Cinesalve

Recover a video recording your camera never finished writing. macOS app.

cinesalve.honorboxx.workers.dev

A flat battery, a pulled card or a crash leaves every frame on the card and only the index missing. Players report moov atom not found and refuse to open the file. Cinesalve rebuilds the index and the recording plays again.

• Every frame exact. A recovered frame lands at the same byte offset and the same size the camera originally wrote, and decodes byte-identical to the original.
• GoPro needs nothing but the broken file. Those recordings carry their own codec configuration, in the H.264 and the HEVC modes the range records, and Cinesalve reads it from the damaged file itself; the sound too, since the camera names every sample ahead of it and its own interleave states the rate.
• Nothing is uploaded. Footage never leaves your Mac. No account, no telemetry, works offline.
• One-time $69. Every Mac you own, for good. No subscription.

Download · cinesalve@proton.me

Check your file free, in your browser: cinesalve.honorboxx.workers.dev/check reads your damaged file and reports exactly how many frames are recoverable and how many seconds that is. Nothing is uploaded — the verdict reads the first and last megabyte, the frame count up to 100.7 MB of the footage and a reference clip in full if you add one — and it runs on any computer, Windows included.

More on the failure itself

• How to fix a corrupted video file on a Mac
• The power went and now the video will not play
• What "moov atom not found" means and how to get the footage back
• Repairing an MP4 without a reference file
• Your GoPro file will not play
• Your DJI file will not play
• Your Canon file will not play
• Your iPhone video will not play
• Your Sony recording will not play
• Your Pixel video will not play
• Your dashcam video will not play
• OBS crashed and the recording will not play
• Getting untrunc working on a Mac
• My test suite was lying to me

Recovered from the damaged file alone

No reference clip, nothing but the recording that will not play. Every sample exact against the intact original, the codec configuration reconstructed byte-identical to the real one, and zero decode errors from ffmpeg reading the result end to end.

real camera	frames recovered	sound	exact	decode errors
GoPro HERO5	456 / 456	892 / 892	100%	0
GoPro HERO6	425 / 425	665 / 665	100%	0
GoPro HERO6, second take	263 / 263	412 / 412	100%	0
GoPro HERO7	170 / 170	266 / 266	100%	0
GoPro HERO8	323 / 323	506 / 506	100%	0
GoPro Karma	200 / 200	313 / 313	100%	0
GoPro Fusion	160 / 160	251 / 251	100%	0
GoPro MAX	175 / 175	274 / 274	100%	0
GoPro HERO6 with BLE	181 / 181	354 / 354	100%	0
GoPro HERO12 Black, 1080p HEVC	217 / 217	169 / 169	100%	0
GoPro HERO12 Black, second take	186 / 186	145 / 145	100%	0
GoPro HERO11 Black, 4K 4:3 HEVC	145 / 145	114 / 114	100%	0
GoPro HERO11 Black, second take	205 / 205	161 / 161	100%	0

An iPhone or an iPad gets there by a different route and keeps its sound. About ten seconds into a take, iOS writes a complete index into the middle of the recording so that a phone dropped mid-shot still has one; a take longer than that is recovered from its own account of itself, with nothing searched for and nothing inferred.

no reference clip	frames	audio	decode errors
iPhone 11, take cut at 50%	397 / 397	560 / 560	0
iPhone 11, take cut at 60%	484 / 484	690 / 690	0
iPhone 11, take cut at 85%	692 / 692	991 / 991	0
iPhone 11, second take, cut at 85%	416 / 416	582 / 582	0
iPad (5th generation), cut at 80%	395 / 395	646 / 646	0

The index stops at the last checkpoint the phone managed to write, but it carries what a reference clip would carry, the codec configuration, the sound description and the way the phone lays its samples down, so the footage after that checkpoint is read with it, exactly as a clip from the same phone would have it read. The take cut at 60% gives 297 frames from the index and 187 more past it, 484 in all, which is what a clip from the same phone recovers through the same bytes. Cinesalve reports how many frames came from each.

