How I Edit Video (2014) Fun & Interesting
It’s not woodworking, but an integral part of the process for me is to make videos of the projects that I do, and that involves editing it into something that is watchable. I’ll say up front that I don’t consider myself an expert on the subject – I’m still learning, but I have developed a method that works well for me. The method applies to editing in general, and not specific to a particular video editing program.
This article is meant for those that have some familiarity of the video editing process and doesn’t cover the basics. Also, this is not an endorsement for the software I use, and as I have no experience with any of the other ones available, this should not be viewed as a recommendation for the products I use.
Most of my videos are made up from multiple clips, sometimes well over one hundred are used to show all of the operations from different angles. A lot of these clips contain only a few seconds of usable action and I’ve found that it’s best to edit each one individually to reduce the length. I then render each clip to the format I use (MP4) at the highest bitrate. Doing this reduces the size and the load on the editing software and computer when assembling all of the clips into a finished video. It can also break the work up into more manageable chunks, without having to spend time going through what was done to catch up.
For this article, I’m using my angle cutting guide video as an example. It is a medium sized project, with only 41 clips recorded with a total size of 4GB. I record everything in 1080p (1920×1080) at the highest bitrate available from my camera. I have found that I get better end results when I start with the highest quality video.
To start, I load all of the clips into the editor:
The images for this article are bigger than normal to make them easier to see. They can also be viewed fullsize by clicking the image.
I use PowerDirector 11 for editing, but the methods used in the article will apply to any video editing software, with a few minor differences.
With all of the clips loaded, I move the first one into the timeline and trim it down to just the part I want to keep. I will normally leave a few tenths of a second ‘buffer’ at the beginning and end to trim in the final edit, and I go into more detail on that later:
When I’m happy with it I can render the clip, save it and delete it from the timeline:
Even though I record in 1080p, I render to 720p to reduce the final size of the video. I do this mainly because I have an expensive internet connection that can be slow at times. Besides that, I think 720p looks great on any screen size and is perfectly acceptable for woodworking videos.
Each clip is rendered at the highest bitrate available. For the software I use, that’s 26300 kbps:
Back to the editor to work on the next clip, and it’s one that has a short voice-over at the start. Whenever possible, I try to do any voice-over sequences speaking to the camera immediately before recording the work. That way, there will be a continuity with the parts where I’m speaking directly to the camera and it won’t sound radically different.
Here, I’ve trimmed the voice-over and split the audio from
the video. The video part will be deleted and I can overlay the voice-over onto the video showing the work:
I lower the volume on the work so that the voice-over can be clearly heard, but still have some sound from the operation:
I will occasionally speed up the voice-over if it sounds like I’m speaking slowly, or if I need to squeeze it into a shorter clip. Increasing the speed by 10% is usually possible:
To further reduce the length of the voice-over, any small pauses can be cut out, as long as the end result sounds natural:
The same can be done with any other parts that you want to cut out, like an ‘umm’ or ‘ah’.
This might sound excessive, but can really make a large difference in the end, making the video flow better and cutting it down to a much shorter overall length.
Another example of overlaid voice-over:
In this sequence I draw two arcs and the time resetting the beam compass to the new radius is trimmed out. Whenever possible, I cut out the parts that are not needed to reduce run time. A couple of brief pauses were also trimmed in the second voice-over, making it flow and fit better.
Here’s one that’s a bit more tricky. I have a fairly long cutting sequence that I want to radically speed up, but my software only allows 2x for the audio. To get past this, I split off the audio from the video of the cutting and increase the audio speed 2x and the video 6x:
I can then trim the audio to fit and it blends in fine with the band saw cutting. I leave the start and end of the cutting at 2x speed, reducing the volume on all. The voice-over is left at full volume and normal speed:
It is also an option to render the cutting sequence without the voice-over, then re-render at 2x again and again to speed it up to the desired amount. The voice-over would be added in the final edit. Most times, what I did here will work fine and the slight audio mismatch will not be noticed.
