Anyone who has sailed, at night knows that night vision is a precious thing. One minute you can see quite clearly, the next - after that clot on the foredeck has shone his torch in your face or someone has opened, the loo door with the light on - you're effectively blind. It can easily take half an hour to recover from a half-second's exposure to light. Wouldn't it be fantastic if we had a cat-like ability to see in the dark without having to rely on lights and torches? Unfortunately, no matter how many carrots you eat, you're never coins to match Titidles' ability to spot a mouse at midnight without some son of technological assistance. But the price of that technology has fallen dramatically over recent years and, if you shop around a bit, its perfectly possible to find night vision devices for less than £100 as well as for over fifty times as much. But do the cheapies really work: And what on earth can justify paying so much more?
 

As one supplier told me when we started collecting equipment for this test: "There's nothing simple about night vision equipment". He was dead right: the technology is highly sophisticated and the underlying science is barely comprehensible to anyone other than a theoretical physicist or the kind of polymath who regards A Brief History of Time as light reading. Yet the basic principle is simple, if you meekly accept that light is made up of minute and mysterious particles called photons. At one end, faint glimmers of light from distant objects are collected by a conventional system of tenses, very much like the objective tens of an ordinary pair of binoculars, and focused onto a thin glass plate, coated with an alloy of rare metals, such as Antimony Caesium and Rubidium. This is the first of the clever bits, because every time this 'detector plate' or 'photocathode' is struck by a photon, it releases an electron. At the other end, the electrons hit a phosphor coated screen - a miniature version of the kind of thing that was at the heart of the old green-screen radars - which converts their energy back into photons, to produce a visible image that you look at through the eyepiece. The only reason it's green, incidentally, is because the human eye is particularly good at seeing green light it could just as easily be red or blue. Of course the only reason anyone bothers with all this business of converting photons to electrons and backagain is because, although a flow of photons cant be amplified, a flow of electrons can - so you get a brighter image at the eyepiece than at the objective. There are two main ways of amplifying the electron flow and it is these that account for the main difference between the relatively cheap 'First Generation' devices and Eater, more expensive versions. The cheap and obvious way is to encourage the electrons to speed up by passing them through a strong electric field. The less obvious, but very much better way, is to increase the number of electrons by passing them through something called a multichannel plate.Essentially, this is a thin glass plate, riddled with countless tiny notes, each about a hundredth of a millimeter in diameter.Although the plate is less than a millimeter thick, each hole is like a miniature tunnel - much longer than it is wide. When an electron enters the tunnel, it ricochets from side to side, releasing more electrons from the surrounding glass. Each of those ricochets around, releasing even more electrons, so that by the time the original electron emerges from the other side of the multichannel plate, it is accompanied by a crowd of other electrons.All of a sudden, with at that going on inside them, second generation image intensifiers don't seen quite so expensive.