Video Production on a Shoestring
My wife Jennie has been cooking for people on special diets for over a decade. In that time she built up incredible skill in making food that tastes good with one hand tied behind her back. No gluten allowed? She knows a lot about nut flours. No sugar allowed? She figured out how to get raw honey to behave.
The ABC network ran a special recently on the people behind YouTube shows and how they are building careers by posting videos. That got Jennie inspired.
This article is written mostly to myself. Here’s how we made the first video, and what we’ll be thinking about in the future.
Make a video. Actually make a video. We have a big family, limited time, limited budget, and limited brainpower. Work within the constraints we have.
Which leads me to my first rule of doing new stuff: Don’t spend any money.
Whenever I get the itch to do something new anymore I try to do it without spending any money. Or to keep the budget at something ridiculously small. So for Jennie’s video, she used a little SD camcorder we picked up at a pawn shop for $80, and the built-in iSight camera in her iMac.
That actually worked out pretty well as you can see when you watch her video. Apple puts decent cameras in Macs. Our kitchen has a very pretty backsplash, and Jennie is quite pretty herself. So the whole thing looks plenty good enough for YouTube. I think you’ll agree.
Since it wasn’t convenient to pick up a 24″ iMac and point it over a mixing bowl (we didn’t actually try it, just sorta assumed) Jennie just used the camcorder. She pointed and shot when she thought we’d want to see more up close. Since she shot the whole thing while I was at work, there was no one to help. It’s a nice constraint, really. The whole thing comes off like she’s showing you how to do this over Skype, but with a few closeups.
My only gripe with this setup was the audio. Jennie speaks loudly enough to be heard clearly, even over the red kitchen aid mixer. But even with that, you can hear that she’s several feet away from the microphone.
Even with what I consider to be so-so audio, we finished and released it. Which is principle two of doing new stuff: Actually finish something.
I was ecstatic when Jennie called me and told me she had finished shooting her video and liked how it came out. The hardest part, creating something of value, was done. And for my gripes about the techinically weak audio, her words are great.
So we were on track to actually assemble the footage and post something.
Now we had made at least one choice that made things complicated. By pausing the video and having closeups we had introduced an extra step. It would be easier if we had just shot the whole thing from the iMac in one go, no camcorder closeups, no pauses. That probably would not have been a very interesting video though. And besides, it’s no problem, we have an iMac with iMovie. What could go wrong now?
So I pulled together the camcorder footage and the iMac footage and started working in iMovie. That’s when I discovered what iMovie ’08 is good for. Turns out, nothing. This is footage shot on the iMac from the iSight camera in Photo Booth, mind you. I would play my second video clip and iMovie would play the audio from the first clip. It was maddening. I did eventually find a way to work around this, but even then, iMovie ’08 can’t actually let you show your closeup shot while leaving the audio from your main shot running. Useless!
I also tried Cinelerra. I have a modest Linux machine I use for development at home. I found Cinelerra to mostly be good for crashing.
After failing to edit the video for a whole evening I was discouraged.
I figured a free Linux video editor was what I wanted until I had more cash to burn on Premiere or Final Cut. So I did more research on Cinelerra. Maybe I could fix it. Well, it turns out no one uses it anymore. Everyone has moved on to OpenShot. That is a thing worth knowing. In 2011, the best video editor in Linux is OpenShot.
I have my gripes with OpenShot. It lacks visual indicators for clip fades. It’s underlying engine is probably capable of fancy audio processing, but very little of that is accessible. And there are still some crash bugs.
However, it edits video. The basics are all there and actually work.
So in a couple of hours I had the whole video edited.
The only problem left was the low audio level.
I ended up using avidemux to do my last compression pass. I found in avidemux an option for compressing and then renomalizing the audio. This got the audio up to what I felt was a passably loud level.
So I lived with this. We needed to put this video out.
I have this theory about audio, see? I think our eyes are far more forgiving than our ears. Watch the original Veggie Tales release sometime. Even though the characters are just blobs and the movements are more geometric than organic, we still suspend our disbelief. We can synthetically produce images all day, and viewers will see real characters in them.
But have we ever seen a cinema debut of synthetically produced voices, for important charaters? C3PO? Human voice actor. Star Trek’s computer? Human voice actor. Toy story? All voice actors. We all thought CGI for a whole movie was neat, but the audio was just plain old voice acting. They didn’t have giant render farms for Buzz’s voice. They just miked Tim Allen.
So that leads me to believe that the priorities of commercial YouTobe video are as follows:
- Be worth hearing.
- Be worth seeing.
- Audio quality. (distant third)
- Video quality. (barely matters)
I think it’s a mistake to obsess about lighting and camera angles unless the audio is perfect. But maybe that’s just me. Maybe I’ll read this in a year and laugh at myself. But I think I’m right about this. I’ll be watching for an inexpensive simple to use lavalier mic that we can plug into the iMac. That won’t get us “radio good” but it should get us almost “TV good.” That’s a huge improvement over “post-processed Skype.”
In the end our little show has been well received in our circle of friends, including someone who I think ought to know better. Which leads to our principle of marketing and growth: Don’t try to compete with the established players.
I think Jennie is talented enough that she will eventually be able to make at least few dollars doing this, and I think she has a shot at some real money. However, we don’t think that we can compete head to head with Mario Batali or Alton Brown.
We’re drawing a smaller circle and focusing just on cooking with Spelt flour. If you want to learn to cook, go watch the Food Channel. But if you are interested in the many benefits, wonders, and intricacies of Spelt flour, come see Jennie. Spelt is becoming easier for us to get, which tells me it’s a growing market. Lucky us.
So we’ll see how this goes. We’ve been delighted with the response to our first “show” and hope to see all of you again when Jennie makes soft pretzels next time.