For this clip, I have a sped up section where I flip the work piece to drill the next hole, and do the same audio trick – speed the video 6x and leave the audio at 2x and trim it to fit:
Again, it may seem extreme, but every little bit helps. Tricks like this can also make the video more interesting to watch, if they are not overused.
In this sequence, I added a still photo and rather than have silence while it’s on screen, I overlaid some ‘room sound’ from the same clip. This is a part of the video where nothing is happening and is relatively quiet:
If I had a voice-over that ran through the part where I show the photo, that would be best, but that wasn’t possible in this case.
When all of the clips have been edited, I can reload them back into the editor:
The overall size has been reduced to a bit more than 1GB (down from 4GB), which is much less stress on the software and computer, increasing the stability of the process.
One thing that I learned early on is that video editing is a massive resource hog and everything that can be done to lessen the load will pay off in time saved.
With all of the clips moved into the timeline, I can see the overall length of the video so far:
Watching it from beginning to end gives me an idea what else should be done to improve the pacing. For this video, it was pretty good at this point – a little rushed through the middle, but it’s probably better to be a bit too fast than too slow.
I almost always start a video with a fade from black, and I do that by ramping up the opacity of the first second or so of the opening clip:
There is a fade transition in PowerDirector, but I’ve had problems with the transitions in past projects, so avoid them now whenever possible. Often the best approach is to do straight cuts between clips, but there are exceptions. When going from one clearly defined operation to the next, I have used fade to black to separate the sequences. A long cross fade is also good for cutting out the middle of an extended operation (like making a rip on the table saw). In the end, the use of transitions is better related to how you want the video to look – little to none for a fast paced, jarring video, to several for slower paced, more polished work.
Getting back to leaving the clips long at the beginning and end, I find it’s usually a good idea to trim out a small section after the clips are put into the timeline. Often there is a bit of a break between that is only noticeable in the audio and doing this cuts that out:
Here’s an example of the fade transition from one sequence to the next. I fade out from the part of the video where the project is being built and into a shot talking to the camera for the closing:
Again, I adjust the opacity on both clips to get this effect.
I generally use a longer fade out to end the video, usually around 2 seconds:
I also fade the audio to zero.
After watching it a couple more times and making more adjustments, it’s ready to render at the same settings as the clips:
The final video is 6-1/2 minutes long and 1GB:
After watching it yet again, I decided that I should add a couple of things at the end. One of those is a caption across the screen in the final scene:
I used the title generator in PowerDirector to do this. Adding a black, slightly blurred border to the letters makes them more visible against the background.
Another way to do this that I’ve often used is to create the caption as an image in my photo editing software and overlay that onto the clip. The image must be a .png with a transparent background. This method gives more flexibility in how it looks and what I can do with it.
After the changes are done and the video is rendered again, I watch it through looking for mistakes. When I’m happy that everything is ok, I do one extra step to finish the video, and that’s to use ColorDirector to ‘clean it up’ a bit. I discovered this program about a year ago and have been using ever since. Admittedly, I did make a mess out of the first few videos, and the software is really not made for this exact purpose – better to correct each individual clip prior to further editing, rather than try to work with the whole video.
I did a lot of fiddling with it to come up with the settings that work well to improve the video by a noticeable amount. As the final rendering of the video is at a much lower bitrate, these improvements are
well worth the extra time involved:
The before and after show the improvement to color and detail with a lot of the ‘haze’ removed.
The final render is at the lower bitrate of 13000 kbps:
I’ve experimented with lower bitrates, but find this setting retains the right amount of detail for a relatively small total size.
The finalized video is 630 MB and ready to upload to my channel:
And here’s the finished product:
Like I said at the start of this article, I’m far from an expert on the subject, but I have had some success with the method shown here. I hope that some of the information presented is useful for those of you that are getting into this type of thing.