Updates

Changelog

Release notes and version history. Follow along as Duplyfi evolves.

v1.2.0

LatestJune 2026

See how unique every copy is — Duplyfi now measures each output, not just transforms it — with a new Hash Verification check on both video and image, plus image-spoofer upgrades.

Uniqueness Verification

  • Every processed file now gets a uniqueness score, so you can verify the result instead of just trusting it
  • New Hash Verification toggle on both the video and image pages — when on, Duplyfi checks each output is different enough from the original and automatically re-processes any that are too similar
  • A uniqueness report appears after processing with a per-file score

Image Spoofer

  • Image uniquification is now available during the free trial
  • Added image metadata injection (device, location, and timestamp)
  • Refined default image transforms and tuned ranges; effect cards are now clickable

Stability & Performance

  • Fixed an intermittent crash during image processing
  • Faster uniqueness checks

v1.1.2

June 2026

Your free trial now starts when you first open the desktop app, with clearer onboarding, an account-aware upgrade flow, and desktop polish.

Trial & Onboarding

  • Your free trial now starts when you first sign in on the desktop app — not at signup — so none of your trial days are spent before you have it installed
  • The dashboard now clearly guides you to download the app and shows your trial status
  • New setup-reminder emails if your trial has not started yet

Account & Billing

  • New upgrade page with account-aware plans and direct checkout
  • Expired trials now get a clear resubscribe path
  • Smoother email verification with a one-click resend option

Desktop App

  • Processing progress now accurately reflects parallel work ("X of N done · K processing") and never jumps backward
  • Save presets with an in-app dialog instead of a system pop-up
  • Clearer guidance when no output folder is set, plus quick access to your output folder when processing finishes
  • Copy and wording refined across the app

Under the Hood

  • More accurate usage analytics
  • Refreshed app icons and social preview image

v1.1.1

May 2026

Bug fixes for overlay effects, FFmpeg compatibility, metadata handling, and uniqueness scoring improvements.

Overlay Fixes

  • Background overlay now covers the full video frame regardless of resolution or active effects
  • Overlay scales to match post-filter video dimensions using scale2ref
  • Fixed partial coverage on portrait and non-standard aspect ratios

FFmpeg Compatibility

  • Noise filter seed: graceful fallback on FFmpeg 4.x (seed requires 5.0+)
  • Luma curves filter: graceful fallback on FFmpeg 4.x
  • Fixed metadata key truncation (com.apple.quicktime keys were being split)
  • Fixed ISO 6709 location trailing slash breaking FFmpeg on Windows and Linux

Uniqueness Scoring

  • Composite scoring: pHash visual base + bonuses for audio, metadata, timing, and encoding effects
  • Simplified to two tiers (red/green) -- no more ambiguous amber
  • Fixed 0% scores caused by frame sampling past trimmed output duration
  • TikTok Native preset now correctly shows HIGH instead of LOW

Desktop App

  • Version number now reads from package.json dynamically (login, settings, What's New)
  • Separate default output directory for image processing (independent from video)
  • Lifecycle email schedulers disabled in development mode

v1.1.0

May 2026

Major update adding 4 new video effects, parallel processing with auto-scaling, GPU acceleration detection, 5 agency-specific presets, and feature gate cleanup removing all daily/batch limits from paid tiers.

New Video Effects

  • Spectral tilt equalizer — supplements audio fingerprint modification for improved audio signature variation
  • Luma curves micro-adjustment — randomized per file for subtle luminance variation
  • Perspective filter — sub-visual spatial warp, less than 1.5% visible distortion
  • QP variation per file — randomizes H.264 qmin/qmax for encoder-level uniqueness

Parallel Processing

  • Concurrent processing using worker semaphore — multiple files process simultaneously
  • Auto-scaling worker count based on available RAM and CPU cores
  • RAM detection at startup to cap workers on low-memory systems
  • GPU acceleration detection: NVENC (NVIDIA), VideoToolbox (macOS), VAAPI (Linux) with graceful fallback
  • Perspective filter availability probe at startup — disabled gracefully if unavailable

Agency Presets

  • 5 new built-in presets tuned for multi-platform distribution
  • Standard (All Platforms) — balanced settings for multi-platform distribution
  • TikTok Heavy — deeper processing for TikTok content
  • Instagram — tuned for Instagram Reels and image content
  • Reddit Safe — conservative settings optimized for Reddit upload requirements
  • High-Volume Speed — optimized for bulk processing with reduced encode overhead

Feature Gate Cleanup

  • Removed all daily processing limits from paid tiers (Starter, Pro, Agency)
  • Removed all batch size limits from paid tiers
  • Preset limits removed from paid tiers
  • Free tier remains time-gated (trial duration) — all features available during trial

Effect Improvements

  • Per-file random seed added to noise filter for distinct noise patterns per output
  • aecho reverb decay parameter tuned to 0.4–0.6 range for more natural audio
  • TikTok preset cutVideo/cutEndVideo range increased to 3–7%
  • iOS software version string fixed to x.y.z format
  • ISO 6709 location metadata field added
  • QuickTime creationdate metadata field added
  • Samsung device/codec alignment fix — S-series now uses HEVC

Platform

  • Version bumped to 1.1.0
  • Auto-update flow verified end-to-end from v1.0.0

v1.0.0

Initial ReleaseMay 2026

The first public release of Duplyfi. A complete video and image repurposing engine built for operators who scale across every short-form feed.

Video Repurposer

  • 35 uniquification effects across 5 categories (Color, Motion, Effects, Audio, Encoding)
  • Batch processing with configurable output counts
  • Multi-threaded processing pipeline for fast rendering
  • Randomized output filenames for untraceable outputs
  • Support for all major video formats (MP4, MOV, AVI, MKV, WebM)

Image Uniquifier

  • Pixel-level noise injection for binary uniqueness
  • Metadata stripping and rewriting
  • Batch image processing
  • Support for JPEG, PNG, WebP, and TIFF

Preset System

  • 5 built-in presets: Gentle, Moderate, Aggressive, Audio Focus, Speed Ramp
  • Custom preset creation and editing
  • Preset export and import (JSON)
  • Preset duplication and deletion

Platform

  • Desktop app for Windows, macOS, and Linux
  • Account-based licensing with free trial
  • Secure JWT RS256 authentication
  • Hardware ID binding per subscription tier (Starter 1, Pro 3, Agency 10)
  • Auto-updates via electron-updater
  • Local-first processing (files never leave your device)