Real world AV over IP problems and solutions

Real world AV over IP problems and solutions
AV over IP is nothing new, yet problems are still occurring, this time it’s around the system rather than the core tech itself. Paul Milligan gathered an expert panel to find solutions to problems.

AV over IP has now reached the stage where it can be classed as a mature technology and has been around long enough that it’s widely trusted across the entire pro-AV spectrum. That isn’t to say failures don’t happen, of course they do, and integrators would have a lot more time on their hands if they didn’t. But because the technology has matured, these failures now happen around the system rather than from the core technology itself, in the form of network configuration, switch selection, firmware management, EDID/HDCP issues, security policies, poor documentation, and an unclear ownership structure between AV and IT teams.

We gathered one of our biggest panel of experts ever to get boots-on-the-ground insight and advice on how to prevent issues occurring, or how to best fix them when they do occur. We began by asking them where problems most commonly arise in deployments, is it in the core tech, the infrastructure, or the way projects are designed and managed? The core technology is genuinely mature at this point states Richard Jonker, VP of commercial business development, Netgear. “Protocols like Dante, NDI, and SMPTE ST 2110 are proven, well-documented, and widely deployed. When something goes wrong today, it’s rarely the protocol that’s at fault.”

AV over IP is sometimes treated like a traditional AV extension system, but it is fundamentally a networked solution says Zec Voislav, product development director, Wyrestorm. “That means it needs proper planning between AV consultants, integrators, IT teams, and the end client from the beginning,” The core technology is no longer the primary bottleneck says Jatin Vasudeo, project engineer at systems integrator Granteq. “Codecs and protocols are stable, and the hardware generally does exactly what it says on the spec sheet.”

The vast majority of problems arise at the intersection of infrastructure and project management. Issues usually stem from a failure to align the AV design with the reality of the client’s network infrastructure early on. When project management fails to bridge the gap between AV integrators and the client’s IT stakeholders during the design phase, serious infrastructure bottlenecks can emerge during deployment. These can include insufficient PoE budgets, switches that cannot handle multicast traffic, or distance constraints between endpoints and IDFs (eg a rack room). In many projects, these issues arise because AV is still treated as an afterthought.

One key issue is the lack of conformity, do security policies and network config vary greatly from client to client? It varies massively says Jon Pain, senior engineering manager - network integration in AV/IT for systems integrator proAV. “Some clients are almost unrestricted and will allow whatever onto their network. But where I think it should be, and where many clients are now, is that every device needs to authenticate onto the network, and the client needs to know what each device is doing and where it’s going.

It’s not just about authenticating the right devices; it’s also about ensuring firewalls are aware of each source, destination, and port, so only allowed traffic can traverse. While that does create difficulties in deployment, it is where we should be as an industry given the current state of IT security.” Moving from an education or innovation centre to a heavily regulated corporate or financial environment is like stepping into a different world says Vasudeo. “Some clients offer relatively flat, flexible networks, while others enforce strict port security, MAC address filtering, and rigid firewall rules separating VLANs. You cannot take a one-shoe-fits-all approach to AV over IP.”

We have covered the thorny issue of firmware before (Inavate April 2025) but just how often are network problems caused by firmware updates? And how can you mitigate them? The key is to treat firmware management as part of a broader operational strategy suggests Kyle Barron, product manager, networked AV, Harman Pro. “Rather than updating devices immediately when new firmware becomes available, organisations should validate compatibility, review release notes, and test updates in a controlled environment whenever possible.”

Those all-too-regular Windows updates are still causing problems years down the line, as this example highlights: “The most widespread problem is Windows updates that have been tested on the ‘site standard’ desktop (as defined by IT department) and then deployed network-wide including to other Windows-based AV appliances that might have dedicated, function-specific configurations which don’t match the ‘standard desktop’. So, the update often has unexpected effects on these operational devices,” explains Rob Moodey, manager strategic partnerships, Matrox Video.

