The vacuum tube, used up to this time in almost all the computers and calculating machines, had been invented by American physicist Lee De Forest in 1906. The vacuum tube, which is about the size of a human thumb, worked by using large amounts of electricity to heat a filament inside the tube until it was cherry red. One result of heating this filament up was the release of electrons into the tube, which could be controlled by other elements within the tube. De Forest's original device was a triode, which could control the flow of electrons to a positively charged plate inside the tube. A zero could then be represented by the absence of an electron current to the plate; the presence of a small but detectable current to the plate represented a one.Vacuum tubes were highly inefficient, required a great deal of space, and needed to be replaced often. Computers of the 1940s and 50s had 18,000 tubes in them and housing all these tubes and cooling the rooms from the heat produced by 18,000 tubes was not cheap. The transistor promised to solve all of these problems and it did so. Transistors, however, had their problems too. The main problem was that transistors, like other electronic components, needed to be soldered together. As a result, the more complex the circuits became, the more complicated and numerous the connections between the individual transistors and the likelihood of faulty wiring increased.