Too Much Crypto

[lain-dono]

https://eprint.iacr.org/2019/1492.pdf

• AES: 9 rounds instead of 10 for AES-128, 10 instead of 12 for AES-192, 11 instead of 14
  for AES-256, yielding respectively a 1.1×, 1.2×, and 1.3× speed-up.
• BLAKE2: 8 rounds instead of 12 for BLAKE2b, 7 rounds instead of 10 for BLAKE2s
  (we’ll call these versions BLAKE2bf and BLAKE2sf), yielding respectively a 1.5× and
  1.4× speed-up.
• ChaCha: 8 rounds instead of 20 (that is, ChaCha8), yielding a 2.5× speed-up.
• SHA-3: 10 rounds instead of 24 (we’ll call this version KitTen, inspired by Keccak family
  member KangarooTwelve), yielding a 2.4× speed-up.

Not being an expert in cryptography, I would guess that using modified AES would do more harm than good. In practice, it is easier to make a mistake in an AES implementation than in a ChaCha implementation. However, having a reduced-round ChaCha will allow it to be used instead of AES where performance matters and safety is less important (gamedev-related protocols for example).

At least there are implementations of ChaCha8 and ChaCha12 in the RustCrypto project.