Why FPS Gains Get Smaller as FPS Gets Higher

Frame rate is the misleading unit

Frame rate is frames per second. Frame time is milliseconds per frame. They're reciprocals — 1000 divided by one gives the other.

That reciprocal relationship is the whole story, because it means equal-sounding FPS gains are worth wildly different amounts:

Upgrade	FPS gained	Frame time saved
30 → 60	+30	16.7 ms
120 → 150	+30	1.7 ms

Same number on the counter. Ten times apart in what you actually perceive — and nothing about the FPS figures tells you that.

Diminishing returns are arithmetic, not opinion

Every doubling of frame rate halves frame time, so each doubling delivers half the absolute saving of the one before:

• 30 → 60 saves 16.7 ms
• 60 → 120 saves 8.3 ms
• 120 → 240 saves 4.2 ms

Whether 4.2 ms is worth money to you is a real judgement — competitive shooters are a different case from single-player games. But the shrinking isn't a matter of opinion.

It's why 30 to 60 changes an experience while 200 to 240 mostly changes a number.

The average hides what you feel

How long a bad frame hangs for is often the more useful output.

A game averaging 60 fps with 1% lows at 45 delivers occasional frames taking 22.2 ms against a normal 16.7 — a 5.6 ms hitch. Put the lows at 30 instead and those frames take 33.3 ms, which is a full extra frame of waiting and reads as a stutter rather than a slightly slower moment.

Inconsistency is more noticeable than a lower steady rate. That's why a locked 60 often feels better than an unlocked 80 that dips, and why chasing a higher average while the lows stay bad is an unsatisfying way to spend money.

If that output shows a dash, your 1% low is at or above your average — which can't happen. Check which figure went in which box.

Your monitor sets a deadline

Your monitor's frame budget is how long the panel gives each frame. A 144 Hz display refreshes every 6.94 ms, so a frame taking longer misses its slot.

Comparing your frame time against that budget tells you immediately whether you're filling the panel — regardless of what the average suggests.

Frames per screen refresh shows where you sit. Below 1 and you're not filling it. Above 1 and the extra frames aren't shown as motion, though they do slightly reduce input latency because the displayed frame is more recent. That's why competitive players deliberately run above their refresh rate.

For single-player games the usual better answer is to cap slightly below refresh and spend the headroom on settings, or on consistency.

What this doesn't do

It converts between units and compares them. That's arithmetic.

What frame rate you get from a given combination of settings, resolution and hardware is a measurement, and only your machine running your game can produce it.

The workflow that works: measure first with an overlay reporting averages and 1% lows, then bring those numbers here to find out what a change is actually worth — before spending time or money on it.

If the average looks fine and it still feels bad

Almost always the lows.

A high average with poor 1% lows means frames arrive unevenly, and the eye notices change more than level. Put your lows in and look at the hitch in milliseconds: anything approaching the length of a normal frame is a visible stutter.

Fixing consistency — capping the frame rate, or reducing whichever setting spikes — usually improves things more than adding average frames.