Resolution is the number that matters least

Bitrate is the budget

Resolution says how many pixels there are. Bitrate says how much information is available to describe them. A 4K stream given twelve megabits has four times the pixels of a 1080p stream and barely more data to describe them with, so each pixel gets less.

That is why a well-encoded Full HD channel routinely beats a starved 4K one, and why "4K" in a channel name guarantees nothing. The honest figures are roughly six to ten megabits for 1080p in H.264, four to six in H.265, and fifteen to twenty-five for genuine 4K in H.265.

The codec changes the arithmetic

H.265 reaches about the same quality as H.264 at roughly half the bitrate. That saving is what made 4K delivery practical at all.

The cost is decoding effort. HEVC is much heavier, and a device without hardware support falls back to its processor — dropping frames, softening during motion, growing warm and getting progressively worse as it throttles. It is not a network problem and restarting the router will never touch it.

This is the most common reason two households on an identical subscription reach opposite conclusions about it. One has hardware decoding; the other does not.

Motion is where encoders are exposed

Compression works by describing change. A studio shot barely changes between frames and compresses beautifully. A camera panning across a stadium changes everything, and the encoder needs far more bits to keep up.

So the same channel can look immaculate during the news and fall apart during sport at the same nominal bitrate. If you want to judge encoding honestly, watch fast motion rather than a talking head.

How to test fairly

At peak time, because that is when infrastructure is genuinely loaded and it is the only test that means anything. On the device you will really use, wired if it can be. On the channels you actually watch. And for ten minutes rather than ten seconds, because a stutter every four minutes will not show up in a quick look and will ruin an evening.