Evolution is an immensely creative process, responsible for amazing creations from the lightening speed of a cheetah to the human brain. Envent is a PC application that uses the principles of evolution and natural selection to automatically develop original, creative circuit designs - from nothing but a "primordial soup" of randomly generated circuits. Without using any advanced knowledge of what an appropriate circuit might look like, it creates unique solutions that, in many cases, would not occur to a human engineer. All you have to do is define the input signals and desired output signals, and it will generate a circuit that fits the bill (at least, as closely as possible – few circuits do their jobs with perfection).

Envent has already recreated several previously patented circuit designs, and had even improved on previously patented circuit designs. Should you develop any patentable circuits, you maintain full patent rights (provided it is done with a properly licensed copy).

Once the parameters are set up, it can run unattended, so you can let it go overnight, and have your results when you get up!

Reference articles

Rocket Scientists Use Darwinian Software to Evolve Better Ion Engine Designs

Robots Evolve Better Homing Techniques

Event mimics the process of evolution, starting from a "primordial soup" of randomly generated circuits*. These will generally perform very poorly, however a few will be better suited the desired task (their environment) than others. These "top dogs" are given a greater chance of passing on their characteristics to the next generation than the rest. On the other end of the spectrum, some will be so poorly designed that they can’t even be successfully analyzed. These circuits are, for all intents and purposes, "killed" and so are not afforded the opportunity to pass their characteristics to the next generation. Then, the metamorphosis begins.

*The user also has the option of "seeding" the evolution with pre-defined circuits that perform the desired function fairly well. Envent will then fill the initial population with those circuits, plus a certain percentage of random variations of them, then fill out the remainder of the initial population with randomly generated circuits. Or, you can simply use it to optimize component parameter values for a given circuit design.

Once the first generation has been assessed and the "fitness" of each circuit is graded, the circuits "reproduce". Reproduction is done in one of several ways – mating, mutation and copying. In the mating process, pairs of well suited circuits generate offspring that are essentially a hybrid of the two parent circuits. In mutation, properties of a single parent are randomly altered to create the child circuit. Copying is exactly what the name implies: the offspring is a duplicate copy of the parent. This is reserved for the best circuits in the generation, and ensures that the evolutionary process doesn’t take a step backwards.

The next generation is then analyzed and graded in the same manner as the first generation, and the process repeats until an acceptable solution is found.