Recovered with a clip from the same camera

Recordings that state neither their own configuration nor an index are detected automatically and read it from any working clip you have, which can be a few seconds recorded on the spot. The last column says what each measurement here was made against: a separate take from the same camera, which is the situation you would actually be in, or the recording's own untruncated self where only one camera-original clip of that model is published anywhere to pair with.

real camera	frames recovered	exact	reference
Sony FDR-AX100E, 4K	1488 / 1488	100%	its own full recording
Sony FDR-AX100E, cut at 85%	2292 / 2292	100%	its own full recording
Sony ILCE-6300, 1080p	354 / 354	100%	a different take
Sony ILCE-6300, cut at 85%	508 / 508	100%	a different take
Sony ILCE-6100, 4K	108 / 108	100%	a different take
Sony ILCE-6100, 4K, cut at 85%	154 / 154	100%	a different take
Sony ILCE-7M3, 1080p	322 / 322	100%	a different take
Sony ILCE-7M3, 1080p, cut at 85%	456 / 456	100%	a different take
Sony DSC-RX10M4, 1080p	200 / 200	100%	a different take
Sony DSC-RX10M4, 1080p, cut at 85%	282 / 282	100%	a different take
Sony FDR-AX33, 1080p	226 / 226	100%	a different take
Sony FDR-AX33, 1080p, cut at 85%	317 / 317	100%	a different take
Sony FDR-AX43, 4K	89 / 89	100%	a different take
Sony FDR-AX43, 4K, cut at 85%	126 / 126	100%	a different take
Sony ILCE-6400, 4K	117 / 117	100%	a different take
Sony ILCE-6400, 4K, cut at 85%	168 / 168	100%	a different take
Fujifilm FinePix HS50EXR, 720p	488 / 488	100%	a different take
Fujifilm FinePix HS50EXR, 720p, cut at 85%	710 / 710	100%	a different take
Sony ILME-FX3, 4K	136 / 136	100%	a different take
Sony ILME-FX3, 4K, cut at 50%	197 / 197	100%	a different take
Sony FDR-AX100, 4K 30p	190 / 190	100%	a different take
Sony FDR-AX100, 4K 30p, cut at 85%	259 / 259	100%	a different take
Sony FDR-AX700, 4K 30p	111 / 111	100%	a different take
Sony FDR-AX700, 4K 30p, cut at 85%	159 / 159	100%	a different take
Google Pixel 7 Pro, 1080p	728 / 728	100%	its own full recording
Google Pixel 7 Pro, cut at 85%	1051 / 1051	100%	its own full recording
Google Pixel 9 Pro, 1080p	210 / 210	100%	a different take
Google Pixel 9 Pro XL, 1080p	90 / 90	100%	a different take
Google Pixel 8a, 1080p 60p	143 / 143	100%	a different take
Google Pixel 6 Pro, 1080p	64 / 64	100%	a different take
Google Pixel 5, 1080p	79 / 79	100%	its own full recording
Motorola moto x4, 1080p	336 / 336	100%	a different take
Motorola moto x4, cut at 85%	480 / 480	100%	a different take
LG-M250H, 720p	170 / 170	100%	a different take
LG-M250H, cut at 85%	243 / 243	100%	a different take
Samsung Galaxy S23 Ultra, 1080p HEVC	340 / 340	100%	a different take
Samsung Galaxy S23 Ultra, cut at 85%	483 / 483	100%	a different take
Xiaomi 15T Pro, 4K HEVC	389 / 389	100%	a different take
Xiaomi 15T Pro, cut at 85%	551 / 551	100%	a different take
DJI Mavic 3 Pro, 4K	722 / 722	100%	a different take
DJI Osmo Pocket, 1080p	534 / 534	100%	a different take
DJI Osmo Pocket, cut at 85%	755 / 755	100%	a different take
DJI Air 3S, 4K HEVC	251 / 251	100%	a different take
DJI Air 3S, cut at 85%	361 / 361	100%	a different take
DJI Osmo Pocket 3, 1080p HEVC	367 / 367	100%	a different take
DJI Osmo Pocket 3, cut at 85%	540 / 540	100%	a different take
DJI FC7503, 1080p	252 / 252	100%	a different take
DJI FC7503, cut at 85%	360 / 360	100%	a different take
DJI FC330, 1080p	376 / 376	100%	a different take
DJI FC330, cut at 85%	534 / 534	100%	a different take
iPhone 11, HEVC	484 / 484	100%	a different take
iPhone 12 Pro Max, Dolby Vision HDR	344 / 344	100%	a different take
iPhone 12 Pro Max, cut at 85%	484 / 484	100%	a different take
iPhone 13 Pro Max, 4K60 HEVC	537 / 537	100%	a different take
iPhone 13 Pro Max, cut at 85%	768 / 768	100%	a different take
iPhone 14 Pro Max, 4K60 H.264	156 / 156	100%	a different take
iPhone 14 Pro Max, cut at 85%	206 / 206	100%	a different take
iPad (5th generation), 1080p	114 / 114	100%	a different take
iPad (5th generation), cut at 85%	168 / 168	100%	a different take
Canon EOS 5D Mark II	208 / 208	100%	a different take
Canon PowerShot SX30 IS	391 / 391	100%	a different take
