[matthewdgreen]

This isn't really accurate. Fault-tolerant classical computing absolutely does exist as a field, and plenty of work has been done there. Some of the early work was done by von Neumann [1] because early computing hardware was extremely error-prone. Over time it turned out that these techniques were not really needed due to the fact that modern solid-state hardware is extremely reliable. The field of quantum computing actually resurrected a handful of ideas that were originally developed for classical computers.

More generally, nobody needs "perfect" classical computing either to make quantum computing work. Given a (quantum or classical) processor with some degree of error, the idea behind these techniques is to "boost" that into a processor with arbitrarily small error. It just turns out that with modern classical processors the error is so small that we suffer it rather than pay the cost of using these techniques.

[1] https://www.cs.ucf.edu/~dcm/Teaching/COP5611-Spring2013/Pape...