Repository navigation
[1.1.2] - The "Pinpoint Diagnostics" Release
This is a major stable release that consolidates the stability improvements from v1.1.1 and introduces advanced hardware diagnostic capabilities.
The update includes a new vector-based grading engine and a refined user interface.
Added
Pinpoint vector grading
The grading engine has been completely redesigned.
Instead of simple threshold checks, the system now evaluates multiple independent health vectors (Life, Writes, Hours, Sectors) and identifies the exact trigger responsible for the final grade.
Dynamic diagnostic highlighting
The S.M.A.R.T. window now includes an integrated statistics panel.
The attribute responsible for a grade downgrade is dynamically highlighted in the interface using the corresponding status color.
Compact diagnostic interface
The S.M.A.R.T. Analysis window has been redesigned to a compact 750×500 layout to improve readability and data density, especially on smaller displays.
Modular architecture
The codebase has been refactored into dedicated modules:
core.py— analysis engine and SMART processingui.py— interface componentsconfig.py— constants and configurationupdater.py— update script
This improves maintainability and internal structure.
Changed
Improved TBW heuristics
Detection for budget SSD controllers (Phison, Silicon Motion) has been refined to improve accuracy in Total Bytes Written (TBW) calculations.
Seagate composite attribute decoding
Stabilized decoding of 48-bit composite attributes used by Seagate drives.
Values are now correctly translated into the industry-style w2/w1/w0 format.
Codebase cleanup
Removed redundant code paths and standardized documentation in English for all low-level logic sections.
Fixed
Mathematical safety guards
Added protection against ValueError in grading logic when processing malformed or empty SMART tables.
Operator consistency
Standardized comparison logic (>=) across the grading engine to ensure predictable evaluation behavior.
Memory management
Resolved an issue where generated QR codes could occasionally fail to render or disappear due to garbage collection.