{
  "status": "shipping",
  "advantages": [],
  "limitations": [],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-11-17",
  "slug": "fpzip",
  "term": "fpzip",
  "aliases": [
    "FPZIP compression"
  ],
  "definition": "fpzip is a predictive codec for spatially correlated floating-point arrays that supports lossless encoding and optional low-bit truncation.",
  "note": "Mechanism: A scan through a regular array predicts each value from previously reconstructed neighbors, then encodes a correction in a mapped integer representation. The 1.3.0 build selects floating-point or integer prediction arithmetic. [implementation]\n\nPrecision: The prec field counts retained representation bits, not mantissa bits. For binary32, prec=24 preserves one sign bit, eight exponent bits and 15 stored fraction bits. For binary64, prec=32 leaves 20 stored fraction bits. Zero selects full precision. [api] [code]\n\nRange: Version 1.3.0 accepts 2-32 bits for binary32 and even counts from 4-64 for binary64, plus zero. Below 9 or 12 respectively, truncation reaches the exponent. This is neither an absolute-error target nor a fixed compressed size. [api] [code]\n\nObserved bytes: An eight-value binary32 fixture takes 36 bytes at prec=9 but 37 at prec=8, including the stream header. Reduced precision need not reduce every small stream. The full-precision fixture takes 59 bytes for 32 raw bytes; this is a constructed example, not a workload estimate. [probe]",
  "related": [
    "fpc",
    "lorenzo-predictor",
    "lossless-compression",
    "zfp"
  ],
  "citations": [
    {
      "id": "api",
      "title": "fpzip 1.3.0: API precision range, headers and floating-point portability",
      "url": "https://github.com/LLNL/fpzip/blob/4a539c06d98b1c029b08324a086d4b75689a2b72/include/fpzip.h",
      "accessed": "2026-11-17"
    },
    {
      "id": "code",
      "title": "fpzip 1.3.0: representation mapping and bit truncation",
      "url": "https://github.com/LLNL/fpzip/blob/4a539c06d98b1c029b08324a086d4b75689a2b72/src/pcmap.inl",
      "accessed": "2026-11-17"
    },
    {
      "id": "implementation",
      "title": "fpzip 1.3.0: scan order, predictor modes and encoded corrections",
      "url": "https://github.com/LLNL/fpzip/blob/4a539c06d98b1c029b08324a086d4b75689a2b72/src/write.cpp",
      "accessed": "2026-11-17"
    },
    {
      "id": "probe",
      "title": "Hesela: fpzip precision ledger and exact-arithmetic Lorenzo examples",
      "url": "https://hesela.dev/research/fpzip-precision/README.md",
      "accessed": "2026-11-17"
    },
    {
      "id": "project",
      "title": "LLNL: fpzip project and Lindstrom-Isenburg 2006 publication, DOI 10.1109/TVCG.2006.143",
      "url": "https://computing.llnl.gov/projects/fpzip",
      "accessed": "2026-11-17"
    }
  ],
  "provenance": [
    "Pinned fpzip 1.3.0, FPZIP_FP_SAFE, GNU 13.3.0; sixteen constructed configurations agree byte-for-byte between Release and Debug.",
    "The native probe covers finite binary32/binary64 values and signed zeros. It does not establish cross-platform compatibility or cover infinities, NaNs and subnormals.",
    "Prediction arithmetic and format compatibility depend on build configuration; decoding requires the matching metadata and codec. [api]"
  ]
}
