Gaming hardware is sold on numbers — frame rates, refresh rates, response times, latency figures. These describe different properties and are frequently conflated.
Understanding what each measures explains which upgrades produce a perceptible difference.
The distinct measures
Frame rate. How many images per second the system produces. Determined by the software and the processing hardware.
Refresh rate. How many times per second the display updates. A property of the display.
These are independent. A high frame rate on a low refresh display means frames are generated and not shown.
Response time. How quickly a display's pixels change state. Affects motion clarity — slow transitions produce smearing.
Input latency. Total delay between a physical input and the corresponding change on screen. The sum of everything in the chain, and the measure that actually determines responsiveness.
Where the perceptible differences are
Human perception of these varies by what's being measured and by context.
Frame rate. The improvement from very low to moderate frame rates is dramatic and obvious to everyone. From moderate to high, it remains noticeable, particularly in fast motion. Beyond that, returns diminish sharply for most people and most content.
Research on visual perception indicates people can detect differences at rates well above what was once assumed, particularly for motion and for latency rather than for smoothness as such.
Consistency matters more than the average. A steady moderate frame rate is perceived as smoother than a variable higher one. Frame time variance — stutter — is more objectionable than a lower stable rate.
Which is why average frame rate is a poor metric, and why measures of the worst frames are more informative about actual experience.
Latency. Perceptible at surprisingly low levels, particularly for direct manipulation. Studies of touch interfaces have found people detecting latency well below what was assumed acceptable.
This is where genuinely competitive players notice differences, and it's the measure least well captured by frame rate alone.
Variable refresh rate
The technology that resolved a genuine problem and is worth understanding.
Traditionally, displays refreshed at a fixed rate regardless of when frames arrived. A frame arriving mid-refresh produced tearing — parts of two frames visible simultaneously. Synchronising to avoid this introduced latency and could halve effective frame rate when the system couldn't keep up.
Variable refresh rate technologies let the display refresh when a frame is ready, eliminating tearing without the latency penalty.
The perceptual improvement is substantial, particularly where frame rate fluctuates, and it's arguably a bigger upgrade than a moderate increase in raw frame rate.
Where the latency actually is
The chain has several components and people frequently optimise the wrong one.
Input device polling. Modern peripherals contribute a small amount.
Processing and rendering. Frame time plus any buffering. Deeper buffering improves smoothness and adds latency — a genuine trade-off.
Display processing. Some displays apply substantial processing, adding tens of milliseconds. This is why a game mode that disables processing matters more than most specifications.
Pixel response. Modest on modern panels.
For anyone concerned about responsiveness, display processing is frequently the largest and most easily addressed component. A television in its default picture mode can add more latency than everything else combined.
Panel technologies
Different characteristics with genuine trade-offs.
Panels differ in response time, contrast, viewing angles and colour, and the technology that leads on one measure frequently trails on another.
Emissive displays, where each pixel produces its own light, have near-instant response and excellent contrast, with considerations around brightness and static image retention.
The practical point: no technology is best at everything, and the right choice depends on whether you prioritise motion clarity, image quality, brightness or cost.
Upscaling and frame generation
Techniques that complicate the picture considerably.
Upscaling renders at lower resolution and reconstructs to a higher one, improving frame rate at some cost to image quality. Modern implementations are good enough that the trade is frequently worthwhile.
Frame generation synthesises intermediate frames. This increases the displayed frame rate without reducing latency, because generated frames are interpolated from existing ones rather than reflecting newer input.
Which produces an important distinction: a high frame rate achieved through generation looks smoother without feeling more responsive. For visual smoothness it works; for competitive responsiveness it doesn't.
Marketing frequently reports these figures without the distinction.
Practical priorities
If responsiveness matters: reduce display processing, use a display mode that minimises latency, and prioritise consistent frame delivery over peak figures.
If visual quality matters: variable refresh rate, adequate contrast, and enough frame rate for stability rather than maximum.
And be sceptical of any single number. The experience is determined by the whole chain, and marketing describes individual components.
Measuring your own latency
Worth doing rather than assuming, since the largest contributor is frequently not where people expect.
Tools exist that measure end-to-end latency from input to display, and some graphics drivers and displays report figures directly.
The common discovery: display processing accounts for far more than expected, particularly on televisions in default picture modes, where processing intended to improve motion or upscaling can add substantial delay.
Switching to a mode that bypasses processing frequently produces a larger improvement than any hardware upgrade, and it costs nothing.