Let’s turn our attention to network switches next, what switch-related mistakes do our panel see most often in projects? “Switch configuration challenges, such as implementing IGMP Fast Leave, can create issues in AVoIP performance if not addressed early on,” says Rainer Stiehl, VP marketing Europe, Extron. A proactive, collaborative approach between switch manufacturers and support teams helps resolve these issues more efficiently he adds: “For example, the Extron Network Technologies Support Group works with switch vendors to verify multicast implementations. This team validates and then publishes a list of network devices based upon network standards.”

 

image: Extron

The central challenge stems from the breadth of switch manufacturers in the market says Andrew Ludke, senior director, product management, hardware solutions, Crestron. “When an AV integrator has the latitude to select networking hardware they have genuine familiarity and expertise with, they retain control of the deployment and are well-positioned for success. In many projects, however, the end customer has already standardised on a particular network infrastructure, and that is where close coordination with the client’s IT team becomes indispensable.”

“The fundamental question the integrator must answer is this: do you have control over the network architecture, or are you required to operate within the customer’s existing infrastructure and switch preferences? If the latter, how effectively are you coordinating within those constraints?” Under specification is the most common problem says Netgear’s Jonker. “Integrators will price a switch based on port count and data rate, overlooking switching capacity, buffer depth, or whether the hardware can actually handle the multicast load they’re asking of it.”

“Beyond that, IGMP snooping misconfiguration is a persistent issue. Getting multicast to behave correctly requires properly configured queriers and leave group timing, and those details often get missed. QoS is another one: AV traffic needs to be prioritised correctly, or it will compete with data traffic and lose.” Next, we asked our panel whether integrators and clients underestimate the switch in AV-over-IP systems? What should they be asking?

We still see situations where both integrators and end users underestimate the importance of the network switch in an AV-over-IP system says Harman Pro’s Barron. “There can be a perception that any standard gigabit switch will suffice, or that networking is solely an IT responsibility. In reality, the switch is a critical system component that directly impacts performance, reliability, scalability, and the overall user experience.”

Like Moodey above, Voislav feels too many buyers look at the number of ports first, but the better questions are around architecture and performance he explains: “They should be asking: what codec or transport technology is being used? What bandwidth does each stream require? Is the system multicast or unicast? What is the required backplane capacity? Are the uplinks sufficient? Is IGMP snooping properly supported and configured? Will AV share the client network or sit on a dedicated AV network? Who owns the switch configuration after handover? Is the switch approved by both the AV manufacturer and the client’s IT team?”

Some integrators interpret the plug-and-play nature of laptops on networks as evidence that all network devices are commodities, i.e. that all switches are the same, says Moodey. “Nothing is further from reality. Detailed examination of the datasheet is required unless you know that it does what you need. If the switch is part of a network rather than being the network, the uplink bandwidth is important. Knowing that a switch has, say, 1Gbps non-blocking bandwidth only tells you what the switch can handle between its own ports.”

“If you are connecting to another switch, you must consider that link. If you don’t, it will become a choke point.” Another common misconception is that the lowest-cost switch is the best value adds Barron. “In practice, choosing the right switch means considering more than just specifications and upfront cost. Factors such as vendor support, warranty coverage, reliability, product maturity, and a proven track record in AV-over-IP deployments are all important considerations. The cost of troubleshooting and downtime can quickly outweigh any initial hardware savings.”

Next, we asked our panel if they were seeing issues with EDID and, if so, what was the best way to deal with them. Yes, EDID issues are still occurring says Vincent Pang, product manager, Pixelhue, particularly in systems with multiple displays of different resolutions or capabilities. “The best approach is to manage EDID proactively rather than relying entirely on automatic negotiation.”

“Automatic EDID exchange works well in many simple installations, but as systems become larger and more complex, with mixed display resolutions, LED walls, switchers, extenders, and converters, manual control becomes increasingly important.” EDID has long been an industry challenge says Extron’s Stiehl, which has seen his company develop tools to help designers and integrators solve the source and display compatibility problems it can cause.

Extron’s EDID Manager is a software tool that puts users in direct control of EDID functionality. Users may adjust video rates, colour space, chroma sampling, colour bit depth, audio support, and overall video format across systems based on Extron products. EDID Minder automates the EDID process by managing EDID communication between devices to ensure sources power up properly and deliver reliable, consistent content to every display.

