Hawkynt

SFC64

Small Fast Counting (SFC64) is a high-quality 64-bit PRNG by Chris Doty-Humphrey designed for the PractRand test suite. It combines a chaotic invertible mapping with a counter to guarantee no small cycles and passes rigorous statistical tests.

Properties

Property	Value
Category	Random Number Generators
Sub-category	Pseudo-Random Number Generator
Security status	🎓 Educational Only
Complexity	Intermediate
Inventor	Chris Doty-Humphrey
Year	2010
Origin	🇺🇸 United States
Source	algorithms/random/sfc64.js

Parameters

Parameter	Supported values
Seed sizes	1 byte (8 bits) to 24 bytes (192 bits)

Capabilities

Flag	Value
IsDeterministic	Yes
IsCryptographicallySecure	No

Security

Status: 🎓 Educational Only

No vulnerabilities are recorded for this implementation.

Documentation

• PractRand: Chris Doty-Humphrey’s PRNG Test Suite
• PractRand RNG Engines Documentation
• NumPy SFC64 Implementation
• NumPy SFC64 Documentation

References

• NumPy Random BitGenerators
• PractRand Statistical Test Suite
• Zig Standard Library SFC64
• TestU01 Statistical Testing

Test vectors

6 vectors ship with this algorithm and run in the test suite. Byte values are hexadecimal.

Vector 1 — Seed (0,0,0,1) after 12-iteration warmup: outputs 13-28 (128 bytes) - verified against Zig stdlib

Field	Value
seed	00000000000000000000000000000000000000000000000000000001
skip	12
outputSize	128
input	null
expected	3acfa029e3cc6041f5b6515bf2ee419c 1259635894a29b610b6ae75395f8ebd6 225622285ce302e2520d28611395cb21 db909c818901599d8ffd195365216f57 e8c4ad5e258ac04a8f8ef2c89fdb63ca f9865b01d98d8e2f46555871a65d08ba 66868677c6298fcd2ce15a7e6329f57d 0b2f1833ca91ca794b0890ac9bf453ca

Vector 2 — Seed (1,0,0,1): Minimal seed with a=1 - raw outputs 1-10 without warmup (80 bytes)

Field	Value
seed	000000000000000100000000000000000000000000000000000000000001
outputSize	80
input	null
expected	00000000000000020000000000000002 00000000000000150000000012000028 00120000240240d4002402417102542b 021302560b404ce517f168db1b98d28a cf0ac2a06233f8978d741151c5de2e2d

Vector 3 — Seed (0x123456789abcdef0, 0xfedcba9876543210, 0x1111111111111111, 1): Mixed values - raw outputs 1-8

Field	Value
seed	123456789abcdef0fedcba98765432101111111111111111000000000001
outputSize	64
input	null
expected	1111111111111101985cfaa8bef49231 ccbddddddddddd4fc1ad5876af21abb3 e2706544fc42af2946e2915c8ab00489 54ac4021de481ea65c9f963e576504a5

Vector 4 — Seed (0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 1): Maximum values - raw outputs 1-6

Field	Value
seed	ffffffffffffffffffffffffffffffffffffffffffffffff00000001
outputSize	48
input	null
expected	ffffffffffffffffffdffffffffffff9 ffdffffffffffff9febffffff6ffffcd fd9723ffca000004f34a23ffaf059595

Vector 5 — Seed (12345, 67890, 11111, 1) after 12-iteration warmup: outputs 13-24 - production-ready sequence

Field	Value
seed	000000000000303900000000000109320000000000002b6700000001
skip	12
outputSize	96
input	null
expected	69550d9d4d386db6eca597ffdc9a4c92 c6eed53dd7b6aec759500586198793fd c856e96e59e62425b285cba09b53e356 393f3e0474f39ebaa1bceb9afa2369a5 ca966f721b7991440018b8868727b44e b946eeae735ae30f83cfabf75c47a394

Vector 6 — Seed (0,0,0,1): Raw outputs 1-8 without warmup - demonstrates initial state evolution

Field	Value
seed	00000000000000000000000000000000000000000000000000000001
outputSize	64
input	null
expected	00000000000000010000000000000002 000000000000000c000000000900001f 000900001b012083001b0120cf024a89 0120024bc721e0b80d0b3628ed8124b6

---

← All algorithms

This site is open source. Improve this page.