• Optimize cloud and NAS sources before export — faster, more reliable processing for iCloud Drive, LucidLink, and network footage • Export presets for YouTube (SDR 16:9) and Instagram/Reels (9:16) — delivery-ready settings in one tap • In-app batch summary: see your processing report, failed items, and total time right in the app • Preflight warnings flag slow-path conditions before processing starts • Clearer per-item error messages for codec, source, and permission problems

• ProRes RAW color now matches professional editing apps • New Look presets — Standard, Flat (for grading), and Punchy • HDR (HLG/Dolby Vision) footage now converts properly to Rec.709 • Fixed: the output folder you choose is always the one used

• Better ProRes RAW color — no more blown-out highlights, full dynamic range • Fixed color auto-detection for drone and third-party footage • Open Gate workflow: shoot 4:3, deliver 16:9 with AI smart crop • Stability and reliability improvements

More sources, identified correctly. Apple Log 2 is now detected using the authoritative metadata key Apple writes into every ProRes file — so footage tagged with LogTransferFunction stays "Apple Log 2" regardless of how it was encoded or re-muxed. Earlier versions relied on color primaries inference and could occasionally misread Apple Log 2 as "Linear." Drone footage and ffmpeg-encoded ProRes now classify correctly. Many cameras and tools write color information in the legacy nclc container format, which AVFoundation's standard API doesn't expose. v1.4 reads the colr atom directly, so DJI ProRes exports, ffmpeg-encoded clips, and legacy camera footage no longer show as "unknown color." Rec.709 recordings from external recorders like Atomos Ninja are now identified by name instead of showing as "unknown color." These files set the color primaries and matrix tags but omit the transfer-function tag — v1.4 falls back to the primaries when the transfer tag is missing. The color-detection badge now shows how each clip was identified — "via Log key," "via metadata," or "inferred" — so you can see at a glance whether a source has reliable color tags before you convert.

v1.3 is a performance and color-quality update for anyone shooting in Apple Log. Batches finish faster on Apple Silicon. M-series Pro, Max, and Ultra Macs now process two clips at once automatically. And when you're exporting one source to multiple aspect ratios in the same batch — 16:9 plus 9:16 for social, for example — the source is read once and fed to every output, instead of starting fresh for each. Sharper HEVC quality from Apple Log and ProRes RAW. Outputs now preserve the full 10-bit color precision all the way through the encoder. Earlier versions silently dropped to 8-bit at the last step, which softened gradients and shadow detail. The fix is automatic on every 10-bit source. Smart crop runs once per video instead of repeating for every aspect ratio you export. Multi-platform social cuts get the time savings without any change in framing accuracy. Progress bars now fill smoothly in one pass without bouncing, and the remaining-time estimate adapts as the batch moves through. The in-app help section has been refreshed to match every current feature. Smaller fixes: iPhone 17 Pro is now correctly identified by name, file access through the sandbox is more reliable on cloud-mounted volumes, and multi-output exports keep audio consistent across every aspect-ratio variant.

NEW: Audio mode picker Settings (macOS sidebar, iOS Settings sheet) now exposes an **Audio** picker with four modes. The selection is recorded in each item's processing log alongside the existing video parameters. - **Pass through (stereo)** — re-encodes the source's audio track to AAC at **192 kbps, 48 kHz, stereo**. This is the default. The bitrate was lowered from 256 kbps in v1.1 to a social-media-friendly value that remains transparent for typical content. - **Mono — single channel** — true 1-channel AAC at **128 kbps, 48 kHz**. AVFoundation downmixes L+R during decode. Smallest file with audio retained; ideal for voice, dialog, and podcast cuts. - **Mono — duplicated stereo** — sums L+R per sample frame and writes the average to **both** channels of a stereo AAC track at **192 kbps, 48 kHz**. Solves the "voice was only on the left mic, but I want it on every speaker" case — lavalier on one channel, single shotgun plugged into the L input, etc. - **Remove audio** — output has no audio track at all. Smallest file overall; useful for muting B-roll or stripping unwanted ambient. ### When to use what - **Pass through (stereo)** for general delivery and any clip where the source mix is already correct. - **Mono — single channel** for voice, dialog, and podcast-style cuts where stereo adds no information. - **Mono — duplicated stereo** for single-mic interview footage where one input is silent and you want the dialog to play on every speaker. - **Remove audio** for silent B-roll, montages, or anywhere the source audio is not wanted. NEW: Smarter per-item progress The progress display in the queue used to climb to 100% and reset for every aspect ratio, so a single source with four ratio variants would appear to start over four times. v1.2 maps the entire run — every smart-crop scan plus every aspect-ratio export — onto a single 0–100% curve, and adds a live activity label under each item. - One smooth fill per source. A four-ratio item fills the bar **once**, not four times. A two-ratio item finishing its first export sits at 50%, not back at 0%. - Live activity label under the file metadata while processing, e.g. **Converting 16:9 (2 of 4)**. - Smart-crop saliency scans now surface as their own status (**Analyzing for smart crop (9:16)**) so the brief pause before each encode is no longer silent. - The step counter (`(N of M)`) only appears when an item has more than one variant; single-ratio items show just the activity label.