FPS and Frame Time Calculator

Frames per second tells you how fast a game runs on average, but it does not tell you how smooth it feels. Two games can both show 60 FPS and feel completely different, because what your eyes notice is the time between frames, and the worst frames matter more than the average. This tool converts between FPS and frame time, and it analyzes a list of frame times from a capture to show average FPS, 1% lows, and stutter.

Everything runs in your browser. Nothing you paste is stored or sent anywhere.

FPS versus frame time

Frame time is how long one frame takes to draw, in milliseconds. FPS is just 1,000 divided by that number. At 60 FPS each frame takes about 16.7 ms, at 120 FPS about 8.3 ms, and at 240 FPS about 4.2 ms. Frame time is the more honest number, because it shows each frame individually. A game that averages 60 FPS can still contain a frame that took 50 ms, and that single slow frame is the stutter you feel.

The conversion also explains why going from 30 to 60 FPS feels like a huge jump while going from 120 to 150 feels small. The first saves about 16.7 ms per frame, and the second saves only about 1.7 ms.

What 1% lows mean

The 1% low is a way to measure how bad the slowest frames are. In this tool it is the average FPS of the slowest 1% of the frames in your capture, and the 0.1% low is the same idea for the slowest 0.1%. If your average is 100 FPS and your 1% low is 90, the game is consistent. If your average is 100 and your 1% low is 40, the game has regular hitches, even though the average looks great.

Programs define lows in slightly different ways. Some use the slowest 1% of frames, and others use the 99th percentile frame time. The two give similar but not identical numbers, so compare results only from the same tool and method.

How to get frame times from a game

1. Use a capture program that can log frame times. PresentMon and tools built on it, such as CapFrameX, can export them to a file, and many overlay tools offer a similar log.
2. Play the same repeatable section of the game for at least a minute, so you capture several thousand frames.
3. Open the exported file and copy the column with frame times in milliseconds, often called MsBetweenPresents or frametime.
4. Paste the values into the box above and press Analyze.

Test the same scene before and after you change a setting, so you can compare like with like. Run at least two or three captures per setting, since a single run can be affected by a background task.

Reading the results

• Average FPS is the total number of frames divided by the total time. It describes overall speed.
• 1% and 0.1% lows describe smoothness. The closer they are to the average, the more even the game feels.
• Worst frame time shows the single longest hitch. A value above 50 to 100 ms is usually clearly visible.
• Stutter frames counts frames that took more than twice the median frame time, which is a simple way to spot spikes.

The refresh rate table

The conversion section also shows how many frames your game produces per monitor refresh. A value of 1 means one new frame for each refresh. Below 1, the monitor shows the same frame more than once, which can look like judder on a standard display. Variable refresh rate technologies are designed to reduce that effect by matching the monitor to the game, but they work best when frame times are steady, which is another reason lows matter.

Common causes of low 1% lows

• Not enough memory, so the system swaps data to the drive. Our guide on why computers slow down over time covers this.
• Heat. A hot processor or graphics card slows itself down. See why your phone gets warm while charging for the same idea in a smaller device.
• Background programs, such as updaters and overlays, that wake up while you play.
• Shader compilation, which happens the first time a game shows an effect and shows up as spikes that disappear on a second run.
• A slow network connection in online games, which can feel like stutter even when frame times are fine. Our piece on why Wi-Fi gets slower with distance explains part of that.

Limits of this tool

The tool analyzes the numbers you paste. It cannot tell you why a frame was slow, and a short capture gives rough lows. It also does not measure input lag, which depends on the whole chain from your mouse to the screen. Use it to compare settings and spot stutter, not as a precise benchmark.