The call came late Friday. Could I shoot a super slow mo shot of a fizzy drink for a European ad agency?

No problem. When would the director arrive in London? " well he's is busy in the edit as the deadline is 7 days away". Could I shoot a single shot as a graphic sting and email a quicktime for approval?

On Monday we began, after spending the weekend emptying local shops of food colouring and soda water. The story board called for a swirling wave of liquid coming from the centre of frame.

The production team had tried this at 120fps on film but preferred the look of a higher speed capture of 500fps.

I chose the APX camera which has 4x3 aspect ratio imager with frame rates up to 2000 frames per second at native 1k x1k resolution. Its download time from camera to computer is fast. It has higher frame rates in 16x9 aspect ratio of 3000 frames per second.

Initially we shot from below a glass tank, with the liquid appearing as a spec and splashing onto the screen. The frames were compressed and emailed to the edit.

Could we create more fizz?

No that from this angle. I suggested we change angle and send them a shot at 2000 fps of lots of splash and fizz.

We rigged and shot this. Client and director loved it but it wasn't quite right for the preceding shot in the commercial.Could we create the fizz and make the liquid come from a white background to fill the screen?

No. The dynamics of moving the liquid or the camera were beyond what we could rig in the time available.

As we had settled on 2000 fps capture as a rate that best captured the splash and fizz we were locked into a duration of subject movement that occurred in real time within just 1/16th of a second.

The maths go something like this. The on screen duration of shot needed to be five seconds. That was fixed. So if we record 2000 fps in one second but only use 125 frames, then those 125 frames were captured in one sixteenth of a second of real time.

Could we move the camera and perform a focus pull within 1/16th of a second? No.

Using formula one triggering equipment from TAG Heur I had achieved good results at 500 frames per second programming my motion control rig to pan with a soccer ball as it was kicked. But that event was a relatively long duration of one second!

So we couldn't achieve exactly what was story boarded using live action. But then the creative team and client really loved the full frame splashes at 2000 fps.

Could we explore a vortex effect with the liquid? We began the following day with compressed air and a selection of pumps and nozzles.

We would shot a setup and email. From shooting a shot to client receiving the email took about 30 minutes.

We gradually fine tuned the dynamics of the liquid until the team realised they had the best effect possible. At the same time as we were shooting they were still refining the edit. We were also working with the special effects supervisor and editor.

When we had a shot everyone was happy with the digital files were converted to HD at Mac Million in London.

As the slow mo camera outputs rgb images grading is necessary. Obviously grading the raw files allows for more scope, however in the time available we graded in HD.

I had controlled the lighting and exposure so the image fell within the dynamic range of the cmos sensor so all that remained was to correct the characteristic green cast and adjust the gamma for TV.

12 setups in two days. Lost count of the number of exploratory shots, perhaps 30.

Unlike a film shoot the client and creative team were not under the pressure that comes with whizzing a roll of 35mm film through a camera. So they treated the lo mo shoot like they would treat a boutique graphics house. Keeping the key personal in touch until they had signed off the shot in the online.

When using frame rates below 500 frames per second it is just possible to see the the movement of the subject. At over 500 it becomes very difficult to see what happens. In our case we needed a piece of suitable action would occurred in just 1/16th of a second.

The collaboration of special effects supervisor editor director and client in no way diluted the creative focus of the brief or hindered the creative boundaries of lighting or cinematography. Even with the client and creative team having an input to the shot during an edit, with the client present and with a deadline looming, the only sticky part of the shoot was the cleanup operation.

From the clients perspective the team were working creatively and someone mentioned "organically".

We are often reminded that the look of digital is not organic,whatever that means, well the technology is not organic but its immediacy creates a fluid exchange of creative responses that certainly is, organic!

Mike Brennan

published in High Definition Magazine 2004