FFmpeg NVDEC Decoder

Purpose: The tool evaluates Nvidia NVDEC-accelerated FFmpeg decoding for a given media input. It targets developers integrating FFmpeg with Nvidia GPUs, media engineers evaluating hardware-accelerated pipelines, QA teams validating codec support, and researchers benchmarking decode throughput. It addresses scenarios where hardware acceleration may be unavailable or misconfigured, providing a structured capability report and actionable guidance for automation. It does not perform actual decoding; it provides a structured capability report and guidance for automation.
Core: The tool relies on a capability matrix and a lightweight throughput model. Required features include: input_path and media properties (width, height, framerate, codec list); GPU target (gpu_model or nvdec_version) and decoding_mode. Outputs include: decode_supported, path (NVDEC, Software, Hybrid), estimated_fps, supported_codecs, and notes. Optional/advanced features include: per-codec performance multipliers, color depth and HDR handling, multi-sample comparisons, and export options (JSON/CSV).
Inputs & Outputs: Inputs include input_path (string), width (int, px), height (int, px), framerate (float, fps), codecs (array of strings), gpu_model (string, optional), nvdec_version (string, optional), decoding_mode (string, required). Outputs include decode_supported (boolean), path (string), estimated_fps (float), supported_codecs (array), notes (string).
Algorithms & Calculations: The system uses a capability matrix and a lightweight model to derive an FPS estimate when NVDEC is available. It normalizes by codec complexity, resolution, and hardware capability, then reports results with caveats. The exact constants are designed for relative guidance and automation workflows.
Error & Edge Cases: Invalid inputs (negative dimensions, zero fps), unknown codecs, missing files, or non-NVDEC-capable GPUs yield safe defaults and warnings in automated pipelines. External data absence results in an informative fallback rather than failure.
Industry/Region & Localization: Units are frames per second and pixels. The model assumes international standards for media encoding and does not depend on a specific locale. Data handling is internal to the tool and respects typical enterprise data security expectations.
Assumptions & Exclusions: Assumes inputs are accessible and valid; no actual decoding is performed; results are estimates intended for planning and automation. It excludes exhaustive benchmarking, memory footprint guarantees, and cross-driver performance comparisons.