Methodology

This page explains exactly how every TechCalcu calculator and test works: the formulas, the data behind them, and what they cannot tell you. Our tools give estimates to help you decide; they are not a replacement for testing your own PC.

General principles

  • Published inputs. Power figures come from manufacturer specifications (Intel PL2, AMD PPT, NVIDIA and AMD board power). Performance scores are approximate tiers we set from published reviews and benchmark aggregates.
  • Simple, visible formulas. Every calculation runs in your browser and is described below, so you can check it by hand.
  • Honest limits. Each tool lists what it does not account for. Browser-based tests (mouse, keyboard) are estimates because browsers limit timing precision.
  • Privacy. Tool inputs stay in your browser; we do not store them.
  • Who maintains them. The tools are built and maintained by Adnan, founder and editor of TechCalcu.

Tools on this page

PC Bottleneck Calculator

Each CPU and GPU in the list has a relative gaming performance score on a 0-100 scale. We set these scores as approximate performance tiers from published reviews and benchmark aggregates, then adjust them for your resolution and RAM:

  • 1080p: CPU score × 0.9 and GPU score × 1.1, because low resolutions lean on the CPU.
  • 1440p: no adjustment.
  • 4K: CPU score × 1.1 and GPU score × 0.9, because high resolutions lean on the GPU.
  • 16GB RAM or less: CPU score × 0.95, because many new games recommend 32GB.

Bottleneck % = (higher score − lower score) ÷ higher score × 100

Under 10% is shown as well balanced, 10-20% as a slight bottleneck and above 20% as a significant one. The lower-scoring part is named as the bottleneck.

Limits:

  • Scores are approximate tiers, not benchmark results for any single game.
  • Real bottlenecks change with the game, settings, FPS target and even the scene.
  • It does not see RAM speed, cooling, background apps or a weak power supply.
  • Confirm a result by watching GPU usage in your own games, as explained in our CPU or GPU bottleneck guide.

Open the PC Bottleneck Calculator

Power Supply Calculator

The calculator adds the maximum sustained power of each part you choose:

  • CPU: Intel’s maximum turbo power (PL2) or AMD’s package power tracking limit (PPT), from the manufacturers’ spec pages.
  • GPU: the official total board power (NVIDIA TGP / AMD TBP).
  • Other parts: motherboard 30-70W, RAM 6-20W, storage 7-20W, fans 6-18W and cooling 3-30W, depending on what you select.
  • Overclocking: multiplies the total by the factor you choose.

Recommended PSU = total × 1.2, rounded up to the next 50W

The 20% headroom covers short power spikes from modern GPUs and keeps the PSU in its efficient load range.

Limits:

  • A higher-quality PSU matters as much as wattage; check independent reviews.
  • Some GPU models are factory overclocked and draw more than the reference figure.
  • If your graphics card maker recommends a larger PSU, follow that recommendation.
  • USB devices, RGB controllers and capture cards add a little extra load.

Open the Power Supply Calculator

FPS Calculator

Each game has a reference frame rate for an RTX 4090 at 1080p High. The estimate is then scaled step by step:

  • GPU: reference FPS × (GPU score ÷ 100), where the RTX 4090 = 100.
  • Resolution: 1080p × 1.0, 1440p × 0.7, 4K × 0.4.
  • Quality: Low × 1.4, Medium × 1.2, High × 1.0, Ultra × 0.85, Ray tracing × 0.5.
  • Caps: games with a fixed frame-rate cap (such as Elden Ring at 60 FPS) are capped.

Estimated FPS = reference FPS × GPU factor × resolution factor × quality factor

Limits:

  • It assumes your CPU can keep up; at 1080p a slower CPU can lower real FPS.
  • Upscaling (DLSS, FSR) and frame generation are not included.
  • Game patches and drivers change performance over time.
  • Use it to compare GPUs and settings, then check real benchmarks for a specific purchase.

Open the FPS Calculator

KD Calculator

K/D = kills ÷ deaths and KDA = (kills + assists) ÷ deaths

If you enter the number of games, the calculator also shows average kills and deaths per game. The session tracker adds up the kills and deaths from every match you enter before dividing, which gives a more accurate overall K/D than averaging each match’s ratio.

Limits:

  • Some games weight assists differently in their own KDA; check your game’s definition.
  • Deaths of zero are treated as one so the ratio can be calculated.

Open the KD Calculator

DPI Analyzer

eDPI = mouse DPI × in-game sensitivity

cm/360 = 360 ÷ (DPI × sensitivity × yaw) × 2.54

“Yaw” is how many degrees a game turns per mouse count at sensitivity 1. We use 0.022 for Counter-Strike 2 and Apex Legends, 0.07 for Valorant, and 0.0066 for Overwatch 2 and Call of Duty (MW 2019 and later). To convert between games, the calculator keeps cm/360 the same: new sensitivity = old sensitivity × old yaw ÷ new yaw.

