The roots of military night vision reach back to the 1930s, when research into television technology produced a tube capable of converting infrared images into visible displays. American armed forces based their first night-vision devices, used in World War II and the Korean War, on this Generation 0 technology [sources: Brickhouse; National Archives].
While a worthy first attempt, the bulky Gen 0 scope had problems, like heavy batteries and a beam that might give away a sniper's position if the enemy was packing an IR sensor [sources: Brickhouse; National Archives]. So much for stealth recon.
By the Vietnam era, the army had moved on to Generation 1 devices, aka "starlight" scopes, which amplified ambient light to turn blackish night into greenish day [source: National Archives]. In a starlight scope, light passes through a lens made up of optical fibers, then strikes a photo-emissive element -- a light-sensitive material that converts photons into electrons. These electrons feed into a series of tiny devices similar to TV picture tubes, each one multiplying the image brightness. It's a bit like pointing a video camera at a television, turning up the brightness, and then repeatedly feeding the camera image back into the TV (which probably wouldn't work, but you get the idea). Gen 1 starlight scopes could amplify the ambient brightness around 40,000 times [source: National Archives].
Gen 1 scopes issued in Vietnam were good to a range of up to 500-1,000 meters (1,640-3,281 feet) in near total darkness, but they remained too pricey and awkward to be considered practical for hand-held weapons. Another knock against them: Shooters could suffer brief blindness from flashes sparked by tracer fire or Dragon anti-tank missiles [source: National Archives].
Light amplification continued as the focus with Generation 2 scopes. These lighter, smaller and cheaper tubes replaced the Gen 1 "mini TVs" with a single electronic device called a micro channel plate (MCP), which output as many as 10,000 electrons for each electron they received as input. These output electrons struck a phosphor screen at the tube's viewing end, generating an image [source: National Archives].
MCP scopes were roughly a quarter the size and half the weight of Gen 1 scopes, and they provided both a sharper image and less distortion than their predecessors [source: National Archives]. Their smaller, more compact design also made the first night-vision goggles possible.
Generation 3 got even better, producing MCP scopes with improved resolution and sensitivity. A metal-oxide ion barrier film on the MCP extended its life span to around 10,000 hours, compared to the 2,000-4,000 hours possible with Gen 2 tubes [sources: Brickhouse; Morovision].
Generation 4 devices removed the ion barrier film to diminish halos and boost sensitivity, signal-to-noise ratio and resolution, but a high failure rate spurred the armed forces to retire the designation for the time being. Some companies, somewhat controversially, continue to apply the term, however [sources: Brickhouse; Morovision].
As the army perfected its tech, it continued to trickle into the consumer sensor and photography markets, where other kinds of shooters thought of new applications for the newly accessible wavelengths.