Questions, answered honestly.

Because frames go straight from the GPU to the hardware encoder without ever touching the CPU — a zero-copy pipeline that keeps Wispframe under 1% CPU while recording 4K60. And it does that while doing a lot: system audio, live mic mixing, redaction (including on-device face tracking), text and webcam overlays, crash recovery, and iPhone PiP are all built in.

Wispframe is distributed exclusively through the Mac App Store — hit Get, and updates arrive automatically. Payments (Pro annual or lifetime) also run through the App Store, billed by Apple to your Apple Account; we never see your card. macOS still asks for the Screen Recording permission on first use, like for any recorder.

Any Apple Silicon Mac (M1 or newer) running macOS 14+. The zero-copy pipeline is built specifically for Apple Silicon's unified memory, so there's no Intel build.

It's measured, not estimated — via mach_task_basic_info, the same source Activity Monitor uses, while the screen is actively churning at 4K60. The trick is that the CPU never sees a pixel: ScreenCaptureKit hands GPU surfaces directly to the hardware encoder. You can reproduce the benchmark yourself with one command — see the Performance page.

Toggle "System audio" in the main window before hitting Record — done. No loopback drivers, no BlackHole, no audio MIDI setup. From the CLI: wispframe record --display 1 --system-audio.

Yes — toggle both. They're mixed live into a single AAC track, and each source has its own gain slider (0–200%), so you can duck the game audio under your voice.

Two ways: press ⌘⇧. (Command-Shift-Period) from anywhere, or click the timer in your menu bar. The main window disappears during recording on purpose — the menu-bar timer is the only sign it's running.

You lose at most the last two seconds. Recordings are written as fragmented MP4 and finalized every two seconds, so the file on disk is always playable — after a kernel panic, power loss, or force-quit, it opens in any video player like nothing happened. No recovery tool needed.

Yes — every connected display in parallel, each to its own file (wispframe record-all). You can also run several independent recordings simultaneously from the session panel.

H.264 for maximum compatibility, HEVC for smaller files, and ProRes when you're taking the footage into an editor. All encoded in hardware, so the CPU cost is the same: essentially zero.

Open the PiP recorder and your iPhone appears in the camera picker automatically via Apple's Continuity Camera — no cables, no apps to install. Requirements: iPhone with iOS 16+, same Apple ID, Wi-Fi and Bluetooth on.

Check the Continuity Camera basics: both devices signed into the same Apple ID, Bluetooth and Wi-Fi enabled on both, iPhone unlocked and near your Mac, and "Continuity Camera" enabled in the iPhone's Settings → General → AirPlay & Handoff. Then reopen the camera picker.

Any of the four corners — pick one before recording. The overlay is composited into the file during encoding, so there's nothing to line up in post.

Studio turns a raw recording into the polished, framed look you see in high-end product demos: your video is shrunk and centered on a padded canvas with rounded corners and a soft drop shadow, over a backdrop of your choice — six gradient presets, a solid color, your own image, or a blurred copy of the clip itself. It's baked into the output file. From the CLI: wispframe studio --in raw.mp4 --out framed.mp4 --bg gradient:Dusk.

Use Redact: drag a box over the region (API keys, emails, names), choose blur or solid blackout, and it's encoded into the output pixels. It is not a reversible filter or a separate layer — the original pixels are gone from the file.

Redact can find faces for you. Blur every face automatically, or click just the ones that matter — each face is detected on-device with Apple Vision and tracked frame by frame as it moves, so it stays covered even when people walk around. Combine it with drawn boxes; pick blur or solid blackout. Nothing leaves your Mac.

Yes. Pick 2–9 displays, windows, or apps in the app (they get numbered badges), choose a layout — grid, side-by-side, rows, PiP, or speaker — and record them in parallel. When you stop, they're auto-merged into one video, carrying the first source's audio. For pixel-perfect placement, the video mixer lets you drag, resize, and overlap sources on a canvas up to 8K, each with its own opacity.

Add text overlays with position, size, and color — emoji render in full color via CoreText. From the CLI: wispframe compose --base in.mp4 --text "LIVE 🔴:100,50:64" --out out.mp4. Redactions always draw above text, so a caption can never un-hide something.

wispframe compose-grid takes 2–9 sources and one layout flag: grid, side-by-side, rows, PiP corners, or speaker view. For precise control, compose places an overlay at an exact pixel rect and time window.

Yes — from the Done row after a recording, or wispframe to-gif --in clip.mp4 --width 480 --fps 15. Great for bug reports and PR descriptions.

Everything — every feature, every codec, and unlimited recording length. The only difference: videos produced by the processing tools (Compose, Studio, Subtitler, Redact, Enhance…) carry a small Wispframe watermark. It's not a trial and it never expires.

Yes — expressly. Free-tier output, watermark included, may be used for any lawful purpose, personal or commercial, and everything you record or export belongs entirely to you on both tiers. The one rule: the watermark itself may not be removed, cropped, or covered — watermark-free exports are what a Pro license buys.

Click the Pro button in the app and choose Annual ($29.99/year) or Lifetime ($79.99, one payment forever). The purchase runs through Apple's in-app StoreKit sheet and removes the export watermark instantly. On a new Mac, sign in with the same Apple Account and click Restore Purchases. The annual plan cancels anytime from the App Store's subscription settings.

Yes — through Apple, which took the payment and is the seller of record. Request it at reportaproblem.apple.com with the Apple Account you bought with; there's no retention script, because the purchase never went through us. EU/EEA customers have a statutory 14-day withdrawal right on top of that. Elsewhere Apple decides case by case — if it declines and you think that's wrong, email us and we'll back your request. Details on the Refund page.

No. Recordings stay on your Mac, full stop. There is no telemetry, no usage analytics, and no crash reporting phoning home. Pro purchases are handled entirely by Apple's StoreKit — we never see your payment details, and the app sends nothing about what you record.

Screen Recording (required — that's the product), Microphone (only if you record your mic), and Camera (only for PiP). macOS asks for each the first time it's needed, and the app deep-links you to the exact System Settings pane.

The site uses cookie-free, anonymous Vercel Analytics — no cookies, no fingerprinting, no cross-site tracking, nothing to consent to. Details in the Privacy Policy.