Convert M4V to PNG

PNG has been lossless since 1996: it stores the frame exactly as the decoder produced it, adds nothing of its own, and survives being opened and re-saved indefinitely. On a photograph that fidelity is a luxury you pay for in file size. On a slide, a diagram or a screen full of interface it is the difference between legible and irritating.

JPEG’s compression works by discarding what the eye rarely misses, and hard edges are exactly where the eye does miss it — the result is a faint halo around lettering that gets worse each time the file is saved. For a screenshot going into a manual, a bug report or a set of documentation that will be edited later, that is not a trade worth making.

Video is stored with colour at a fraction of the resolution of brightness, because eyes are far more sensitive to one than the other. That works beautifully on faces and landscapes and is brutal on coloured text: red or blue lettering on a slide loses its edges before the file ever reaches you, and no still format puts them back.

The second cause is bit allocation. An encoder spends fewer bits on frames where things are moving, so a slide captured while a presenter was mid-transition is materially softer than the same slide two seconds later once everything settled. PNG preserves whichever of those you hand it, faithfully — which is the point and also the limit.

If the frame is a slide, take it a second or two after the transition finishes rather than during it. If it is an interface, take it while nothing is animating — no menu sliding, no cursor moving, no spinner turning. Encoders reward stillness with detail and the difference is often dramatic.

Since a conversion is cheap, the practical method is to try three timestamps a fraction of a second apart and keep the best. That takes under a minute and reliably produces a sharper image than any amount of deliberation about which format to use.

Video bought or rented from the iTunes Store carries FairPlay encryption. A browser has no key for it, so nothing can be decoded and the conversion stops with a message saying nothing in the file could be converted. That is a lock rather than a fault.

Screen recordings, QuickTime and iMovie exports, camera footage and course-platform downloads are all unprotected and behave normally. Since this page’s material is overwhelmingly recordings people made themselves, the question rarely bites here — but it explains the one file on the shelf that refuses.

No downscaling happens: a screen captured at 2560 by 1440 gives a PNG at 2560 by 1440, whatever size the player window happened to be when you found the moment. That is the fundamental advantage over pausing and taking a screenshot, which captures the player at its on-screen size, complete with a control bar if it has not faded.

It matters most for exactly this material. A slide screenshotted from a half-size player is a slide nobody can read at full size in a document; the same slide extracted from the file is sharp enough to crop into.

The timestamp control takes seconds and returns the frame being shown at that moment — the last frame whose own start time falls at or before the number. Each run produces exactly one image; there is no sequence export and no interval mode.

For a recorded talk that means one conversion per slide, which sounds tedious and in practice is not: type the second, download, type the next. Keeping a list of the timestamps as you go is worth more than the images, because a recording you come back to a month later is one where finding the moments again is the slow part.

PNG carries transparency and the output has an alpha channel, but a video frame is a solid rectangle of pixels — there is nothing in the source that could be transparent. Every pixel comes out fully opaque.

If the goal is a screenshot with the background removed, or a logo lifted off a slide, that is editing work in an image editor rather than something a conversion supplies. Extract the frame losslessly first, which is what this page does, and cut it out afterwards from an image that has not already been degraded by a lossy save.

Larger than the JPG of the same frame, and how much larger depends entirely on the content. PNG compresses runs of identical pixels extremely well, so a slide with a plain background and some text can be remarkably small, while a frame of a busy camera shot can be several times the size of a JPEG of it.

That split matches the advice. Screen and slide content is both the case where PNG is technically right and the case where it costs least; photographic content is where it is expensive and where a JPG would have been fine. Let the contents of the frame decide rather than a rule.

The frame is decoded and written inside the browser tab on your own machine, so the M4V is never sent anywhere. The network tab during a conversion is the proof, and the free tier takes files up to 100 MB.

The material this page attracts is frequently internal: a recorded planning meeting, a demo of an unreleased build, a support session containing somebody’s account details on screen. Extracting a frame from those on a stranger’s server is a decision somebody would eventually have to explain, and not making it is simpler.