From our perspective, the choice between software or hardware options for AV-over-IP in control rooms depends on the specific project requirements and use case scenarios. In some cases, a software implementation may be preferable, while in others, a hardware approach may be more suitable. Capture (Encoding) and Streaming of Image using AV-over-IP.
In hardware solutions, workstations are configured in such a way that the operator's desktop is duplicated on two outputs: one is connected to a monitor (or multiple monitors), and the other is connected to an encoding device (encoder) that captures the signal, converts it into a video stream, and sends it over the local network (streaming). If there are no available video outputs on the workstation (for example, if all existing video outputs are connected to separate monitors), the video cable from the workstation is first connected to a 1-to-2 splitter, which divides the signal into two - the first goes to the monitor, and the second goes to the encoding device. This way, the image not only appears on the screen but also gets transmitted over the network.
In software solutions (as an example, let's consider the mature and proven Polywall Streamer solution that appeared in 2019), the capture and streaming of video signals are performed using the resources of the workstation itself: the processor, graphics card, and network adapter. Unlike hardware encoders, software streamers are not tied to physical video signals, which gives them much more freedom. For example, with a single workstation, multiple video streams can be created, each transmitting the content of one of the monitors or application windows.
In some cases, it is useful to provide the control room operator with remote control of the keyboard and mouse of the workstation or PC whose image is being transmitted to the video wall. Certain models of AV-over-IP hardware solutions (IP-KVM) allow for keyboard and mouse switching, while software solutions of this class use the outdated VNC technology for video transmission, which does not match the concept of "video over the network" in terms of image quality. However, development does not stand still, and within a year or two, similar functionality can be expected from software solutions in the AV-over-IP class.
Hardware solutions also have significant limitations. They work with the signal from a single physical video output, without the ability to "crop" a portion of the image or combine multiple monitors into a single stream. Software technologies do not have such limitations at all - it is possible to capture any part of the desktop or all screens at once. Additionally, it is possible to work with virtual video objects, such as virtual PCs or windows of individual applications, even if they are obscured by other windows.
A similar situation arises with scaling the number of workstations: it is very easy to connect another user, for example, with a laptop, to the video wall if the software approach is followed - just install the software streamer on the laptop. When choosing a hardware approach, additional equipment units will need to be purchased, installed, and configured. For example, in the case of a laptop, an additional hardware streamer, a port for its network connection, and several cables will be required.
Capturing images from cameras connected to the operator's PC using the software approach is done directly on the workstation, without the need for any additional equipment. Additional encoding devices will be required for streaming from cameras using hardware methods. USB cameras need to be first connected to the PC and then one of its video outputs should be connected to the encoding device. HDMI cameras can be directly connected to it. IP cameras equipped with a network port can be connected directly to the network.
It is important to note that user interfaces of hardware solutions are designed to be understood and convenient for system engineers, and end-user operators cannot interfere with the settings and operation of pre-configured hardware solutions. On the other hand, user interfaces of software solutions are designed with end-users in mind, providing them with a wide range of possibilities. Operators can, for example, adjust the quality settings of the transmitted video on-the-fly, such as reducing the video traffic or increasing the resolution and image clarity. In the case of hardware implementations, operators would need to involve a trained AV engineer to make such changes.
The same applies to audio accompaniment: when using software solutions, it can be transmitted along with the video stream or not transmitted at all, and users can change these settings themselves on-the-fly. In hardware-based implementations, the intervention of an AV engineer would be required, and often the selection of other, more expensive end nodes.
Another capability that can be easily achieved using a software approach is disabling the mouse cursor. When displaying images from multiple workstations on a video wall, having multiple cursors on a large screen can confuse control room personnel. Software video solutions allow removing unnecessary cursors. Such an option is not available in hardware solutions.