Programming
Computing that can run backwards.
Reversible computing is built from steps that are one-to-one and invertible, so no information is ever destroyed. Because nothing is erased, Landauer's principle says such a computation need not dissipate the heat that forgetting demands. That is the route to "zero-energy programming" — reversible programming languages, reversible debugging, and adiabatic, energy-recovery hardware.
Latest news
Reversible Computation 2026 convenes in Torino
The field's main annual conference returns on 9–10 July, spanning reversible languages, circuits, quantum computing and the thermodynamics of computation.
ProgrammingVaire moves to its second reversible chip
After net energy recovery on silicon in 2025, the startup turns to a second chip aimed at logic and competitive performance, with a sharper prototype due in 2026.
ProgrammingVaire tapes out a chip that recycles half its energy
A 22nm test chip recovers about half its switching energy through a resonant power clock instead of dumping it as heat.
Explore
Principles →
Landauer's toll, computing without forgetting, and energy-recovery hardware.
Academia →
The foundational papers, from Landauer 1961 and Bennett 1973 onward.
Commercial →
The young industry building near-zero-energy reversible and adiabatic chips.
Open Source →
Reversible programming languages and reverse-debugging tools you can try today.
Key ideas
Logical reversibility
Bennett (1973) showed a computation that throws nothing away — every step one-to-one and invertible — can run forward and then backward to uncompute its scratch work, leaving memory clean.
Landauer's principle
Erasing one bit must release at least kT ln 2 of heat — about 2.75 zJ at room temperature. Avoid erasure and you avoid the toll.
Adiabatic switching
Energy-recovery circuits charge and discharge nodes gently through a resonant clock, returning most of the energy to the supply rather than dumping ½CV² as heat.