← Back to blog

Does 120Hz improve reaction time?

You miss a mini-game by a hair, and the first thought is never "I was too slow". It is "the screen lagged". Comfortable. Occasionally true. A phone running at 60Hz paints a new frame every 16.67 milliseconds. At 120Hz, every 8.33 ms. Somebody decided that heartbeat for you.

Which leaves the only question worth the read: do those milliseconds mean anything against a human reaction time sitting around 273 ms? Short answer, yes, but far less than gaming phone marketing suggests, and far less than the night you cut short. Long answer with numbers, below.

How many milliseconds does a 60Hz screen actually cost you?

The math is not mystical. A 60Hz display pushes 60 frames per second, one frame every 16.67 ms. A 120Hz display halves that to 8.33 ms. When the game decides to show a signal, that signal waits for the next frame to exist on screen. Best case, it waits nothing. Worst case, a full interval. On average, half of one: roughly 8 ms at 60Hz, 4 ms at 120Hz.

So the average gain attributable to refresh rate alone is about 4 ms. Not 16. Not "twice as fast", whatever the spec sheet implies. Four milliseconds, inside a budget that runs close to three hundred.

The panel is not the whole circuit though. Between your finger and the pixel that changes, there is touch detection, game logic, rendering, buffering, and finally the pixels lighting up. DXOMARK measures that full chain with a robot tapping the glass and a high-speed camera watching, and the spread is ugly: in their gaming use case, one device responds in 3 ms while another takes 10 ms. Three times slower, on hardware sold to people who all believe their phone is fast.

Refresh rate or touch sampling rate: which one matters?

This is the most common confusion on gaming phones, and it is maintained on purpose. Two frequencies live side by side, and they do different jobs.

• Refresh rate (60, 120, 144Hz) is output: how fast the screen shows you something new.
• Touch sampling rate (120, 240, 720Hz) is input: how fast the digitizer scans where your finger is. At 240Hz it re-reads your finger every 4.16 ms.

On a touch game the second one matters at least as much as the first, and it is the one nobody checks before buying. Phones built for gaming push it hard: the ROG Phone 7 Ultimate reaches 720Hz sampling, the RedMagic 7S Pro goes to 960Hz. Ordinary flagships sit around 240Hz, which is already plenty for a reflex game where you tap once, decisively, at the right instant.

The honest caveat: a 4 ms difference on an isolated tap is well below human perception. You do not feel one tap arriving 4 ms sooner. What you do feel, on continuous gestures like a drag or a hold, is tracking smoothness, because the engine receives more sample points along your thumb's path. For a mini-game that asks you to hold steady, that shows. For a mini-game that asks you to tap on cue, much less.

Your screen costs you a few milliseconds. Your five-hour night costs you fifty.

Does high refresh rate actually raise your score?

Serious measurements exist, and they refuse to please either camp. A study on the relationship between display refresh rate and eSports reaction time compared 60, 120 and 240Hz and found a measurable gap, including between 120 and 240Hz, favouring the higher rate in competitive play.

Measurable is not dramatic. The gap lives in single-digit milliseconds, while your own brain varies by tens of milliseconds between two consecutive attempts. Human Benchmark says as much on its own statistics page: the test is affected by the latency of your computer and monitor, and a low-latency display improves the score. Their 273 ms median, drawn from more than 81 million clicks, therefore already bakes in the average latency of the world's hardware. That is the number you are chasing, not a lab ideal.

Practical takeaway: if you play reflex games on a phone locked at 60Hz that can do 120, open display settings and unlock it. That is free, and it removes a few milliseconds of noise. If you are considering a 900 dollar phone to buy 4 ms, keep the money and go to bed earlier.

What costs you far more than your display

Before blaming the panel, look at what genuinely slows you down. The orders of magnitude are not up for debate: hardware plays with single milliseconds, your state plays with tens of them.

• Sleep. One short night is enough to add several tens of milliseconds to your reaction time. We broke the mechanism down in the real cost of sleep deprivation on reaction time. No screen buys that back.
• Guessing. Anticipation is a bet. When you leave early you stop measuring your reflex and start measuring your luck. A good reflex game punishes that, and it should.
• Finger position. A thumb that lifts three centimetres between taps adds its own travel time. Stay in contact, or just above the glass.
• Series fatigue. Your first twenty attempts and your last twenty tell different stories. Focus decays, and it decays faster than your hardware ages.

If you want the real lever instead of the shop window, it sits elsewhere: in training reaction speed consistently, five minutes a day on unpredictable signals. Gains there are measured in tens of milliseconds, not in fours.

How to check whether your phone is slowing you down

A simple protocol, doable tonight, worth more than any spec sheet.

• Run ten measurements on a reaction test, keep the median, throw away the first attempt, which is always bad.
• Change exactly one variable: switch the display from 60 to 120Hz in settings, touch nothing else.
• Run ten more measurements under the same conditions, same hour, same level of tiredness.
• Compare medians. A 3 to 6 ms gap is consistent with physics. A 40 ms gap did not come from the screen, it came from you.

This protocol earns its keep beyond the hardware question: it teaches you to measure instead of to feel. Our piece on how to test your reflexes properly covers the four ways to do it without fooling yourself. The rule fits on one line: a single measurement proves nothing, a median over ten attempts starts to mean something.

Hardware as an excuse, or as a detail

120Hz is a genuine improvement. It makes motion cleaner, it removes a little noise between your gesture and its consequence, and across dozens of prompts in a row that missing noise is felt. It is simply not where your score is decided. Four milliseconds gained on the panel against fifty lost to a bad night: the arithmetic settles itself.

That is also why a well built reflex game refuses to cheat on legibility: the cue has to be blunt, the timing has to be clean, and failure has to come from your slowness rather than a muddy animation. In NERVE, 20 mini-games hit you without warning, each success makes the next one harder, and each mistake eats one of five shared lives. The Daily mode hands everyone the same challenge on the same day, which gives an honest comparison between players, screens included. The global leaderboard settles the rest, and the score card is there to be shared, with or without a hardware alibi.

One question stays open, and it beats the 60 versus 120 argument: at what level of practice does hardware genuinely start to matter? As long as your median swings 30 ms from one week to the next, the phone is not the subject. The day it settles within 5 ms, come back and negotiate with your screen.