Canon PowerShot SX30 IS, cut at 85%	558 / 558	100%	a different take
Canon PowerShot G7 X Mark II, 1080p25	252 / 252	100%	a different take
Canon PowerShot G7 X Mark II, cut at 85%	360 / 360	100%	a different take
Canon EOS R10, 1080p	525 / 525	100%	a different take
Canon EOS R10, cut at 85%	749 / 749	100%	a different take
Canon EOS M200, 1080p	81 / 81	100%	a different take
Canon EOS M200, cut at 85%	117 / 117	100%	a different take
Canon EOS 90D, 1080p50	78 / 78	100%	a different take
Canon EOS 90D, cut at 85%	114 / 114	100%	a different take
Canon EOS 200D, 1080p25	302 / 302	100%	a different take
Canon EOS 200D, cut at 85%	432 / 432	100%	a different take
Canon EOS Rebel T7i, 1080p	204 / 204	100%	a different take
Canon EOS Rebel T7i, cut at 85%	288 / 288	100%	a different take
Canon EOS 6D Mark II, 1080p	419 / 419	100%	a different take
Canon EOS 6D Mark II, cut at 85%	593 / 593	100%	a different take
Canon EOS 80D, 1080p	215 / 215	100%	a different take
Canon EOS 80D, cut at 85%	306 / 306	100%	a different take
Canon, 720p, mono sound	266 / 266	100%	a different take
Canon, 720p, mono sound, cut at 85%	372 / 372	100%	a different take
SANYO Xacti WH1	837 / 837	100%	a different take
SANYO Xacti WH1, cut at 85%	1195 / 1195	100%	a different take
SANYO Xacti SH1, interlaced	731 / 731	100%	a different take
SANYO Xacti SH1, cut at 85%	1056 / 1056	100%	a different take
SANYO Xacti FH1, 1080p60	1566 / 1566	100%	a different take
SANYO Xacti FH1, cut at 85%	2226 / 2226	100%	a different take
Casio EX-S10	842 / 842	100%	a different take
Casio EX-S10, cut at 85%	1166 / 1166	100%	a different take
Fujifilm FinePix XP70	3494 / 3494	100%	a different take
Fujifilm FinePix XP70, cut at 85%	4951 / 4951	100%	a different take
Fujifilm X-T20, 1080p59.94	381 / 381	100%	a different take
Fujifilm X-T20, cut at 85%	554 / 554	100%	a different take
Fujifilm FinePix S8350, 1080p59.94	558 / 558	100%	a different take
Fujifilm FinePix S8350, cut at 85%	808 / 808	100%	a different take
Panasonic DC-GH5, 4K	111 / 111	100%	a different take
Panasonic Lumix DMC-TZ60	3274 / 3274	100%	a different take
Panasonic Lumix DMC-TZ60, cut at 85%	4572 / 4572	100%	a different take
Panasonic Lumix DMC-SZ1	570 / 570	100%	a different take
Panasonic Lumix DMC-SZ1, cut at 85%	829 / 829	100%	a different take
Nikon COOLPIX P300, 1080p	840 / 840	100%	a different take
Nikon COOLPIX P300, cut at 85%	1187 / 1187	100%	a different take
Nikon D3100, 1080p	282 / 282	100%	a different take
Nikon D3100, cut at 85%	408 / 408	100%	a different take
Nikon 1 J3, interlaced	444 / 444	100%	a different take
Nikon 1 J3, cut at 85%	630 / 630	100%	a different take
Nikon Z6III, 4K HEVC	46 / 46	100%	a different take
Nikon Z6III, 4K HEVC, cut at 85%	62 / 62	100%	a different take
Nikon Z 50, 4K	125 / 125	100%	a different take
Nikon Z 50, 4K, cut at 85%	184 / 184	100%	a different take
Pentax X-5, 1080p	120 / 120	100%	a different take
Pentax X-5, cut at 85%	179 / 179	100%	a different take
Olympus, 1080p	399 / 399	100%	a different take
Olympus, cut at 85%	569 / 569	100%	a different take
Ricoh WG-80, 1080p	817 / 817	100%	a different take
Ricoh WG-80, cut at 85%	1168 / 1168	100%	a different take
Samsung SMX-F70 camcorder, 720x576	14380 / 14380	100%	a different take
Samsung SMX-F70, cut at 85%	20313 / 20313	100%	a different take
YI action camera, 4K	408 / 408	100%	a different take
YI action camera, cut at 85%	578 / 578	100%	a different take
iCatch action camera, 1080p60	417 / 417	100%	a different take
iCatch action camera, cut at 85%	592 / 592	100%	a different take
Novatek dashcam, 1080p60	3125 / 3125	100%	a different take
Novatek dashcam, cut at 85%	4429 / 4429	100%	a different take
BlackVue DR770X dashcam, 1080p60	454 / 454	100%	a different take
BlackVue DR770X dashcam, cut at 85%	648 / 648	100%	a different take
QuickTime, Apple's own muxer	420 / 420	100%	a different take
1080p consumer camera	1356 / 1356	100%	a different take
1080p camera that names no maker, mono sound	227 / 227	100%	a different take
1080p camera that names no maker, cut at 85%	325 / 325	100%	a different take
GoPro HERO6, cut at 85%	603 / 603	100%	a different take