Limits:

  • Values are for hip-fire; zoom and ADS multipliers differ per game.
  • Fortnite and Rainbow Six Siege use extra sensitivity settings, so only eDPI is shown for them.
  • Games occasionally change their sensitivity scales in updates.

Open the DPI Analyzer

Mouse Speed Test

While the test runs, the page adds up the distance your pointer travels in screen pixels and divides it by the elapsed time, giving an average speed in pixels per second.

For the polling estimate, the page counts raw pointer reports (the pointerrawupdate event in Chrome and Edge, or coalesced pointer events in other browsers) and divides them by the time the mouse was actually moving.

Limits:

  • Pixel speed depends on your DPI, Windows pointer speed and display scaling, so compare results only on the same setup.
  • Browsers may group or cap pointer events, so polling readings above about 1000Hz are unreliable.
  • For an exact polling rate, check your mouse software.

Open the Mouse Speed Test

Keyboard Latency & Reaction Time Test

After you start an attempt, the box turns green after a random delay of 1 to 3 seconds. The page records the time between the colour change and your input, and averages all attempts.

Limits:

  • The result is mostly your reaction time (often around 200-250 ms), plus monitor, keyboard and browser delay. It cannot isolate keyboard latency on its own.
  • To compare keyboards, run many attempts on each with the same monitor and browser and compare the averages.
  • Frame rate and display refresh rate affect how fast the green colour appears.

Open the Keyboard Latency & Reaction Time Test

Keyboard Polling Rate Tester

USB keyboards send key changes only at fixed time steps: every 8 ms at 125Hz, 4 ms at 250Hz, 2 ms at 500Hz and 1 ms at 1000Hz. When you tap keys quickly, the gaps between your key presses and releases cluster on multiples of that step.

The test records the browser timestamp of every key press and release, takes the gaps between them, and checks which step size (8, 4, 2 or 1 ms) they line up with best. If none fits, your keyboard is faster than the browser can resolve.

Limits:

  • Holding a key down does not work: it measures Windows key repeat (about 30 per second), not polling.
  • Browsers that round timestamps to 1 ms cannot confirm 1000Hz.
  • It cannot tell 1000Hz, 4000Hz and 8000Hz apart reliably.
  • USB hubs, wireless dongles and power saving can lower the result.

Open the Keyboard Polling Rate Tester

APM Test

APM = number of clicks ÷ test length in seconds × 60

The test counts every click in the test area during the 10-second run and converts it to a per-minute rate.

Limits:

  • In games like StarCraft II or Dota 2, APM also counts keyboard commands, so in-game APM is usually higher than click-only APM.
  • Mouse switches and double-click faults can add or miss clicks.

Open the APM Test

Kohi Click Test

CPS (clicks per second) = total clicks ÷ test length in seconds

The timer starts on your first click and the result is ranked from beginner to legendary based on your CPS.

Limits:

  • Mouse debounce settings and switch condition affect results.
  • Some servers limit or flag very high CPS, so check the rules before using techniques like butterfly or drag clicking.

Open the Kohi Click Test

Pixels to Inches Converter

inches = pixels ÷ PPI, pixels = inches × PPI, diagonal = √(width² + height²) ÷ PPI

The default of 96 pixels per inch is the CSS reference value used by browsers and by Windows at 100% scaling. For print, 300 PPI is the usual target; for a monitor, use its real pixel density.

Limits:

  • A screen’s physical PPI depends on its size and resolution, for example a 27-inch 1440p monitor is about 109 PPI.
  • Printers and image editors may call this value DPI; for converting pixels it means the same thing.

Open the Pixels to Inches Converter

Changelog

  • 21 October 2026: DPI Analyzer rebuilt with per-game yaw values for correct cm/360 and a sensitivity converter (the old version applied Windows pointer speed and fixed divisors). Keyboard Polling Rate Tester changed to a tap-timing method (the old version measured Windows key repeat, about 30 per second on every keyboard). Mouse Speed Test polling estimate now counts raw pointer reports while the mouse is moving. Every tool page now shows its formula and limits.
  • 19 October 2026: Corrections to the Bottleneck, PSU and FPS calculators, and text fixes on the mouse, keyboard, DPI, KD and APM tools.

Report a problem

If a result looks wrong, email contact@techcalcu.com with the tool, your inputs and what you expected. Corrections are listed in the changelog above. See also our editorial policy.

Last updated: October 2026