Lazy_MkPFS

A fast, automated, and user-friendly PS5 PFS image builder.

Disclaimer: This is a personal project and is not intended to be a replacement for the original tool. It is built as a customized, streamlined, and optimized version designed for ease of use, speed, and seamless integration into other Python projects.

Note on macOS: macOS support is currently implemented in the codebase but has not been officially tested yet. Use at your own risk.

Based On

MkPFS by PSBrew

Key Upgrades and Features

Compared to the original project, Lazy_MkPFS includes several major quality-of-life and performance upgrades:

• Massively Faster Performance: Optimized Linux directory scanning (os.scandir), direct streaming compression (no temporary spool files for the final image), and native OS-pipe IPC for multiprocessing.
• Advanced Compression Backends: Added support for zlib-ng and Intel ISA-L (zlib-isa) for significantly faster compression speeds.
• Cross-Platform exFAT Creator: Automatically generates and wraps exFAT images for maximum PS5 compatibility.
• Windows Specific Improvements: Utilizes OSFMount for raw image mounting, executes the Windows format command silently in the background, and implements an automatic retry loop during unmounting to prevent "Access Denied" errors caused by Windows Explorer or Antivirus locking the volume.
• Low RAM Verification: Streams and hashes files block-by-block during verification, preventing MemoryError crashes on massive images.
• Batch Processing: Easily pack an entire folder of games concurrently with a single command and clean multi-progress tracking.
• Auto-Dependency Installation: Automatically detects and installs missing core dependencies (like cryptography) on the first run.
• Modular and Importable: Cleanly packaged as a lazy_mkpfs module, making it trivial to import and use the packing logic in your own Python scripts.

---

Installation

Clone the repository and run any script. The tool will automatically install required core dependencies (like cryptography) on the first run.

git clone https://github.com/Nazky/Lazy_MkPFS.git
cd Lazy_MkPFS

Terminal usage (CLI)

1. Pack a Game Folder

# Pack a folder using Intel ISA-L (fastest) with the exFAT wrapper (Default)
python main_folder.py "./Games/PPSA01474" "./Output/RAC.ffpfsc" --zlib-isa --zlib-level 3

# Pack a folder using zlib-ng WITHOUT the exFAT wrapper (creates standard .ffpfs)
python main_folder.py "./Games/PPSA01474" "./Output/RAC.ffpfs" --zlib-ng --no-exfat

2. Pack a Single File

# Pack a single .exfat or .ffpkg file into a compressed .ffpfsc image
python main_file.py "./game.exfat" "./game.ffpfsc"

3. Batch Pack Multiple Games

# Pack all games inside a directory concurrently
python main_batch.py "./Games" "./Output" --zlib-ng --zlib-level 5 --workers 4

4. Verify an Image

# Verify the integrity and hashes of a packed image
python main_verify.py "./Output/RAC.ffpfsc"

CLI Arguments Reference

Argument	Applies To	Description
source_folder / input_folder	folder, batch	Source directory containing the game or games.
source_file	file	Source file to pack (.exfat or .ffpkg).
output_image / output_folder	all	Output path for the image or batch results.
--zlib-level [1-9]	folder, file, batch	Zlib compression level (default: 6).
--zlib-ng	folder, file, batch	Use the zlib-ng backend for faster compression.
--zlib-isa	folder, file, batch	Use the Intel ISA-L backend (fastest).
--min-compress-size [MB]	folder, file	Minimum file size in MB to attempt compression.
--no-exfat	folder, batch	Disable exFAT wrapper (creates .ffpfs instead of .ffpfsc).
--cpu-count [N]	folder, file, batch	Number of CPU cores to use (0 = auto).
--workers [N]	batch	Number of parallel workers for batch processing.
--no-ram	folder, file, batch	Disable RAM-based writing (forces disk spooling).
--ekpfs [HEX]	verify	EKPFS key in hex (for encrypted images).
--new-crypt	verify	Use alternate newCrypt key derivation.
--verbose, -v	all	Enable verbose output.
--quiet, -q	all	Suppress non-essential output.

Python API (Importing)

Because Lazy_MkPFS is fully modular, you can easily import its core functions into your own Python projects.

from lazy_mkpfs import pack_folder, pack_file, pack_batch, verify_pfs
from pathlib import Path

# ==========================================
# 1. Pack a Folder
# ==========================================
stats = pack_folder(
    source_folder="./Games/PPSA01474",
    output_image="./Output/RAC.ffpfsc",
    zlib_backend="zlib",  # Options: "zlib", "zlib-ng", "isa-l"
    zlib_level=3,
    exfat=True,            # Creates the exFAT wrapper automatically
    verbose=True
)
print(f"Completed in {stats.elapsed_seconds:.2f}s | Gain: {stats.actual_gain_pct:.2f}%")

# ==========================================
# 2. Pack a Single File
# ==========================================
stats = pack_file(
    source_file="./game.exfat",
    output_image="./game.ffpfsc",
    zlib_backend="zlib-ng",
    zlib_level=6,
    compress=True
)

# ==========================================
# 3. Batch Pack Multiple Games
# ==========================================
results = pack_batch(
    input_dir="./Games",
    output_dir="./Output",
    workers=4,             # Number of parallel workers
    zlib_backend="isa-l",
    zlib_level=3,
    exfat=True             # Apply exFAT wrapper to folders
)
print(f"Success: {results['succeeded']} | Failed: {results['failed']}")

# ==========================================
# 4. Verify an Image
# ==========================================
verify_pfs(
    image="./Output/RAC.ffpfsc",
    verbose=True
)

---

Credits

• PSBrew for the original MkPFS project and PFS format research.
• pycompression for the amazing compression libraries:
  • python-isal (Intel ISA-L)
  • python-zlib-ng
• Declan Kerr for the windows testing.

---

TODO

• Clean the code
• Fix MacOS support
• Make a GUI (peoples love GUI for some reasons)
• Test with big games (this project only have been tested with small games, from 3gb to 50gb)