For the original ring-light Razer Kiyo, start with a supported video mode and a simple OBS scene. Get a stable, well-lit face image before adding effects or extra sources. This walkthrough uses OBS on Windows; menu labels and available camera controls can vary with the driver and OBS version.
Confirm the original Kiyo and choose a mode
The original Kiyo is model family RZ19-02320. Razer lists 1920 × 1080 at 30 frames per second and 1280 × 720 at 60 frames per second. Newer Kiyo models have different capabilities. Use the original-model support page for manuals and any applicable firmware.
On small screens, swipe the table sideways. With a keyboard, focus the table and use the arrow keys.
| Mode | Useful starting situation | What to inspect |
|---|---|---|
| 1080p, 30 fps | Large face view, presentation, mostly seated speech. | Fine detail, focus and whether motion is acceptable. |
| 720p, 60 fps | Smaller face overlay where smoother motion matters. | Actual movement in the recording and sufficient light. |
Setting the whole OBS output to 60 fps does not turn a 30 fps camera source into 60 distinct camera frames. Set the camera source and the final recording deliberately.
Build a clean OBS camera scene
- Connect the Kiyo directly to a suitable computer USB port for the initial test. Close other apps that may be using it.
- In OBS, create a simple scene and add a Video Capture Device source. Select the Kiyo explicitly.
- Open that source's properties. Choose a supported resolution/FPS combination; use Custom if the default produces the wrong mode.
- Fit the source to your intended composition. Keep the original aspect ratio, then crop if needed.
- Make a short local recording and play it back before configuring a live broadcast.
The OBS video-source documentation explains device selection, custom resolution/FPS and Configure Video. Unsupported combinations can produce a blank source. Available controls depend on what the camera driver exposes.
Use one deliberate microphone source. If a separate USB microphone is carrying your voice, check that the webcam microphone is not also being captured. The Seiren X setup guide explains the audio-routing test.
Adjust light before pushing exposure
Place the camera near eye level and face a soft light source. A bright window behind you can leave the face underexposed. Begin with modest ring-light brightness and enough room light to keep the camera from struggling. Razer says the ring light activates while the camera is in use and its dial adjusts brightness.
Record a face turn and a hand movement. If the face is bright but moving edges smear, add useful light and revisit exposure in Configure Video where available. Exposure is the time used to gather light for an image; longer exposure can blur movement. Raising brightness in software alone cannot recover detail that was never captured.
For glasses reflections, slightly change the camera/light angle and reduce the ring brightness. Compare the eyes in the recording after each change. If focus repeatedly shifts to the background, simplify the background and try manual focus if the installed controls provide it.
Run a repeatable one-minute check
Keep the same seat position and lighting while changing one setting. Save a short note with resolution, FPS and the changed control. A visibly better preview is useful; a recorded file confirms what the audience will receive.
- Blank video: select the camera again, close competing camera apps, check camera permissions, and return to a supported mode.
- Flicker or bands: Razer recommends setting the camera's frequency control to match the local 50/60 Hz environment. Test the actual lights as dimmers and LED drivers can behave differently.
- Intermittent freezing: repeat with a direct USB connection and a scene containing only the camera; this separates source trouble from a complex recording workload.
- Settings change after restart: reopen source properties and compare your saved notes before recording.
Download the camera and microphone test log to record the baseline, one change and the playback result.