When Creative Director Greg Hahn of New York’s Gretel started thinking about creating opening titles and category opens for Spike’s Guys Choice awards show, aired last month, he gravitated toward the idea of shooting ultra-high-speed images. He knew about high-speed film cameras but had never before managed to match the time and budgetary requirements with the right project. Fortunately, his research for the Spike TV project led him to the Phantom line-up of digital high-speed cameras from Vision Research, which would prove to be much more convenient than shooting film. (He also looked at the similarly capable Weisscam HS-1, but decided to use the Phantom based on endorsements from "friends and friends of friends.") Hahn considered the HD version of the camera, but using an SD camera – provided by Abel Cine Tech – was more manageable.

Hahn co-directed with Darius Maghen. Watch some of the footage, below, then read the interview below to find out what Hahn learned on the high-speed shoot.

What interested you about a high-speed shoot?

In the kind of work I do, generally, I don’t like a lot of smoke and mirrors. The idea of an image being boiled down to its essence fits with my general approach to design and film work. The high-speed image is sort of the opposite of smoke and mirrors. It strips away distractions, and lets you focus on the moment. Not to mention that it looks really cool.

Did you have to pour a lot of light on your subjects to get the shots?

The shutter speed is so fast that you do. We had two 10Ks, basically – two banks of lights. It was an array, so it might have been five 2Ks or something like that. Right away we knew we wanted this slick, high-end look, like a car commercial. The ideas we were shooting were kind of caveman ideas. Spike Guys Choice is like an awards show for guys. The image for “Biggest Ass Kicker” is a guy getting punched in the face. It seemed sort of fun to shoot that really elegantly. The subject is crude but the shoot is high-end.

When we started researching high-speed footage we found a lot of materials are often shot this way, against a black background, and we were trying to figure out why that is. Is it necessity or coincidence? It’s partly because when you’re shooting something like glass breaking it’s nice to have a black background as a foil for the action. You can pick up all the details. Ultimately we did go that way. We ended up, most of the time, with two banks of lights for, essentially, rim lighting and we would dial it in for each shot. Some had a soft key light, but for the most part we went with rim lighting. I was impressed, because I had heard that you do need a ton of light and it concerned me about the subtlety we’d be able to get, whether we would get the right balance of information and then fall-off with the rim lighting, and we were able to get exactly what we had hoped for. We had a lot of reference and very specific frames going in. We had a target area we were shooting for, and Aaron Philips, the DP, really nailed what we wanted.

When you saw playback on set, were you surprised by the look of the shots? Or were you able to imagine clearly what you’d be getting?

The amazing thing is the playback is what it is. Right away, you see exactly what you’re ultimately going to be seeing. We did very little correction to the footage in either speed or color, so what we saw on set was pretty close to what we ended up with. That was definitely the best part of using that camera. You get instant gratification.

We had a pretty tight schedule and a lot to shoot. We tried to think it through: “OK, we need something in each shot that will really show off the slow-mo aspect.” We had a couple of shots early on where there wasn’t enough movement to even register at that speed. The camera shoots at up to 4000 fps. Most of what we shot was 1000 fps, and at that speed a person moving normally almost doesn’t register. So for some of the early shots we realized we needed to add a little more movement, a little more flair.

And if you shoot an action that’s, say, two seconds, that’ll last a long time in playback [each second of 1000 fps capture stretches to a little more than 30 seconds when played back at 30 fps]. We’re creating a bunch of segments that will be 10 seconds at the most. The trickiest part is making sure that out of the 30 or 40 seconds you’re looking at, there are 10 juicy seconds in there.

The images are stark and monochromatic. Is that a characteristic of the high-speed camera or a deliberate design decision?

We definitely were going for a very limited palette. Everything was styled in black, white, silver – and then generally each shot had an accent of red. Color-wise, if you look at skin tones and at the reds we did use, there was pretty true color. We shot a bunch of red balloons and silver glitter, and my sense is that if you shot a very colorful scene you would get true color with it. My general feeling by the end of it was that the camera can do pretty much what any normal film camera, or high-end HD camera, can do.

The one issue we ran into – and I’m not sure whether this was on our end or if it was a function of the camera – because the shutter speed is so fast and the aperture is so big your depth of field is fairly shallow. I imagine there is a way you can trick it and get a deeper field. We weren’t really concerned with that. Some of our shots have a deeper field than others, so you do have some control. My sense from the outside is, overall, the depth of field may be a bit shallower than if you were not shooting at high speed.

There’s a shot of an athlete, and all he does is run toward camera and jump in the air, and we were able to pull focus on this guy running toward the camera. We had drawn the shot up as a guy crashing through a pane of glass, but the expense of the glass and having it shipped – and we’d only have two shots at it because the huge panes of glass were $1,000 each plus shipping – we ended up not doing that. And in the end, the guy jumping through the air was amazing, because of the ability to see how the body moves in the moment.

What were you recording to?

The camera itself is effectively a box. You can get an eyepiece attachment for it, but we basically just had the lens mounted on it, so you’re looking at a box with wires running to a laptop. It’s basically always running. You call action for action, but when you call cut, it’s actually backlogging from where you stopped the cache. At 1000 fps it was roughly a 20-second cache. So you shoot, you say cut, it holds the cache backlogged from where you called cut, and then you pull out the section that you like and set an in point and an out point for the portion of the shot you liked, and then you download that shot. The cache is temporary. After each shot you still need to download the shot.

How do you set the in and out points?

There’s a tech there working on the laptop, and essentially all he’s doing is hitting your out point when you call cut to stop the cache. And then you go into that cached footage and pull out your shot. You could download that whole cache if you wanted to, but to save on download time it helps to set an in and an out point for your shot. One shot had a dog barking, and the dog barks multiple times, so it made sense to just call cut and download the whole shot rather than try to go in and pick out the best bark.

While it’s downloading, we’re also backing up to tape, just in case. That’s a redundancy. It’s a nice thing to have, but not a necessity.

So you leave the shoot with your selects fully digitized?

Yep. With a big drive, just full of clips. Now when the clips come in, the camera has a proprietary codec. In order to get it into a format that we like, like QuickTime, you need to use their proprietary software, which they provide with the camera. You end up with a sequence, and you have to re-import that and then spit it out as whatever format you’re interested in.

Did you hit any roadblocks working with the footage?

There were some glitches. You can output raw images, just like on a digital still camera, so you have a ton of information, but obviously the image size is huge. So there are different types of compression you can use. The compression that we used gave us some artifacts that we were able to pretty easily work around. I hate to say anything bad about the camera, since it was such a great experience, but that’s something to be aware of.