Cheap Shotcut Randomizer

Desktop app that shuffles and generates randomized playlists from Shotcut projects, then batch-renders them through melt with a full render queue.

Note: AI upscaling and frame interpolation features live in CheapUpscaler.

Features

Randomizer

• Shuffle playlists — randomly reorder clips per playlist, optionally avoiding consecutive clips from the same source
• Generate smart compilations — simulated-annealing selection from multiple source playlists with duration and clip-count weights, plus a per-playlist target duration
• Non-destructive — generated projects are written as separate temp files; your source project is never modified

Render Queue

• Queue-first workflow — the queue is the home page; a full-page stepper (Source → Video → Audio → Tracks & Range → Review) configures each job
• Encoder selection — x264/x265 on CPU, or your detected hardware encoders (NVENC/QSV/AMF) with the correct per-encoder quality flags
• Stock export presets — the MLT/Shotcut preset library (YouTube, HEVC, ProRes, …) found next to your melt install, searchable with recommended picks on top
• Output overrides — resolution, aspect ratio, frame rate, and audio quality-based VBR
• True pause/resume — pausing a render suspends the melt process in place; resuming continues without losing progress
• Pre-flight checks — missing media, duplicate output targets, and low disk space are caught before the render starts
• Post-completion actions — move the output to a folder or reveal it in Explorer when a job finishes
• Reliability — persistent SQLite queue, crash recovery, safe-mode retry, keep-awake during renders

Find Files

• Search clips across playlists by filename, spot duplicates, and jump to every occurrence on the timeline

Usage

1. Randomizer — load your .mlt project, shuffle a playlist or generate a compilation; both hand off to the render stepper with the generated track pre-selected
2. Add Files on the Render Queue page opens the same stepper for any .mlt project
3. Start Queue — jobs process in the background with live progress, ETA, and per-job actions

Generated projects are stored in the app's temp folder and cleaned up automatically once their render finishes.

Requirements

• Windows 10/11
• Shotcut (provides melt and the export preset library)
• FFmpeg/FFprobe (bundled with Shotcut and SVP; auto-detected, or set paths in Settings)

Building

Prerequisites: .NET 11 SDK

dotnet build
dotnet test

Publish a self-contained single-file build (~90 MB, no .NET needed on the target machine):

.\deploy\publish.ps1 -Version 1.0.0

Output: deploy\out\CheapShotcutRandomizer\CheapShotcutRandomizer.exe

Tech Stack

• Blazor + Avalonia desktop via CheapAvaloniaBlazor
• MudBlazor UI
• Entity Framework Core + SQLite (render queue persistence)
• melt/MLT (rendering), invoked with Shotcut-parity consumer properties
• CheapHelpers (utilities, media processing)

Related Projects

• CheapUpscaler — AI upscaling (Real-ESRGAN, Real-CUGAN) and frame interpolation (RIFE)
• CheapAvaloniaBlazor — Blazor desktop framework
• CheapHelpers — shared utility library