OBS Canvas vs Output Resolution CPU Overhead

Understanding the performance impact of your OBS Studio settings is crucial for achieving a smooth, lag-free stream or recording. This article explains the technical differences between “Canvas (Base) Resolution” and “Output (Scaled) Resolution,” specifically focusing on how these two settings affect your CPU overhead and system resources.

Defining the Terms

To understand how these settings affect your CPU, you must first understand what each resolution does in OBS Studio:

• Canvas (Base) Resolution: This is the size of your virtual workspace in OBS. It dictates the dimensions of your preview window and is typically set to match your gaming monitor’s native resolution (e.g., 1920x1080 or 2560x1440).
• Output (Scaled) Resolution: This is the final resolution of the video file being recorded or streamed to platforms like Twitch or YouTube (e.g., 1280x720).

How Canvas Resolution Impacts Performance

Canvas Resolution has almost no direct impact on your CPU overhead.

OBS Studio utilizes your graphics card (GPU) to render the canvas, composite your sources (like webcams, gameplay, and overlays), and apply visual effects. A larger Canvas Resolution (such as 4K) will increase VRAM consumption and GPU utilization, but it will not noticeably increase CPU usage.

How Output Resolution Impacts Performance

Output Resolution has a massive impact on CPU overhead, especially if you are using a software encoder.

When you stream or record, your encoder must compress every pixel of every frame in real-time. The number of pixels scales exponentially with resolution: * 1080p (1920x1080): ~2.07 million pixels per frame * 720p (1280x720): ~0.92 million pixels per frame

If you are using x264 (CPU encoding), lowering your Output Resolution from 1080p to 720p reduces the encoder’s workload by over 50%, resulting in a massive reduction in CPU overhead. Even if you use a hardware encoder (like NVIDIA NVENC or AMD AMF), a lower Output Resolution reduces the strain on the dedicated encoding chips, though the impact on the main CPU remains minimal in hardware encoding scenarios.

The Performance Cost of Downscaling

When your Canvas Resolution and Output Resolution do not match, OBS must resize the video frame before encoding it. This process is called downscaling.

Downscaling is handled by the GPU, not the CPU. In OBS, you can select different downscale filters under the Video settings: * Bilinear: Fastest, but can look blurry. Uses the least GPU resources. * Bicubic: A good balance of sharpness and performance. * Lanczos: The sharpest, but uses the most GPU resources.

Because the GPU handles this scaling process, differing Canvas and Output resolutions do not increase CPU overhead. However, if your GPU is already running at 100% capacity while gaming, this scaling process can cause rendering lag.

If you match your Canvas Resolution and Output Resolution perfectly, OBS skips the downscaling process entirely, saving minor GPU resources.