Trimming the Script
Friday, September 30, 2005

We're ten minutes over in length...

When we started our Spawn: Animation journey, it was decided that we'd go for a running time of approximately 75 minutes. The first step to reaching that goal is in the length of the script, which as it stands now, is 81 pages. Okay, so maybe from the get-go we were pushing the envelope. But the minute-per-page estimate is just that, an estimate, so in all reality we were pretty close to being right on target with length ... on paper. The final length of the show is determined after the dialogue is recorded and the storyboards are finished.

There are a number of factors that can change the timing of the show. Set design plays a big role in timing. For instance, if an environment is large, it will take more time to move across it; whereas, a scene in a small space may take less time to cross -- or it may not even be necessary for a character to move to address someone or something.

In addition to the physical space, the voiceover talent plays a huge part in timing. How each person delivers his or her lines is the primary factor in the length of an animated production. A character with a southern drawl takes longer than a fast-talking city kid to say the same sentence. Although we could have timed the dialogue before the actors recorded, our efforts may have been wasted, as it may have changed after the actor reading each part gave their performance.

Now we have reached the point in production where the storyboards and dialogue are complete and we are able to time the show. This means sitting down in front of the computer with a stopwatch and turning pages as the dialogue is heard. However, as I mentioned, you cannot gauge the length of the production via the dialogue alone. You must account for dramatic pauses and scenes without any dialogue at all. Assigning a running time for each page with script and storyboards in hand is crucial to get a realistic estimation of length. One way of doing this is by actually acting out the movements. For example if we have a scene in a church that has been built on the computer, we then take those dimensions and translate them into real space. Using those lengths, someone actually times himself walking the distance and adds the time in accordingly. It's not very scientific and there is a whole lot of room for error, but getting the most accurate timing for the animation is a big step.

In our case, our run time estimate was too long -- meaning that before going to physical animation, cuts had to be made. The show timed out at about 85 minutes, and we needed to be closer to 75-80 minutes in length.

Film Roman sent over a list of suggested cuts to trim approximately 8 minutes. With their list in hand, it was our turn to go through the storyboards and make some decisions. The main thing Todd was looking for was redundancy in scenes. Finding similar shots in multiple scenes enables you to cut the fat without really losing anything -- because in editing you can reuse a shot from Act 1 in a later act, and have it go unnoticed.

This process on our end took about five hours. Since we had sort of a cheat sheet from Film Roman, we were able to move through the thousands of boards fairly quickly.

All in all, I think some nice compromises were made to trim things down without losing too much of the artistic integrity of the animation, and without losing any of the key story elements. If all goes well, we should be moving into physical production of the animation by the first week of October.