Tutorial

How to choose the video quality you download

A single video usually exists on a platform in half a dozen versions, from a preview-sized file to the sharpest rendition that site will hand out. The list you get after parsing is that menu, and choosing from it is a trade between picture and data. Here is how to read it.

  • No account needed
  • Phone and desktop
  • Public links only

SniffVid TeamPublished Last reviewed

Bot protection

Paste a link above to see every available video, audio and cover download.

By using SniffVid you agree to our Terms of Service and Privacy Policy

The short answer

  • A resolution label describes the frame size, not the sharpness: bitrate decides how much detail survives movement.
  • The file size shown beside each entry is the most useful single clue about bitrate, and it is what to compare first.
  • Platforms gate their best renditions behind a sign-in or a subscription, and a parser reads the page as an anonymous visitor.
  • The jump in cost between 1080p and 720p is large, and the difference is hard to see on a phone screen.
  • Nothing is upscaled and nothing is compressed on the way through, so the tier you pick is the tier you keep.

Written from what the parsers SniffVid runs actually return, and checked on a desktop browser and a phone. The limits that apply to every page are stated on each one rather than left in the FAQ.

Step by step

The same three moves every time, whatever the platform does with its streams.

  1. 1

    Parse, then read the video column

    Each entry carries a resolution label, a codec and often a size. Those three numbers are the whole decision, and they come from the source rather than from a guess.

  2. 2

    Compare size against resolution

    Two entries at the same label can differ in size, which usually means a different codec or bitrate. The larger of the two carries more detail in fast movement.

  3. 3

    Pick the smallest one that still looks right

    For a clip you will watch on a phone, the middle of the list is usually the sweet spot. Take the top of the list when you are archiving something or feeding an editor.

Four bars of decreasing length labelled 1080p, 720p, 480p and M4A, showing relative file size.
Relative size, not relative quality. 1080p carries roughly twice the data of 720p and about five times that of 480p, and the audio track alone costs a fraction of any of them.

SniffVid lists what the parsers can actually reach. If a quality is missing from the list, it is missing at the source: the tool never invents a stream and never upscales a file to a label it has not earned.

What the labels actually describe

A resolution label describes the number of pixels in the frame, not how good the frame looks. How much detail survives movement is decided by the bitrate, which is why a smaller-looking label can sometimes beat a larger one.

  • 1080p means 1920×1080; 720p means 1280×720; 480p means 854×480
  • The label says nothing about smoothness, sharpness or compression artefacts
  • The file size shown beside an entry is the most useful single hint about bitrate

Why the highest quality can be missing

Platforms gate their best renditions. Some formats are reserved for signed-in accounts, some are tied to a subscription, and some are simply not exposed to an anonymous visitor at all. SniffVid parses as an anonymous reader and never asks for credentials.

  • Sign-in-only qualities are invisible to any tool that does not hold your account
  • Region-locked renditions may be absent while lower ones resolve normally
  • Parser coverage changes, so a quality missing today may appear later

What each tier costs you

The gap between tiers is larger than most people expect, which is why the newest option is rarely the right default. The figures below are typical ranges for a ten-minute video rather than guarantees, because platforms choose their own bitrates.

  • 1080p: roughly 100–300 MB for ten minutes of ordinary footage
  • 720p: roughly half of that, and hard to tell apart on a phone screen
  • 480p: a fraction of the size, fine for slides, talking heads and note-taking
  • Audio only: typically a few megabytes regardless of the length of the video

Choosing for the job at hand

There is no single best answer, because the best file depends on where it will be watched and whether it needs to survive as an archive.

  • Something to watch on a phone: the middle of the list, or audio only if the picture is incidental
  • Reference material for an edit: the highest resolution offered, in the container the editor prefers
  • Archiving your own upload: the highest quality the platform exposes, kept somewhere you control
  • Saving data while travelling: a lower tier downloaded in advance beats a higher one that fails on hotel Wi-Fi

Reference

These tables are shared across the site, so a figure you saw on one page is the figure you see here.

Typical size of a ten-minute download by quality
QualityTypical size (10 min)Best for
1080p100–300 MBArchiving, editing, or watching on a large screen
720p50–150 MBPhones and laptops, and anything that has to download quickly
480p20–60 MBSlides, talking heads and note-taking, where fine detail hardly matters
M4A5–15 MBListening, transcription, or saving data on a long video

Background and edge cases

Why the same label can look different twice

Two entries at 1080p are not the same file. They may use different encoders, and often do: a platform can publish an H.264 rendition for compatibility and a more efficient codec beside it. The more efficient one packs more detail into fewer bytes, which is why a smaller 1080p file is sometimes the better of the two.

  • A label shows frame size, while the codec decides how efficiently it is stored
  • Compare like with like: same codec, then same resolution, then size
  • The largest entry is not automatically the best-looking one

What the numbers do not tell you

Size and resolution are visible before you download; the encoder's settings are not. Bitrate allocation, keyframe spacing and whether the source is a re-upload are all invisible, and they matter more than the label in fast-moving footage.

  • A re-uploaded video has already been through one lossy pass, whatever the label says
  • Static scenes hide compression damage; movement and gradations reveal it
  • Downloading the next tier up is the only way to compare reliably, and the sizes tell you what that costs

Choosing when the connection is the constraint

On a metered connection the choice reverses: the highest tier you can afford to download now beats the tier you would prefer but cannot finish. Downloading in advance is the whole reason to keep a copy, and a partial file is worth nothing.

  • 720p downloads in roughly half the data of 1080p, which matters on mobile plans
  • Audio-only is the right tier for anything you will not watch
  • A failed download costs the full size of the tier you attempted, not the part that arrived

Frequently asked questions

How this was checked

How this was checked

The size ranges come from comparing the track sizes the parsers report for ordinary footage across several platforms and resolutions. They are ranges because every platform chooses its own bitrate — no single figure here is a promise about your video.

  • The steps were run in a current Chrome, Edge and Firefox on desktop, and in Safari on iOS, against the live parsers.
  • Sizes in the tables are typical ranges for the resolutions named rather than measurements of one particular video, because each platform chooses its own bitrate.
  • Every limit stated here — merging on a phone, converting between formats, private videos — was confirmed as a limit rather than a fault before it was written down.

Who maintains it

Written and reviewed by SniffVid Team, the people who build and run SniffVid. There is no freelance pool and nothing is sponsored: a page exists because a question keeps coming up, and it is rewritten when the answer changes. Read the editorial policy

Published · Last reviewed · Content last reviewed

Sources and further reading

Primary documentation for the claims on this page. Nothing here cites another downloader as an authority on how downloads behave.

Keep reading

Other tutorials, and the tool pages for the sites people download from most.

Tool pages for the biggest platforms

Four sites have a page of their own with the downloader built in. For everything else, the tutorial above is the way in.