AV systems for video production, streaming, live events, and security are loaded with computers, and control rooms are loaded with keyboards, monitors, and mouse(s). KVM switches allow one or more users to access multiple computers, saving desk space and increasing convenience. In fact, they are so useful I would put one in just about any system with multiple computers!
Most of my KVM installs are small (no more than five computers and five users) and contained in a limited area, so I have stuck with direct-connect KVM matrix switches for their simplicity and performance vs. cost. I used the Adder DDX series for several systems and liked it enough to publish a review in 2023 (https://www.svconline.com/author/eric-wenocur). Unfortunately, the DDX was discontinued, and Adder’s only options now are IP-based. There are still some direct-connect (over CAT cable) switches around, but it’s getting pretty hard to find a sophisticated version that supports 4×4 or larger.
The DDX series consisted of a matrix chassis, small Computer Access Modules that lived right at the computers, and user stations that could mount or sit on the desktop. By contrast, the ADDERLink XDIP can be configured as either a Transmitter (computer) or Receiver (user) node, and the dedicated matrix is replaced by a simple Layer 2 network switch. XDIPs can also be used in pairs as KVM extenders–though rather expensive ones.
The XDIP is a 1/2-width 1RU box that can be rack-mounted, shelved, etc. A bit of space is required near each computer, but the dual rack-mount tray is actually rather nice, with thumbscrews to hold the devices. (Other Adder KVMs still use 0RU CAM modules at the computer.) The rear panel is well-labeled so connecting computers and kybd/video/mouse is straightforward. XDIPs can be powered via POE and purchased without external power supplies.
One aspect of IP-based KVM that I was unhappily anticipating was having to connect each unit and assign IP addresses (or discover what address they got via DHCP) before being able to use anything. I was pleased to find that the XDIPs will discover and coordinate themselves, using either DHCP or self-assigned addresses, and are immediately ready for use!
On first power-up the front power light is red, which indicates that the unit has not yet been configured. This threw me at first until I checked the manual. Once configured the power light will be green for a Transmitter node, blue for a Receiver. The front panel Link, USB and VID lights are also very useful for confirming proper operation or troubleshooting.
XDIPs start with the On-Screen Display asking if it is to be a Transmitter or a Receiver, so user stations will likely be set up first. XDIPs can be configured as Transmitter nodes by directly connecting a kybd, mouse and monitor to the unit, or via the Admin menu of any Receiver. As units are connected they are added to the system and will be seen when screens are refreshed.
XDIP settings can be customized in many ways, including name, static IP address, hot-keys, device list order, passwords, access restrictions, etc., depending on whether the node is a Transmitter or Receiver. On the whole, the GUI is sensible and functional, and the manual is quite helpful. The hardest part about configuring XDIPs is remembering which functions are hidden under which little gear icon!
Interestingly, each module not only remembers its settings, but also how it was last connected in the matrix. If a Receiver is moved to another location, it will start up connected to the same computer it was last controlling. This came in handy when I was trying to set up a graphics tablet connected to a specific computer (via a USB extender) and was moving a single Receiver unit between two different locations. Since the receiver was providing video for the tablet, it would immediately show the right computer, without needing a keyboard or mouse to select it.
Another nice function of the XDIP is the ability to put a “local” computer at a user position and have it appear in the same list as the computers out on the matrix (the local computer is only visible at the Receiver node where it’s connected). Conversely, you can connect a kybd, mouse, and monitor to the ports on a Transmitter node to control its computer locally, while still being on the matrix.
Using the matrix is as easy as pressing CTRL+ALT+C (the default shortcut) at any user station which brings up the On-Screen Display. Use the mouse or arrow keys to move up and down the list of computers and select one to control. There are other configurable ways to bring up the OSD and select computers as well.
Connections are HDMI for video (up to 1920×1200), USB A and B for keyboard, mouse, and other controllers, and analog audio via the usual mini-TRS jacks. Note that the XDIP only emulates HIDs (human interface devices), so other USB devices, such as memory sticks, or the graphics tablet mentioned earlier, will not work via the XDIP matrix. The fancier products can do full USB 2.0 emulation.
As for the network switch, I most recently used an inexpensive NETGEAR 24-port “smart” switch with POE, and it worked perfectly. I did not even look at the switch GUI, the defaults were fine. I like the idea of a dedicated switch for KVM, but the XDIPs can also connect to a switch that is used for other purposes. The KVM data will pass between switches as well (after accounting for specific switch configurations).
My experience with Adder KVM products goes back many generations, and I have always been happy with them. Support, when rarely needed, has been very good. The XDIP system worked perfectly out of the box in multiple installs. No support required.
