A collection of notes related to my current video editing process – in concept up to date and adjusted as my various tools and processes change:
- Collect the raw footage and audio
In my case, my camera records at an unfortunate 15 fps, and I have to record audio independently. This creates some initial challenges before I can even start editing. With my current hardware, I have the camera set to record from a tripod, typically located above the scene, and fixed on a specific point – I generally just let it run without modification during the game session or whatever I am filming. At the same time, I have my microphone, the H1 Zoom, mounted on a tripod as well and positioned below the camera. Since it records in stereo, this seems to give the best orientation to the film sound. Since the microphone is so sensitive, both the camera and microphone are normally on a separate table or stand, so that any activities in the game don’t cause additional knocks and noise on the recording.
When recording, I try to create an audio and visual key near the start of the game (typically rolling a handful of dice in front of the camera). This gives me a clear sync mark for when I later try to line up the audio with the video. I also always plan to do the same thing near the end of the session, though I almost always seem to forget that step, and end up having to just use some event in the game. This second sync point allows me to adjust the audio speed for the slight drift that seems to be present in the camera recordings – otherwise the audio drifts out of sync by approximately 1 second for every hour of footage. Not a big deal over a short stretch, but more than a half second or so, and it becomes very distracting.
Once I have the raw footage, I move it from the camera/microphone and onto my computer, where I can create and open a new project in Davinci Resolve (version 12.5 currently). The first task here is lining up the audio with the video, a process which isnt too time-consuming, but is a bit fiddly. First I locate my first sync point, and line up the audio and video, trimming everything that comes before so both points are at 0. Then I locate (or choose) a second sync point near the end of the footage. Using a excel spreadsheet, I calculate how much slower or faster the audio needs to be adjusted, so that the second sync points line up. Normally, this works out to about 0.08% increase, which doesnt seem like much, but is a surprisingly large impact on the overall project. With the second sync points lined up, I can then find the end of the session content, and trim out everything after to remove the excess length.
Because the audio and video files are broken into 2 GB filesizes (due to a limitation in the storage), this results in a single extra long project, composed of numerous audio and video files. I have found that this is very troublesome to work with, and keep everything in sync, so my next step is to render the project, in order to create a single large video and audio file. That way, future edits don’t shift pieces unexpectedly. Once this “Pass 0” is complete, the raw footage is no longer needed, and the file size is around about 20 GB.
- Edit the footage
With Pass 0 complete, I start the primary edit. Again, I go into Davinci Resolve, and create a new project (or re-use a prior project), and then import the consolidated video file. At that point, step 1 is a color correction. Needless to say, my camera isn’t the best at colors, so my white tends to be rather yellow. With a single large file, color correction is quick and simple – I just use the auto-correct option on the entire film. Yes, I could likely get better results if I tweaked and adjusted the settings manually, but I honestly have no idea what I am doing there, and the auto-correct setting seems pretty solid.
With color correction complete, its on to basic content editing. In this stage, which is by far the longest, I repeatedly listen and watch the footage, and trim and cut out any sections which are unnecessary. For the most part, this is side conversations, rules discussions, and all of the other distractions that are caught by the camera by just allowing it to record. What I have found is that about 80% of my raw footage is unnecessary, so I try to aim for 10 minutes out of every hour of content – but that isnt a hard rule. It is really dependent on the subject matter and how the story is being told. In reality, I am trying to tell a complete story, and cut out the fluff. In my case, I also find I have a lot of pauses and “um’s” in my speech pattern, which is good to cut out at a minimum.
Depending on the computer I am using for the editing and my time available, this process may take a couple of days. In particular, as the number of cuts grow, your system will consume more and more memory. If I run out of time, or start to get memory concerns, that indicates a stopping point, at which point I re-render the video into a new smaller single file. Once that is completed (Pass 1 or Pass 2), then I can repeat this step until the movie is down to a lean cut of the footage in a single output file.
- Formatting and special effects
With the video down to a single file, I can then add the intro and outro sections. Both of these are based off of templates, I simply open the template, adjust any variables needed, and render a short movie. Then I can bring the intro and outro movies into the larger project, and connect all 3 pieces (intro, movie, outro).
As part of the outro, I include a 20 second span of “blank” content, nothing but a background image and possibly some music or audio. This is for later use within YouTube, for the content links (Subscribe and View Other Content links). YouTube always defaults to overlaying the last 20 seconds of your video, so if you don’t include the blank scene, it will overwrite credits or actual content. This way I have a clean slate to add those pieces at the end.
The last step within Davinci Resolve is the “special effects” – though it is more of just basic artistic additions. Depending on the content, I add graphics, text, or other flourishes to the video, keyed to the audio and footage. For example, in my Pathfinder game sessions, I add images of the active player characters, any NPCs they may be interacting with, and anything else relevant. Then as the scenes change, the graphics fade and shift to the new focus characters. This helps (at least in my mind) the audience keep track of who is talking and acting, and also serves to distract from the background of the footage itself. In the case of Star Wars Armada, I add in fleet lists, so that anyone watching the video understands what fleets are engaged.
With the special effects added, and any music or sound effects added, the film is rendered one last time. This combines all of the last elements, and gives me a single final file. By this point, a 40 minute video is down to about 2-3 GB from the orginal 20+ GB. This file is then finally uploaded to YouTube, where it is processed and then ready for publishing.
- YouTube publication
On YouTube, there are a couple more steps, but they are fast compared to the time for the video to upload and for YouTube to process the video. I update the various settings on the YouTube page itself – the video title, date recorded, if it will be public or private, etc. In particular, I select “Creative Commons – Attribution” in the License and Rights ownership, since I use YouTube’s provided royalty-free music for most videos. I also fill in the video description with a basic summary of the content, and add a pre-compiled list of Amazon associate links (which I keep in a separate Notepad file), along with the Creative Commons attribution text. All of this information can be filled out while the video is uploading and processing, so that it is immediately published when the processing is complete. This also automatically creates a YouTube notification for any followers of the channel.
If the video fits one of the playlist topics, that can be selected as well, and for Monetization I select overlay adds and skippable video ads. Generally, I do not add mid-video adds – I don’t like them myself and I don’t want any viewers to be bothered with them either.
Once all that is done, I make a note of the video URL, and create a similar posting on my website, to link to the same thing, scheduled to be posted either that day or shortly after. The WordPress link also creates a Twitter and Facebook post, to help with the advertising. Since the video tends to be published earlier than the WordPress posting is published, this gives a second link to the content.
The Twitter feed is updated automatically by both actions, as a basic setting within YouTube and WordPress – though I will likely disable the YouTube update in the future just to avoid the duplication.
Once the YouTube file is processed, the final step is to add the Endscreen Elements, which is why there is a 20 second gap at the end. I add a link for the audience to subscribe, and a link to my playlist with similar content, in case someone has watched to the end, and wants to see something more. At that point, the video is complete, and I just wait for any comments.