Synthetic suite

Cut at known points so the original is available to compare against, frame by frame.

test case	video frames exact	audio frames exact
H.264 720p 30fps, cut at 68%	203 / 203 (100%)	100%
H.264 720p 30fps, cut at 90%	269 / 269 (100%)	100%
H.264 1080p 24fps, cut at 55%	107 / 107 (100%)	100%
H.264 480p 60fps, cut at 45%	217 / 217 (100%)	100%
H.264 without audio, cut at 60%	144 / 144 (100%)	—
HEVC 720p 30fps, cut at 60%	146 / 146 (100%)	100%

What it recovers

Picture, exactly, on H.264 and HEVC in MP4 and MOV — what consumer and prosumer cameras record.

Sound comes back on every camera in the suite that recorded any. On a GoPro, a Canon EOS, a Canon PowerShot, a Casio EX-S10, a Fujifilm FinePix, a Fujifilm X-T20, a Nikon COOLPIX, a Nikon D3100, a Nikon 1 J3, a Nikon Z6III, a Panasonic DC-GH5, a Panasonic Lumix TZ60, a Panasonic Lumix SZ1, a Pentax X-5, a Ricoh WG-80, a Samsung SMX-F70, an Olympus, a YI action camera, an iCatch action camera, a SANYO Xacti, a Sony camcorder, a Sony a6300, a Sony a6100, a Sony a7 III, a Sony FX3, a Sony FDR-AX700, a Sony RX10 IV, a Sony FDR-AX33, a Sony FDR-AX43, a Sony a6400, a consumer camcorder, a recording written by Apple's own muxer, an iPhone, a Google Pixel, a Motorola moto x4, an LG phone, a Samsung Galaxy S23 Ultra, a Xiaomi 15T Pro, a DJI Osmo Pocket, a DJI Osmo Pocket 3, a dashcam on a Novatek chip and a BlackVue dashcam it comes back whole, every frame at the position the camera wrote it: a GoPro and a BlackVue name the size of every sample in the bytes ahead of it, the Canon MOV cameras, the Fujifilms, the iCatch camera, the Olympus, the GH5, the Pentax, the Ricoh, the Nikon D3100, the Nikon Z6III and the Sony write uncompressed sound that is placed from the pictures by arithmetic rather than searched for, and the Canon MP4 cameras', the camcorders', the Apple-muxed recording's, both Lumix compacts', the COOLPIX's, the Nikon 1's, the YI's, the Xacti's, the Casio's, the iPhone's, the moto x4's, the LG's, the Galaxy's, the Xiaomi's, both Osmo Pockets', the Novatek dashcam's and the Pixel's compressed sound is read frame by frame, with the system's own decoder saying where each frame ends. The metadata an iPhone writes behind each block of sound, and the recovery index it keeps inside its own recording, each name their own length, so they are stepped over to the next picture rather than read as sound. On a camera it has not measured, where bytes between pictures cannot be attributed to any track, Cinesalve recovers the picture and says it left the sound out rather than writing bytes it cannot verify into your recording.

If it does not recover your footage

You get your money back. Email cinesalve@proton.me with the licence key and a sentence about what happened, and the refund goes through the same day.

That is an easy promise because you know the answer before you buy: the free checker reads your own file and reports how many frames are recoverable. Nobody should pay to find out whether their footage survived.

Installing

Download the zip, unzip, drag Cinesalve.app to Applications. macOS 13 or later, universal binary, Apple silicon and Intel.

First launch needs one approval, once:

1. Double-click Cinesalve.
2. Open System Settings → Privacy & Security, scroll to the bottom, click Open Anyway next to Cinesalve.
3. Confirm.

Control-clicking and choosing Open does not work — Apple removed that shortcut in macOS 15. The terminal equivalent is one line:

xattr -dr com.apple.quarantine /Applications/Cinesalve.app

Or install with Homebrew, which handles upgrades:

brew install --cask lucidelarp/cinesalve/cinesalve
xattr -dr com.apple.quarantine /Applications/Cinesalve.app

Getting help

Open an issue here, or email cinesalve@proton.me. Say which camera and recording mode produced the file, and include the app's message if it showed one.