capframex.com

CapFrameX reports several performance metrics because no single number describes how a game feels. FPS tells you the average throughput, frametimes tell you the consistency, the median and percentiles describe the distribution, and the x%-low figures summarize the worst part of that distribution. If you only look at average FPS, you can miss stutter that a frametime or 1%-low reading would expose. The explanations below apply to any CapFrameX capture, since the tool is built on Intel's PresentMon for frametime capture and uses Rivatuner Statistics Server for its overlay.

Frametimes vs. FPS: two views of the same data

Frametime is the time between two consecutive frames, usually measured in milliseconds. FPS is simply the reciprocal of frametime, scaled to one second:

• 16.7 ms per frame ≈ 60 FPS
• 8.3 ms per frame ≈ 120 FPS
• 33.3 ms per frame ≈ 30 FPS

Because the relationship is non-linear, a single long frame does not move average FPS much, but it is very visible as a hitch. That is why CapFrameX keeps the raw frametime series and derives FPS from it rather than the other way around.

CapFrameX's own blog notes a related quirk: analysis can show FPS values lower than what you see in-game. This happens because the tool measures the actual presented frames over the captured interval, including frames that were dropped or delayed, while an in-game counter may report a different sampling window.

Median and percentiles

Once you have a list of frametimes, you can describe its distribution:

The median is useful because it is not dragged around by a few very long frames the way the average is. If your average FPS looks fine but your median frametime is noticeably higher, the average is being propped up by a burst of fast frames while most of the session ran slower.

Percentiles work the same way in reverse for frametimes: the 99th percentile frametime is the value that 99% of frames stayed under, so it captures the slow tail. A low 99th percentile frametime means even your worst frames were not far from the rest.

What x%-low means

The x%-low metric (commonly 1%-low and 0.1%-low) is the average FPS computed from only the slowest x% of frames. It answers the question: "How bad were the worst moments?"

• 1%-low: average FPS of the slowest 1% of frames. A good indicator of occasional stutter.
• 0.1%-low: average FPS of the slowest 0.1% of frames. Sensitive to rare, severe hitches such as shader compilation stalls or asset streaming spikes.

A system can post a high average FPS and still have a poor 1%-low if it hitches regularly. Comparing average FPS against 1%-low is one of the fastest ways to judge whether a frame rate is smooth or merely high on paper.

How to use these together

1. Capture a run in CapFrameX while playing the scene you care about.
2. Look at average FPS and median frametime first to gauge the typical experience.
3. Check the 1%-low and 0.1%-low to see how the worst frames compare.
4. If average FPS is high but 1%-low is far below it, investigate the frametime graph for spikes rather than chasing a higher average.

The same logic applies when comparing two configurations: match the scene, capture length, and settings, then compare the full set of metrics rather than a single headline number.

CapFrameX records FPS, frametimes, and sensor data while you play a game, then lets you analyze and plot the results or share them through its cloud service. To do this you need three pieces in place: CapFrameX itself, Intel's PresentMon (the capture layer CapFrameX is built on), and RivaTuner Statistics Server (RTSS), which provides the in-game overlay. Once those are installed, you start a capture while the game is running and the tool logs the frame data for later analysis.

What you need before recording

The site describes CapFrameX as a "Frametimes capture and analysis tool based on Intel's PresentMon" with an "Overlay provided by Rivatuner Statistics Server." That means the actual frame timing data comes through PresentMon, and the on-screen overlay you see during play comes from RTSS — so both need to be present and working for a capture to succeed.

Recording a capture

1. Install the prerequisites. Get CapFrameX, and make sure PresentMon and RTSS are available as described above. The site's primary call to action is a Download link, which is where you obtain the tool.
2. Launch your game first. Capture depends on a game actively rendering frames, so start the game and get to the scene or gameplay you want to measure.
3. Start the capture in CapFrameX. Begin recording while the game is running. CapFrameX will log FPS, frametimes, and sensor data for the duration of the capture.
4. Play through the segment you want to measure. Keep the capture running for the period that represents the workload you care about.
5. Stop the capture. The recorded data is then available for analysis.
6. Analyze or share. Use CapFrameX to analyze and plot the results, or share them with friends through the cloud service.

