Reviewing a streaming player is not a processor benchmark. The measurements that separate one box from another concern format support, display negotiation and the time from input to picture.
Codec and format support sets the ceiling
A player can only show what it can decode. Support for a given codec must exist in dedicated hardware, because general-purpose decoding of high-resolution video is not practical on these devices.
That makes format support a fixed property of the model rather than something a software update adds. A box lacking hardware support for a newer codec will not gain it later.
Reviewers therefore test playback of each relevant format directly rather than trusting a specification sheet, since apps sometimes fall back to a lower rung without saying so.
The display handshake decides the picture
Every time a player sends video to a television, the two negotiate resolution, frame rate, color range and high dynamic range signaling.
Failures here produce most of the complaints attributed to picture quality: washed-out color from a mismatched range, or a television stuck in the wrong mode for the content playing.
A useful review checks whether the box switches output frame rate to match the content, since a mismatch causes the periodic stutter viewers describe as judder.
Time to picture is the honest speed test
The meaningful speed measurement is the interval between selecting a title and seeing stable video, including app launch, authentication and buffering.
This depends on storage speed, memory pressure and how the app was built, so it varies dramatically between apps on the same hardware and must be measured per app.
It also degrades over time as apps grow and storage fills, which is why a box reviewed at launch may behave differently a year later.
Audio passthrough is easy to get wrong
Many households route audio to a receiver or soundbar, which requires the player to pass an encoded audio stream through untouched rather than decode it.
Boxes differ in which formats they pass, whether they do so reliably after an app switch, and whether they resynchronize picture and sound correctly afterwards.
Lip-sync drift after switching apps is a common and reproducible defect, and it is the kind of thing only found by testing rather than by reading a feature list.
What a review cannot establish
App availability and behavior are controlled by the app publishers and change constantly, so a review captures a moment rather than a permanent state.
Long-term support is similarly unknowable at review time, and it matters more than anything measurable, since a box without updates loses formats and apps as the market moves.
Honest reviews say so, and separate the hardware's capabilities, which are fixed, from the software situation, which is not.