Ballistics and mining engineers have long used high-speed imaging to study how small changes affect the way explosives detonate. Industrial engineers use the cameras to slow down high-speed machines so they can study manufacturing problems. Automotive test engineers examine body panels during a crash, brake rotors under stress, windshield wipers in the rain, and turbulence in engine cylinders. The common thread that runs through these applications is that they all involve solving problems that happen too fast for the human eye to discern. "Slowing down" the image enables an engineer to divide one large problem (e.g., poor "survivability" ratings in a crash test) into several smaller, more manageable ones (e.g., air bags that don't inflate properly plus steering column that doesn't collapse plus bumpers that collapse too well). Choosing a high-speed camera for any of these applications involves weighing frame rate, sensitivity, resolution, time constraints, system integration and cost. If these trade-offs sound familiar, it's because increasing a camera's frame rate only adds a few twists to the standard camera-buying conundrum. Brake Study Stops the Action Scientists at Oak Ridge National Laboratory in Tennessee are using high-speed infrared imaging to study an expensive automotive warranty problem: brake judder.