative system.
One additional organization note: tapes are labelled with color Avery circles; each color corresponds to an episode. And tapes are labelled with a Brother tape labeler. In the future I’d like to have a tape database and bar code labeler, but with tapeless media on the horizon I don’t know if I’ll ever get the chance.
The more I think about this new format, the more I believe that the most component is going to be a year-round viewing experience. In addition, I think these bite sized episodes will need to be released at random intervals. That is to say, not knowing when the next part of the story will be unveiled is the key element that will make NewTV must watch entertainment.
As an example: imagine being absorbed in Game of Thrones, as told in NewTV. The story would unfold as if it were happening in real life. You’d be on Facebook and you see your friends posting: “RIP Ned”. You’d fire up whatever NewTV service is hosting our Game of Thrones and find out what happens next.
I think this unpredictable release schedule is going to be critical for one important reason: if episodes are not released unpredictably what differentiates NewTV from regular appointment television?
Netflix is partnering with Buzzfeed to offer weekly 15 minute stories and I’m curious to see how they fare. My guess is not well. When people sit down on the couch at night to “tune in” I think they want long form content, or to string multiple short shows together, hence the popularity of the ‘binge’. What would NewTV be offering if you could just sit down and watch seven 10 minute scenes in a row?
What I like about this analysis is that you can draw interesting inferences from the data:
Thor’s keywords suggest that in the Avengers movies (not including the films Thor headlines, like Ragnarok and The Dark World), he’s more action-oriented than most other characters. With the exception of his relationship with Loki, he tends to focus on tangible artifacts that drive the plot forward. Like Loki’s scepter, the Tesseract, and the mind stone.
Check out how Vision and Scarlet Witch have some similar words- they’re talking about fear an awful lot. I’m hoping they stay synced in Infinity Wars. Interestingly, I did a sentiment analysis as well, and Vision had the most lines with a negative sentiment. It’s not because he’s a constant downer, but because he calls situations like he sees them and reflects sometimes on the futility of the human heroes he comes to love. He sees the extra, extra big picture, and I get the sense it disturbs him.
This analysis is very from the less useful one I critiqued last November. As analytic tools become more prevalent, it is important that our understanding of their uses grows with it. Glad to see work like this being done.