Another problem that can arise with AV-over-IP deployments is with HDCP, how often did our panel encounter problems? And could they offer any tips or guidance? HDCP compliance issues are largely avoidable when the system is designed with HDCP in mind from the outset suggests Crestron’s Ludke. “The complications arise when integrating technologies that were never engineered to support it.” Crestron’s DM NVX AV-over-IP platform, built around HDMI and USB-C interfaces, was architected from the ground up with full HDCP support, to eliminate these issues he adds.

HDCP has, since its inception, plagued the integrator trying to get systems to work– and anyone else who ‘tweaked’ working systems says Matrox’s Moodey. “The ‘not enough encryption keys’ problem largely went away for hardwired systems as the design of devices evolved. When AV over IP started, the problem was back with a vengeance, the number of potential connections on the network overwhelmed the number of keys.”

“A different approach was devised with the result that individual brands ‘did their own thing’, requiring the same brand at both ends, not something that HDCP was ever intended to cause.” And this is where IPMX comes in he adds, “IPMX addressed this by creating HKEP (the HDCP Key Exchange Protocol) which everyone could adopt because it was an open standard. There are many AV environments where HDCP is not needed, so IPMX certification does not require all devices to implement HDCP.”

“However, if HDCP is implemented, it must be done the IPMX way, by supporting HKEP. That is a big implementation headache removed.” Next, we asked our panel about network documentation: if it is not completed correctly, or in some cases doesn’t exist at all, does it create significant problems? It’s the single biggest cause of prolonged downtime says Granteq’s Vasudeo. “Without comprehensive IP and MAC address schedules, clear port mapping, and accurate topology diagrams, a minor troubleshooting task turns into a forensic investigation.”

Paul Tuthill, manager of network integration for AV/IT at proAV finds it can considerably eat into the schedule on lots of projects. “It’s fairly standard for clients not to issue documentation proactively, so we end up spending considerable time workshopping with the client to extract the necessary information. That actually requires us to have our own documentation prepared before those conversations begin, so we can lead them, outlining things like inter-VLAN routing needs, VLAN setup, and the firewall rules that will be required.”

“And that’s complicated by the fact that AV products aren’t yet standardised, so some of that information is learned in the field rather than coming from the manufacturers themselves.” Finally, we asked our panel if a lack of clear demarcation of tasks between AV and IT teams was still a challenge? How can you stop it becoming a ‘blame game’ between the two disciplines?

“If an AV integrator just says, ‘it works on our side’ and throws it back, the other party is left doing all the painful diagnostic work, which damages relationships and gives everyone a bad impression of how AV integration is managed,” says Tuthill. When a screen goes black, AV blames the network, IT blames AV and the site operations team are caught in the middle says Vasudeo. “You stop the blame game by establishing a scope matrix before a single cable is pulled.”

“Hold joint kick-off meetings where both teams agree on exactly where AV’s responsibility ends and IT begins: e.g. AV owns the endpoint devices and their configuration; IT owns the switch port, VLAN routing, and PoE delivery.” This issue is still a challenge says Pain, which is why systems integrator proAV has its own network integration team specifically to bridge that gap.

“You need people who understand both AV and IT, and you need genuine collaboration between the two teams, not just on initial configuration but throughout troubleshooting as well. From the AV side, we need to give clients a clear picture of what we’re seeing and what our requirements are, not just sending documents over for them to dig through but actually opening up the conversation. From the IT side, they need to understand how AV is working and be open about their network configurations so we can all get to the bottom of what’s happening.”

Final word goes to Matrox’s Moodey, with a warning to all: “AV is finding itself to be a specialist branch of the infrastructure function, just as desktop IT is. And that infrastructure is also being used by security, telecoms and other historically separate systems. Networking is thus fast becoming ‘a core competence’, not ‘a technology’. AV folks have to recognise that too, those who cling to the signal flows of the hardwired, and circuit-switched era will increasingly become marginalised and under-employed.”

images
panumas nikhomkhai/shutterstock.com
DC Studio/shutterstock.com
Gorodenkoff/shutterstock.com

Article Categories