The time-lapse sequence has become quite popular in how-to videos to replace the old cooking show stand-by, “Here’s one I prepared earlier.” The explosion of DIY videos on the internet is really great but often the time-lapse sequence of the thing being built is not all that interesting. The key to making a worthwhile time-lapse segment is the same as the key to a good camera move:

Make the middle interesting, otherwise use a cut.

If all that matters is the before and the after, use a cut or transition. You wouldn’t want to use a long, slow pan from one item of interest to another if the area between them is visually uninteresting. The same holds for the time lapse; if it’s just people moving around really quickly and you can’t tell what’s going on, you probably can just cut to the chase.

The best time-lapse footage comes from something where you can see a long arduous process come to life, like a building being erected in a couple minutes or a pumpkin rotting. Here’s a great example from my friend Clay.

The term “pulldown” has been around, and implemented since films were being edited on videotape. It is only now, with almost every prosumer camcorder on the market having the ability to record in 24p format, that pulldown has entered into conversations among amateur filmmakers. Before discussing pulldown, a few basics about the NTSC video signal.

In the United States, the National Television Standards Committee (NTSC) has defined a standard for all analog (non HD) television systems to operate which is based on a 60Hz frequency of alternating current. A video image is made up of 525 horizontal lines of information. An electron gun inside your television scans a single frame of video twice in 1/30th of a second. Each scan is called a field, and this process is referred to as “interlacing”. It starts from the top line, called the “upper field” or “odd field” and alternates fields (1, 3, 5 etc.) until it finishes at the bottom. (This process occurs in 1/60th of a second). The electron gun then returns to the top of the screen and scans the “lower fields” or “even fields” and completes the image. This process produces exactly 30 frames every second. This was true for black and white television. As many of you know, however, todays television signal is now slightly slower running at 29.97 frames. This was a side effect of the invention of color television, and a phasing issue between the color carrier, and the sound carrier. Thus the color carrier was slowed .03 frames per second creating a very complicated scenario for todays filmmakers. Okay, now for another equally confusing topic… pulldown.

Film cameras, and many video cameras, as mentioned before, are capable of recording at 24fps. Pulldown is a process of eliminating frames created during the telecine process (film transferred to video) for computer editing. This allows an editor to edit a 24fps project without having to worry about making a cut on a “fake frame”, a frame created in telecine which does not exist in the originally shot film material. When a film project is completed, an edit decision list (EDL) is created from the computer, which is then sent to a negative cutter who uses the list of cuts generated from the computer to make an actual film spliced version of your movie (yes the process still occurs). If you did not pulldown your video when capturing your material into the computer, every edit has a 1 in 5 chance of being a “fake frame” thus making your EDL incorrect for the negative cutter. (The telecine process adds 6 frames of video to your film material creating your 30fps).

2:3 pulldown is the most common form of pulldown, which means that the first frame of your film is represented by the first 2 fields of video, the second frame of film by 3 fields of video, third frame, 2 fields, fourth frame, 3 fields and so on. Take a look at the chart.

Telecine machines are also capable of transferring material for a 3:2 pulldown. All this means is that digitized material must be set to first capture at a B frame, rather than A.

All this being said, what does pulldown do for amatuer filmmakers utilizing 24p technology. Nothing — unless of course, your material is being transferred to film or another playback method capable of running at 24 frames per second. The final playback method of your finished project should be considered at the time of acquisition. Although it may be tempting to shoot in 24p, the filmmaker should remember that if the final playback ultimately will be for DVD/TV that run at 29.97 fps, so will your movie.

I am a big fan of Indy Mogul (indymogul.com) but I was a little disappointed with one of their recent episodes of Q and Erik, their weekly filmmaking question and answer show. In response to a reader question about achieving a film look, Erik claimed that frame rate was one of the two big factors differentiating the film look and the video look. He suggested looking for a camera that shoots 24P, or 24 progressive frames per second. This is a commonly held misconception about video and I’d like to try and clear this up a little.

Film cameras in the US run at 24 frames per second. Television and video runs at 30 frames per second. So, what is the difference? I’ll explain.

One of the ugly downsides to the video image is what is often referred to as “the video edge.” The edges of objects seem perhaps too crisp or hard, especially around bright areas. The video edge has two main sources, one is loosely related to frame rate and the other is not.

1.  The charge coupled device (CCD) in the camera, the chip that registers the image, is divided into distinct pixels that are isolated from one another. If light hits one but not the other, the edge between them is very distinct. Film does not have this problem.  The light receptive molecules on a piece of film are scattered randomly and they overlap. Also, light can pass through the film emulsion and bounce of the back of the film to expose nearby areas of the film. This is called halation and is the reason for the “glow” around bright objects in a film frame. Different film stocks will halate differently. Pro videographers will often use a filter such as a Tiffen Pro Mist filter to soften video’s hard edge.
2. The common misconception, (promoted by video manufacturers who cannot solve the first problem) is that the video edge comes from a lack of motion blur. It makes sense that more motion blur could soften the video edge. The fewer frames per second, the longer the exposure of each frame, and thus the more motion blur. That is the party line. Is it true? Not exactly.

