Tuesday, May 27, 2008

Class Capture vs. Screen Capture

Over the winter break this year, ECIT purchased and installed an Accordent Capture Station in one of our classrooms. The desire was to have a space in which video presentations can be easily captured as we move toward pushing more and more stuff to iTunes U for the public release this fall. The classroom now has two smartboards, a remote control camera, wireless mic, a Crestron control system, and a stack of computers and receivers.

When in class capture mode, one of the smart boards turns off, but the Accordent system captures everything that happens on the other board at approximately one frame per second. Simultaneously, it fires up the camera and starts recording video (30 fps). The resulting presentation, which is viewable in Real, has two screens: the smart board and the video of, presumably, you or the class. This is nice because it allows those watching a presentation to see not only the documents, spreadsheets, or PowerPoint/Keynote presentation but also the speaker's facial expressions, hand motions, etc. You get a better sense of what it might be like to watch such a presentation in real-life. What's more, we now have a dedicated space made for video capture so faculty at Emory no longer have to worry about getting a video camera taken to their classroom. The headaches saved by this feature alone probably make it worth thinking about using.

In order to test the system and think about its capabilities, I recently had the chance to record a med school lecture on the heart in the space. Even better, this last Friday I spent an afternoon filming my own presentation on--you guessed it--timelines. Essentially, I discuss the basic features of the interactive timelines I've been playing with this year and demonstrate how easy it is to add information to such a timeline within Google Docs. You can watch the presentation here.

Now, it just so happened that I recorded a presentation on timelines that was more or less identical to this one I made for my Timeline Tutorial using Camtasia. Having already made one presentation was helpful in the sense that I more or less knew what I wanted to say when I got put in front of a camera. More importantly, it gave me the chance to think about the relative advantages and disadvantages of both systems.

And there are plenty of disadvantages to the Accordent class capture system.
  1. In my mind, one of the main reasons to make a video recording of a class in which you use a presentation on a computer screen is to capture faithfully what you put on the screen. Unfortunately, because the Accordent only captures a smart board image once every second, the presentation side of things is not so much a video as a rapidly advancing slideshow. You can lose a lot of important information if you don't have a perfect capture of what's on the screen, as I learned in making the med school lecture, which used a PowerPoint presentation with a lot of movies. The value of watching the heart in action was lost as motion was eliminated. If a presenter is made aware of this limitation of the system, he or she can move the mouse and/or screen more slowly and compensate for this time delay. I think I actually did okay with this in my presentation, but there is a definite loss.
    • Camtasia, on the other hand, captures exactly what is happening on the screen and does so at an adequate frame rate that the full motion of the mouse or videos your are displaying is preserved.
  2. The camera on the Accordent has a limited range in which it can recorded. This means that instead of calling it a "class capture" system, the Accordent should more properly be called a "lecturer capture" system. There are zoom and pan controls on the Crestron for the camera as well as 5 presets for the camera. But the range of the room that can be captured effectively is smaller than I would like. While recording myself, I had to mark where on the floor I could stand to be sure that I was within the camera frame. If someone else had been working with me, they could have used multiple camera angles. But the process for doing so is more complex than it should be: it requires watching a computer screen while simultaneously positioning the camera using the Crestron unit. Of course, the Crestron is touch-screen only so you have to look at it to use it. This results in shifting back and forth between two screens. Add to this the fact that the camera controls are on the herky-jerky side, and you have a situation in which the best production results from keeping the camera and the lecturer planted in one place.
    • Of course, with Camtasia you are similarly rooted in one place if you want to record video of yourself as you're talking about what you're showing on a screen as you are dependent on a webcam.
  3. If you watch the Accordent presentation, you'll eventually notice that the volume of what I'm saying varies widely with where I'm looking. When recording the video for the med school, I had noticed that the lapel mic didn't have a very large range. I tried to compensate for this in my own production and positioned the mic so as to pick me up when I turn my head to look at the screen. Unfortunately, when I'm looking at the camera, the volume drops. More experimentation with the mic will likely help us iron out these problems.
    • Still, in Camtasia I don't have to look away from the screen as I narrate what I'm talking about. This means that I can use a stationary mic and avoid similar problems.
  4. Another problem with the Accordent system is that it takes a very long time for it to start recording once you hit the "record" button. It makes sense: it has to start two different video feeds and an encoder. Still, it is always surprising how long it takes. It requires having another person give you the high sign when things are finally rolling. Otherwise, you end up missing the first 30-60 seconds of whatever has been said. If you are recording by yourself, you end up watching the Crestron and looking down for the first five seconds of the presentation.
    • With Camtasia, on the other hand, when you hit "record" you are going immediately. Once you hit "stop," there's a bit of a lag as you wait for the encoding to wrap up. But it's nothing like the time required for the Accordent to wrap up its recording.
  5. So you've got a bit of dead space in the beginning of your Accordent presentation? Can't you just lop it off? Here's the real problem with our Accordent experience so far: a lack of editing tools. You have to do everything in one take. No trimming. No patching in a part where you misspoke. It's all or nothing. This isn't a radical oversight on the part of Accordent so much as it is due to the problems that would be involved in syncing the patched in/edited video with the smart board screen that has been recorded. When I was recording the medical lecture this was a real problem as a cell phone went off and we had to start a 20-minute segment over again. We've been told that there might be a tool we can buy from Accordent to edit our presentations. But since we've already spent tens of thousands of dollars on this tool, you'd think that this would come gratis.
    • Of course, Camtasia is explicitly designed to allow you to edit your screencasts. You can go back and rerecord your audio commentary. You can cut in title screens (you'll notice that I had to use PowerPoint for titles in my Accordent presentation), and zoom to different parts of the screen. And because you can use a webcam in conjunction with recording the screen, you can also have a recording of the speaker (albeit, seated at computer) that plays simultaneously with what you have captured on the screen.
  6. A final difficulty with the Accordent system is that it outputs exclusively to Real formats. This means that it won't play nice with iTunes and iTunes U. You can--it appears--put either the video or the presentation video into iTunes U. But the simultaneous videos can only be seen on a computer.
If it's not obvious at this point, I have to say that I prefer the results and experience of using Camtasia over the Accordent system. Compared to the latter, the former really embodies the ECIT mantra of being simple, easy, and scalable. Not that Camtasia is the easiest piece of software that I've used, but it only took a few hours to get comfortable with its different options. And it's a much richer tool. Camtasia also has the advantage of being much less expensive than the Accordent.

Of course, the Accordent is a new tool. We're still trying to figure out how to best use it, what sorts of presentations play nice in the space, and how to get faculty and grad students to make use of it. Still, the gauntlet has been thrown and for the moment, I know which way I'm leaning.


JGM said...

About how much did Accordent cost?
Do you have to pay more to stream to say 100 users? Accordent "sounds" good but something seems to be missing.

Brian said...

I can't tell you exact figures (as I don't know them), but the system was in the five-figure range. The number of people you could stream to only depended on your network capability.