[Cryptography] Some quantum computers might need more power than supercomputers

[Peter Gutmann]

Jan Dušátko <jan at dusatko.org> writes:

>May this work should be useful
>
>Estimating the Energy Requirements to Operate a Cryptanalytically Relevant Quantum Computer https://media.licdn.com/dms/document/media/v2/D4E1FAQHzi9RTyvl3lg/feedshare-document-pdf-analyzed/B4EZ18e.bvJAAY-/0/1775909989427e=1776902400&v=beta&t=srE4GEiT0eANSV2erykLD4aOHQkZG2wON6vfx33d_g0

It's definitely useful, but possibly not in the way you were thinking:

  We did not attempt to estimate the (enormously high) upfront cost to build
  or purchase a CRQC or the ongoing costs of maintenance, operator labor,
  administration, and other variables.

  With so little concrete data available, we are forced to be highly
  speculative and to extrapolate our very few empirical data points far beyond
  the regimes where they are likely to hold.

  It is entirely reasonable to question the utility of this final estimate,
  given the number of separate assumptions, approximations, and extrapolations
  that went into it – some of which have orders of magnitude of uncertainty.

Kudos to the authors for being up-front about the level of uncertainty
involved.

Peter.