A film camera has a shutter shaped like a disk with a pie piece cut out of it. It rotates at 24 times per second to give 24 frames per second. During the time when the pie piece cut out is over the frame, the film is being exposed. The time when the film is not exposed is used to pull the next frame into place. Generally, shutters use a 180 degree cut so the actual exposure time for a given frame is 1/48 second. An ordinary consumer video camera recording 30fps will actually record 60 interlaced fps. Because there is no film to pull through the gate, the video camera doesn’t need to use a shutter and thus can expose each interlaced frame for the full 1/60 of a second.

The above is a naive generalization of the issue. In reality, video cameras do have an adjustable shutter speed to cut down on light and film cameras also have adjustable shutter angles. There is a distinct difference in the look of a short shutter, but it’s not the thing that separates film and video. For example, do you think Saving Private Ryan looked like it was shot on video? No, but it did have that sharp crispness to the image because of a small shutter angle.Good video cameras today can shoot 30P, or 30 progressive frames per second. The exposure can be set to 1/30 of a second which is even slower than a film camera do, but it doesn’t eliminate the video edge. 30P is a better choice than 24P for most videos. But here is the real killer…

The 24P camera will stretch your 24 frames into 30 frames by interlacing the progressively shot frames, thus counteracting the fact that it shot progressive frames in the first place. Interlacing probably contributes more to the bad video look than frame rate so why would you add interlacing just to lower the frame rate?

If you bought a 24P camera, don’t feel bad. It probably shoots great video at 30P.

My Conclusion: 24P is a gimmick. Shoot 30P unless (and only unless) you plan on getting a film print made, in which case, shoot 24P advanced (Google this if you want to know how it differs from vanilla 24P).

There are always rumors circulating about the release of real time TC generator effect for avid, to which all are inevitably untrue. As a professional avid editor, you maintain a certain close relationship with the tech guys out there in Tewksbury, and amidst our regular conversations, I kept asking the same question. “When are you making a RT timecode gen?” — Always to which they responded “Soon”. Frankly, I got sick of asking, and decided to make my own. It was mind numbingly easy to achieve and have never had to render another TC window burn again. Here it is –

Step 1: This will be the last TC effect you will ever have to render. Make it a long one. Create a sequence of however long you will need your TC to run. Mine is 1 hour. Apply the effect with no video in the background, and render.

Step 2: Create a quicktime of your sequence.

Step 3: Import this quicktime into Avid. This will be your TC Burn In .

Step 4: Mark an “In” point at the first frame of your TC Burn. Mark an “In” point to the corresponding Timecode in your sequence. Edit your TC clip as the upper most track in your sequence.

Step 5: Apply a Picture in Picture Effect to your TC Burn. (Note: This is a real time effect and does not need to be rendered!) Crop the Window Burn to reveal the Timecode. Use the postion to place the TC wherever you like on the screen.

That’s It! A perfect Real time TC Burn that MacGyver would be proud of. A few pitfalls to be aware of however. Make sure that your Burn In Timecode and your sequence Timecode match — meaning both are Drop Frame, or both Non Drop, however you are working. Also, if you let it, Avid will try and render your effect… No! It will play perfectly fine unrendered. Deselect this track, while rendering your sequence.

I am a big Photoshop junkie. I think it’s a great way for people working in video to gain a deep understanding of color and composition. After coming up with some cool new effect in Photoshop, my first thought is, “Wow, how do I get this effect in video?” Granted, there is a lot of overlap between the filters provided by Photoshop and those in most standard video software, but there’s still a lot of stuff that your average video software just doesn’t do as well as Photoshop. AfterEffects comes close but it still can’t do everything. That’s why I was really happy to find this great little tutorial describing how to apply a set of Photoshop effects to your video.

One effect I can’t wait to try out with technique is to fake miniature macro photography. To make this look extra cool, have the people in the scene move at half speed. Then play the video at double speed so everyone looks a little twitchy like hand crafted stop motion.

It feels like the first post should include some kind of manifesto. Certainly, filmmakers are no strangers to manifestos. It does seem a bit pretentious though doesn’t it?

Instead, I think we’ll just start with a quick intro to what to expect from this site. Agile Filmmaking is our entrance in to the growing new worldwide conversation about how movies get made. It is geared toward the next generation of independent filmmakers. The posts will try to provide helpful info on the art and the craft of filmmaking.

The title, Agile Filmmaking, is a methodology for getting films from idea through post-production. It is about taking the many complex tools and skills needed to make movies and getting them to work together in an economical way.