Bitrate vs resolution: what actually drives video quality
7 min read
Resolution gets all the marketing attention, which is unfortunate, because it is the less important of the two numbers that determine how a video actually looks. A 1080p file can comfortably out-perform a 4K one, and understanding why makes it much easier to choose sensibly between quality options.
The two numbers
Resolution is how many pixels make up each frame — 1920 × 1080 for 1080p, 3840 × 2160 for UHD. It is a measure of potential detail.
Bitrate is how much data is used to describe those pixels each second, measured in megabits per second. It is a measure of how much information survived compression.
Resolution sets the size of the canvas; bitrate determines how much paint you have. A large canvas with too little paint produces a worse picture than a smaller canvas painted properly, and that is the entire relationship in one sentence.
Why compression forces the trade-off
Raw uncompressed video is enormous — 1080p at 30fps is roughly 1.5 gigabits per second, which is impossible to stream and impractical to store. Every video you watch has therefore been compressed, usually by a factor of several hundred.
Codecs achieve this mostly by describing what changed between frames rather than storing each one whole, and by discarding detail the eye is least likely to notice. Bitrate is the budget for that process. A generous budget lets the encoder keep subtle detail; a tight one forces it to throw away more, and the losses become visible.
The characteristic symptoms of an insufficient bitrate are worth recognising: blocky patches in dark or smoothly graded areas, detail turning to mush during fast motion, banding across skies and gradients instead of smooth transitions, and a general softness as though the image is slightly out of focus. If a video looks wrong in these specific ways, resolution is not the problem.
The interaction
Here is the crux: a higher resolution needs a proportionally higher bitrate to look equally good. Four times the pixels need roughly four times the data.
So take a fixed budget of, say, 8 Mbps. Spend it on 1080p and each pixel is well described — the result looks clean. Spend the same 8 Mbps on 4K and the same data is spread across four times as many pixels, each getting a quarter as much. The 4K file has more pixels and less information per pixel, and it will usually look worse than the 1080p version at the same bitrate.
This is why "4K" alone is not a quality claim. A 4K file at a low bitrate is a large file containing a soft, artefact-ridden image. A 1080p file at a healthy bitrate is a smaller file containing a clean one.
Rough bitrate guidance
For H.264, the most common codec, sensible working figures are around 5 Mbps for 720p, 8–12 Mbps for 1080p at 30fps, 12–20 Mbps for 1080p at 60fps, and 35–50 Mbps for 4K at 30fps. Higher frame rates need roughly proportionally more.
Newer codecs change the arithmetic. H.265 and VP9 achieve comparable quality at something like half the bitrate of H.264; AV1 does slightly better again. A 6 Mbps H.265 stream and a 12 Mbps H.264 stream are broadly equivalent, which is why bitrate figures are only meaningful alongside the codec.
Content matters too, and more than the numbers suggest. A talking head against a plain background compresses beautifully — very little changes between frames. Falling snow, rippling water, confetti, or a handheld shot through foliage defeats compression almost completely, because nearly every pixel changes every frame. The same bitrate that looks flawless on an interview can fall apart on a snowy landscape.
Constant, variable and constrained
Encoders spend their budget in different ways. Constant bitrate uses the same amount of data throughout, which is predictable but wasteful — simple scenes get more than they need while complex ones get less.
Variable bitrate allocates according to difficulty, spending more on complex sequences and less on simple ones. It produces better quality per megabyte and is the sensible default for most purposes.
Constrained variable bitrate varies within a ceiling, which is what streaming services use — quality adapts, but never exceeds what the connection can carry.
There is also CRF, or constant rate factor, which targets a consistent visual quality and lets the bitrate go wherever it needs to. For archiving, this is usually the right approach: you specify how good it should look and accept whatever file size results, rather than fixing the size and accepting whatever quality falls out.
Choosing in practice
When a tool offers you several qualities, the useful question is not "what is the highest number" but "what is the highest number this source actually supports". Downloading a 4K option for a video that was uploaded at 1080p and upscaled gains you nothing but file size — the detail was never captured.
Match the resolution to the screen: 1080p for phones and tablets, where 4K is imperceptible; 4K only for large displays with genuinely 4K sources. Then, if a choice remains, prefer the higher bitrate at the resolution you picked over a higher resolution at a starved bitrate.
A quick sanity check: divide the file size in megabytes by the duration in seconds, multiply by 8, and you have the approximate bitrate in Mbps. If a supposedly 4K file works out at 5 Mbps, it is 4K in dimensions only — and a 1080p version of the same material would very likely look better.