The expected result is a captured dataset of FPS and frametime values (plus sensor readings) that you can review as plots and metrics rather than relying on a momentary on-screen number.

Understanding the metrics you'll see

CapFrameX's own blog posts explain that the tool reports several performance metrics, including frametimes, FPS, median, percentiles, and x%-low values. One featured post specifically addresses a common confusion: why analysis can show FPS values lower than what you saw in-game. This matters because the FPS number displayed live during play and the FPS computed from a full frametime capture are not the same measurement — the analysis reflects the complete distribution of frames, including slow ones, which is exactly why frametime-based metrics like percentiles and x%-low are useful for judging smoothness rather than just average speed.

Common sticking points

• No overlay or no data: Since the overlay comes from RTSS and capture comes from PresentMon, a missing or misconfigured one of these is a likely cause of a capture that doesn't start or shows nothing.
• FPS looks lower than in-game: This is expected behavior that CapFrameX documents directly — the analysis is derived from frametimes across the whole capture, so it can differ from the live in-game counter.
• Capture without a game running: The tool records frames while a game runs, so there is nothing to capture if no game is rendering.

For the exact current download, installation steps, and any configuration specifics, use the Download link and the metrics-explained blog posts on capframex.com, since those reflect the tool's current state.

CapFrameX often reports a lower average FPS than the in-game counter because the two tools measure different things over different time windows. CapFrameX captures every frame's frametime through Intel's PresentMon and then derives FPS from that complete frame data, while an in-game counter typically shows a smoothed, instantaneous rate. The gap is expected, not a bug, and it usually means CapFrameX is giving you the more honest number for the captured period.

Where the difference comes from

Frametime vs. FPS as the raw measurement

CapFrameX records frametimes — the time each frame takes to render — and calculates FPS from them. The site's own blog post "The challenge of displaying performance metrics as FPS" addresses exactly this question: why analysis shows FPS values lower than what you saw in the game.

An in-game counter is usually a running average or a smoothed instantaneous value, updated frequently and often rounded. It answers "how fast does it feel right now." CapFrameX answers "what did every captured frame actually cost," which is a stricter question.

The averaging method matters

A simple average of FPS values is not the same as the reciprocal of the average frametime. Because FPS is a rate (frames per second), short slow frames — the stutters — pull the average down more than they appear to in a smoothed counter. If CapFrameX reports a lower average, it is often because those slow frames are included rather than averaged away.

The capture window is complete, not selective

The in-game counter runs continuously and you tend to read it during the smooth parts. CapFrameX analyzes a defined capture window that includes loading transitions, shader compilation hitches, and scene changes if they fall inside it. A capture that includes a single heavy spike will show a lower average than the counter you glanced at during normal play.

How to read the numbers correctly

The site's "metrics explained" post covers frametimes, FPS, median, percentiles, and x%-low, and is the reference for how CapFrameX calculates each one.

Practical steps to make a fair comparison

1. Capture a representative window. Start recording after the scene has loaded and shaders have compiled, so the capture reflects gameplay rather than startup.
2. Compare like with like. Put the in-game counter and CapFrameX side by side over the same segment. If the counter is a smoothed average, expect it to sit above the CapFrameX average.
3. Look past the average. Check median and x%-low. A high average with a poor x%-low means the game feels worse than the average suggests — and that is the information the in-game counter is hiding.
4. Repeat the capture. One run can be skewed by a single background task. Two or three captures of the same route give you a stable picture.

When a lower number is actually a problem

A lower CapFrameX average is normal. It is worth investigating when:

• The gap is large and consistent, not just a few percent.
• x%-low is far below the average, indicating regular stutter.
• Frametime shows periodic spikes at a fixed interval, which can point to a background process or a driver issue rather than the game itself.

In those cases the difference is not a measurement artifact — CapFrameX is showing you a real performance characteristic that the in-game counter does not surface.

The takeaway

Treat CapFrameX as the detailed record and the in-game counter as a quick glance. When they disagree, CapFrameX is usually right about the captured window, because it works from complete frametime data rather than a smoothed rate. Use the average for a headline figure, and median, percentiles, and x%-low to judge how the game actually feels.

