Before you start
Webcam resolution on a product listing and the track delivered to a browser are different statements. A camera advertised as 1080p may currently deliver 640×480. Record that difference before assigning a cause.
Choose the intended device. A built-in and external webcam may both appear, and an app can select another one. Keep camera, connection and room conditions fixed for the first comparison.
Is the problem delivered width and height, soft motion, or what the call recipient sees? These require separate checks. A local track setting does not answer the entire image-quality question.
Run the check
Open the webcam test, grant access and select the intended camera. Size request says Auto because this tool does not ask for fixed dimensions. Delivered track comes from MediaStreamTrack.getSettings and describes this session, not the sensor's maximum capability. Track maxima lists separate browser capability limits, not a guaranteed combined mode.
Record the webcam resolution mode before closing another camera user or changing a USB connection. Restart the stream after one change and compare its settings. A changed value is evidence about this setup, not proof that one particular bandwidth limit or sharing mechanism caused it.
Move through the preview to verify frame delivery. Preview area is the video element's size in CSS pixels and may include letterboxing. It is not a second camera mode. A static frame does not become live because track settings report dimensions.
If another browser reports a different mode, record both selected devices and applications. Requests and processing paths can differ. Neither local reading establishes what a remote participant will see.
Read your result
The delivered dimensions are what this browser track reports. They do not prove the camera supports or lacks another mode. A 640×480 result calls for a controlled device and settings comparison, not an immediate driver-replacement conclusion.
Could a shared connection affect the result? It is one hypothesis, but this page does not measure USB bandwidth. Compare a documented direct connection with the existing one while keeping the camera and request fixed.
A call app can request or transmit a different mode after this browser test. Dim light can affect exposure and perceived sharpness without changing width and height. Separate those observations before naming the limiting layer.
Measurement limits
MediaStreamTrack.getSettings reports the current track. It cannot enumerate every sensor capability or promise advertised 1080p in every browser and route. Do not turn a delivered value into a universal device specification.
A high-resolution stream can look soft because focus, lighting and processing are separate properties. The page does not measure optical resolution or a call service's encoded image quality.
What a remote participant sees is outside this preview. Conferencing software can change the image again. Keep local track readings and the meeting app's observation labelled by source.
What to do next
If closing another camera user changes the mode, record both observations. Keep that app choice stable for the next call. The change does not prove permanent hardware damage.
If a direct connection changes the mode, repeat once before changing drivers. The browser cannot prove a hub alone imposed a bandwidth limit; report the connection comparison without assigning an unmeasured cause.
If no tested setup delivers the expected mode, consult the model's current specifications and support. Record selected device, connection and track settings. A browser comparison cannot itself establish a firmware defect or warranty entitlement.
640 × 480 vs 1920 × 1080
Size examples for explanation, not measured camera results.
| Dimensions | Aspect ratio | Pixel count |
|---|---|---|
| 640 × 480 | 4:3 | 307,200 |
| 1920 × 1080 | 16:9 | 2,073,600 |
Width × height gives the pixel count: 2,073,600 ÷ 307,200 = 6.75. That is an arithmetic ratio, not a claim of 6.75 times the sharpness or bandwidth.
Full reference scene. Dashed lines mark the crop used in the 16:9 example.
Same scene, with the top and bottom removed in this crop example.
Original vector illustrations, not actual webcam screenshots. Both frames use the same scene at the same displayed width; the 16:9 frame illustrates a possible central crop only. The artwork cannot compare camera sharpness.
- Crop: part of the scene is removed to fill the frame, as illustrated above.
- Black bars (letterbox / pillarbox): the composition is kept with empty bars above and below or at the sides.
- Different field of view: another capture mode can show more or less of the scene.
Dimensions alone cannot tell you which strategy a camera or app uses. Compare visible landmarks in your own setup before recording a crop, black bars or a changed field of view.
Printable observation record
Print this page with your browser’s Print command and fill the spaces by hand. Keep “Not measured” or “Not found” when you have no evidence; these are blank observations, not a diagnosis.
In the existing webcam test, start the camera only when you choose to allow access, select your intended device, then copy the Delivered track width and height into their separate columns below. Read Preview area (CSS px) separately. A dash means no value to record yet; Size request: Auto and Track maxima are not the delivered mode.
For the repeat, keep device, app, lighting and connection fixed except for the single condition you write below. Restart the stream after that change, copy the delivered dimensions again, and record a remote call observation separately. Leave unknowns unknown. This guide requests no camera access and saves no entries; review the privacy policy before using the free tool.
- □ Confirm device, app, lighting and connection are fixed for the comparison.
- □ Change only the single condition written in the record.
- □ Keep track dimensions, CSS preview and remote observations separate.
- □ Repeat once; retain unknowns instead of assigning a hardware cause.
| Run / webcam test | Track width (px) | Track height (px) |
|---|---|---|
| Baseline | Not measured: __________ | Not measured: __________ |
| Repeat | Not measured: __________ | Not measured: __________ |
| Observation / source | Baseline / notes | Repeat / notes |
|---|---|---|
| Selected device and application | Not recorded: __________ | Not recorded: __________ |
| Lighting and connection | Not recorded: __________ | Not recorded: __________ |
| Advertised dimensions / source | Not found: __________ | Not found: __________ |
| Preview area / CSS pixels | Not measured: __________ | Not measured: __________ |
| Remote application / recipient observation | Not measured: __________ | Not measured: __________ |
| One condition changed | Baseline: __________ | Only change: __________ |
| Repeat result / remaining unknowns | Not measured: __________ | Not measured: __________ |
Symptom table
| Symptom | Likely layer | Next action |
|---|---|---|
| 1080p request delivers 640×480 | Negotiation or concurrent use | Read track settings and close other camera users |
| Aspect ratio changes | Fallback mode selection | Compare supported modes and application constraints |
| Frame rate falls at higher size | Bandwidth or exposure tradeoff | Compare frame rate at each negotiated mode |
Common questions
Why is my 1080p webcam sending 640x480?
A delivered 640x480 mode is a track observation, not a complete diagnosis. Check selected camera, close other camera users and repeat under the same light and connection. Application constraints can matter, but this page cannot measure USB bandwidth or prove firmware damage from one stream.
How can I check what resolution my webcam is really sending?
Open the webcam test, choose the intended device and read width, height and frame rate from its active track after permission. These getSettings values describe the browser's current stream. They do not list every supported sensor mode or what a call participant receives.
Does a higher resolution always look better on a call?
No. Resolution is one property; light, focus and processing affect the preview, and a call service may encode another stream. Compare the same camera and room in the app you will use rather than treating larger dimensions as guaranteed picture quality.