HDR Display Capability Check

Verify if your monitor is outputting true HDR signal.

Visual comparison (for "subjective judgment")
Recommendations: Full screen, turn off Night Sight/Eye Protection, turn on system HDR, and increase the brightness; then observe "whether the highlights are obviously brighter and whether there are layers in the shadows."
Black(0)
rgb(0 0 0)
SDR White (1.0)
rgb(255 255 255)
SDR gray scale (0→1)
"HDR Try" highlight gradient (requires CSS Color 4 + HDR output link)
Note: "The browser can display HDR" and "can decode HDR video" are two different things; and even if it supports it, it may be affected by the system HDR switch, color management, external monitor, and player path.
Local video loading (recommended HDR video shot with your device)
iPhone/some Android "HDR/Dolby Vision" materials are the most valuable for reference; if the picture is gray or the highlights are not bright during playback, it may be that the HDR output link is not turned on or is being transferred to SDR.
resolution
-
duration
-
hint
Try to play it in full screen and observe the highlights
Display/browser capabilities
Conclusion (experience):More likely SDR output
dynamic-range: high
unknown
color-gamut: rec2020
unknown
color-gamut: p3
unknown
CSS color(display-p3)
unknown
CSS color(rec2020)
unknown
CSS rec2020 > 1(HDR)
unknown
Reminder: Some systems will force SDR under "External Display/Screen Mirroring/Power Saving Mode/Night Shift"; even if the display is HDR, it may be downgraded.
HDR video decoding capability detection (MediaCapabilities)
"Supported" doesn't necessarily mean displaying in HDR; it's closer to "decodable." Whether HDR output still depends on the system/monitor/player path.
FormatSupport decodingfluencyPower saving
SDR: H.264/MP4 (1080p@30)---
HDR10: HEVC Main10/MP4 (4K@30, PQ/Rec2020)---
HLG: HEVC/MP4 (4K@30, HLG/Rec2020)---
HDR: VP9 Profile 2/WebM (4K@30, PQ/Rec2020)---
HDR: AV1/WebM (4K@30, PQ/Rec2020)---
Explanation: smooth/powerEfficient can often be used to "guess whether to use hard solution/hardware acceleration", but it is still subject to actual playback.

How to use this page to quickly determine whether HDR "really takes effect"

First look at "dynamic-range: high" and "color-gamut: rec2020" on the right side: both are "yes" at the same time, which makes it more like an HDR output link.
Look at "CSS rec2020 > 1 (HDR value)" again: If it is "Yes", use the highlight color block (1.2/1.6/2.0) in the top "Visual Comparison" to compare with SDR white. Normal HDR will be "brighter" and have more "highlight impact".
Even if the "Decoding Support" shows that it supports HDR10/HLG, it may be converted to SDR playback by the system/browser (a common phenomenon is that the overall image is gray and the highlights are not bright). Be sure to use the HDR material you shot/downloaded to confirm.
External monitors/screen projection/power saving mode/eye protection mode/night viewing often trigger SDR degradation; it is recommended to first use the built-in screen, turn off night viewing, increase the brightness, and go to full screen before observing.

HDR Guide

Combine capability scan with visual high-light checks.

Step 1

Make a “capability judgment” first

about 10 seconds

Determine whether the HDR output link may be enabled through the browser's media query/CSS support.

Observe the results such as "dynamic-range: high" and "color-gamut: rec2020" on the right
If dynamic-range: high is No, it can basically be processed as SDR (unless the browser/system does not expose this capability)
If rec2020/p3 is yes, it means that there is at least the possibility of wide color gamut output
Step 2

Do visual highlight contrast

about 20 seconds

Observe "whether the highlights are brighter and whether the dark parts have layers" through color blocks and gradients.

Click "Full Screen Contrast", increase the system brightness, and turn off Night View/Eye Protection
Compare SDR White vs “HDR Highlight (1.6/2.0)”: If there is almost no difference, the common reason is that it was converted to SDR
Compare SDR grayscale: whether the dark parts (close to black) can be separated into layers and whether they are "blurred"
Tip: The "highlight impact" of different devices varies greatly; please give priority to "same-machine comparison" (HDR on/off, different browsers, external/internal screens).
Step 3

Live broadcast confirmation using HDR material

about 30 seconds

Loading a local HDR video (shot/downloaded on a mobile phone) and actually playing it is one of the most reliable ways to judge.

Upload a video that you confirm is HDR (for example, a clip recorded by "HDR Video" on your mobile phone)
Observe after full-screen playback: whether highlights (lights/reflections/sky) are significantly brighter, and whether details are still retained in dark areas
If "the whole is gray/the highlights are not bright/the color is strange", it is mostly due to HDR→SDR conversion or color management problems; you can change the browser/change the player path to verify

FAQ

Focusing on the differences between HDR output, color gamut, decoding and actual look and feel, we have sorted out the most common judgment and troubleshooting paths.

1.

Can this page "measure the maximum brightness of the screen (nits)"?

cannot. Browsers generally cannot directly read the absolute upper limit of brightness (nits) of the screen. This page provides a combination of "capability detection + subjective comparison": use dynamic-range/color-gamut/MediaCapabilities to determine the possibility of support, and then use highlight contrast and HDR material live broadcast to confirm whether it is really output in HDR.

2.

dynamic-range: high = Yes, does it have to be HDR?

Not necessarily, but it’s a strong signal. dynamic-range: high reflects the browser's judgment of "high dynamic range output capability"; actual HDR is also affected by the system HDR switch, external monitor, color management, and player path. It is recommended to do "Highlight Color Block Comparison" and "HDR Video Live Confirmation".

3.

The decoding detection shows that HDR10/HLG is supported, but the picture looks like SDR. Why is this?

Very common. Decoding support only means "can decode this type of code stream", it does not mean that the output will be displayed in HDR. In many cases, browsers/systems will perform tone mapping on HDR content and convert it to SDR (showing that the overall content is gray and the highlights are not bright). Please make sure that system HDR is turned on, night view/eye protection is turned off, and try to play it in full screen on the built-in screen before making a judgment.

4.

Why are HDR results different on the same device on different browsers?

Different browsers have different HDR pipelines, color management, hard decoding strategies, and implementation of CSS Color 4 / MediaCapabilities. Especially HEVC/HDR10 has greater differences under different system licenses. It is recommended to compare Chrome/Edge with Safari/Firefox at least once (if available).

5.

CSS rec2020 > 1 (HDR value) shows "not supported", what does it mean?

This usually means that the browser does not support "HDR highlight values ​​beyond 1" expressed directly in CSS, or the current output link does not allow it. Even so, the device may still implement HDR through the "video playback pipeline" (such as when playing HDR video). Therefore, this capability is more biased towards "web page rendering HDR highlights" and is not equivalent to "video HDR".

6.

What footage should I use to verify HDR effects?

It is recommended to use HDR video shot with your mobile phone ("HDR Video/Dolby Vision" for iPhone/some Androids), or test footage that you confirm is HDR10/HLG. Loading and playing it as a local file, and then observing the highlights and shadow details, is one of the most intuitive and reliable ways.

7.

Will external monitor/screen projection affect HDR judgment?

meeting. External monitors, docking stations, screen casting protocols, and system "mirror/extended" modes can all cause HDR degradation or color management changes. It is recommended to test on the built-in screen first for benchmarking, and then test the external link for comparison.