CapFrameX is a Windows benchmarking tool for recording and analyzing game performance. It captures FPS, frametimes, and sensor data while you play, then lets you analyze, plot, and share the results. It is built on Intel's PresentMon for capture and uses RivaTuner Statistics Server (RTSS) for its overlay. If you need detailed, reproducible performance data rather than a single average FPS number, this is the kind of tool you are looking for.

What it captures

While a game is running, CapFrameX records:

• FPS — frames per second over time
• Frametimes — the time each individual frame takes to render
• Sensor data — hardware telemetry captured alongside the run

The tool is described as a source for "detailed and reliable benchmarks," which is the core distinction from simple FPS counters: it keeps the raw per-frame data so you can analyze it after the fact.

How it works under the hood

Two components do the heavy lifting:

Because capture is based on PresentMon, the data reflects what the presentation pipeline actually delivers, not an estimate. The overlay is provided separately by RTSS, so RTSS needs to be present for the overlay to function.

Why frametimes matter more than average FPS

CapFrameX's own blog posts address a common confusion: why analysis can show FPS values lower than what you see in-game. The explanation is that average FPS hides frame-to-frame variation. Two runs can share the same average FPS while one feels smooth and the other stutters. CapFrameX therefore reports metrics such as:

• Frametimes — the per-frame timing that reveals stutter
• Median — the middle value, less skewed by outliers than the mean
• Percentiles — the distribution of performance across the run
• x%-low — the average of the worst x% of frames, a common measure of worst-case smoothness

The site explicitly documents how CapFrameX calculates FPS, so the numbers are traceable rather than opaque.

What you can do with the results

• Analyze a captured run using the built-in metrics
• Plot the data to visualize frametime spikes and trends
• Share results with others through the tool's cloud service

That last point matters for comparisons: instead of trading screenshots, you can share the underlying capture.

Who it is for

CapFrameX suits users who need reliable, detailed benchmarking data — for example, comparing two GPUs, testing a driver update, tuning in-game settings, or verifying that a system maintains consistent frametimes under load. If you only want a quick FPS readout, a simpler overlay is enough; if you need to prove or diagnose performance behavior, the per-frame data here is the point.

To get started, download it from capframex.com and run a capture while playing your game.

You can get CapFrameX from its official site, capframex.com, where the download entry and the cloud-sharing service for benchmark results are both hosted. This applies if you want to record FPS, frametimes, and sensor data while gaming, then analyze the results or send them to others. The site also links to documentation and blog posts explaining how the metrics are calculated.

Downloading CapFrameX

The homepage presents CapFrameX as a "Frametimes capture and analysis tool based on Intel's PresentMon," with its overlay "provided by Rivatuner Statistics Server." The download entry sits on that same page.

Before you start, note two dependencies the site names explicitly:

• Intel PresentMon — the capture foundation CapFrameX is built on.
• Rivatuner Statistics Server (RTSS) — supplies the in-game overlay.

The site describes the tool's purpose as recording "Frames per Second (FPS), Frametime and Sensors while running your favourite game," then analyzing and plotting the results. The download link is the starting point for all of that; the page does not state pricing or login requirements for the download itself, so treat those as unconfirmed rather than assuming the tool is free or account-free.

Sharing Results Through the Cloud Service

The same source description states you can "share them with friends using our cloud service." That is the distribution path for benchmark data: capture a run, analyze it in the tool, then publish it through the cloud service so others can view the results.

The site does not document the exact sharing steps, account requirements, or link formats on the pages captured here. If you need those specifics, check the site directly rather than relying on secondhand instructions.

What Else the Site Offers

Beyond the download and cloud service, capframex.com carries explanatory material. Two featured blog posts are listed:

The second post addresses a common confusion directly: "Why does my analysis show fps values that are lower than what I've seen in the game?" If your recorded numbers look off compared to the game's own counter, that article is the intended explanation.

Who This Fits

This setup suits anyone who needs to distribute or present benchmark data — reviewers, hardware testers, or players comparing configurations — because the cloud service turns a local capture into something shareable. If you only want a quick FPS readout, the overlay plus the tool's analysis may be more than you need; the sharing feature matters most when someone else has to see the numbers.