Capture card vs hardware encoder: what's the actual difference?

The short answer: a capture card gets video into a computer, and a hardware encoder puts video onto the network without one. If your production runs on a PC, you need a capture card. If you want a box that streams by itself, you need a hardware encoder. This is the most common question we are asked about the Magewell range, and the diagram below shows the one decision it turns on.

A capture card (the USB Capture, Pro Capture and Eco Capture ranges) gets video into a computer. It connects to an HDMI or SDI source on one side and to a PC over USB, PCIe or M.2 on the other, and presents that signal to software such as OBS, Teams or Zoom as a video source it can select. The computer does the streaming, recording and encoding; the card's job is to deliver the picture and sound to it. You need a capable PC, because its performance and software settings decide the quality of what you send out. In return you get everything the software can do: mixing, graphics, recording and conferencing.

A hardware encoder (the Ultra Encode range and the Pro Convert encoders) does the opposite. It is a standalone box that takes HDMI or SDI in and sends a network stream out, with no computer involved. Ultra Encode models stream to platforms and other sites over RTMP, SRT and NDI HX2; Pro Convert encoders put a source onto a local network as NDI High Bandwidth. The encoding happens inside the device. This is the right choice when you do not want a PC in the signal chain: at a remote venue, in a rack with no monitor or keyboard, or anywhere simplicity matters more than the flexibility of software.

Rule of thumb: if you are already building on a PC-based production setup (OBS, vMix or another software switcher), you want a capture card. If you want a self-contained box that streams or converts with nothing else attached, you want a hardware encoder. Some installations use both: capture cards in the production PC, and an encoder to send the finished programme on.