Skip to content

Library is actually slower than scrypt-js with setImmediate polyfill #17

@desfero

Description

@desfero

It appears the library is actually around 25% slower than scrypt-js when the polifill for setImmediate is provided (otherwise it fallbacks to setTimeout which is slower due to being run in the tasks queue not micro tasks queue).

Input: {"P":"pleaseletmein","S":"SodiumChloride","N":65536,"r":8,"p":1,"dkLen":64}
  scrypt-pbkdf — mean time: 626ms ±1.84% (9 runs sampled)
  scrypt-js — mean time: 485ms ±3.34% (9 runs sampled)
  scryptsy — mean time: 23649ms ±267.28% (8 runs sampled)

Input: {"P":"pleaseletmein","S":"SodiumChloride","N":131072,"r":8,"p":1,"dkLen":64}
  scrypt-pbkdf — mean time: 1292ms ±4.10% (7 runs sampled)
  scrypt-js — mean time: 946ms ±2.58% (9 runs sampled)
  scryptsy — mean time: 10943ms ±186.85% (6 runs sampled)

Input: {"P":"pleaseletmein","S":"SodiumChloride","N":262144,"r":8,"p":1,"dkLen":64}
  scrypt-pbkdf — mean time: 2529ms ±1.60% (6 runs sampled)
  scrypt-js — mean time: 1888ms ±1.64% (7 runs sampled)
  scryptsy — mean time: 5179ms ±0.14% (5 runs sampled)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions