In this new edition of Industry Voices, we speak with Erling Hedkvist about his professional journey, the context behind the creation of manifold, and key market trends such as the transition to IP, the expansion of broadcast-grade infrastructure into enterprise environments, and the key topics shaping industry conversations this year

*This content is included in the April issue of TM BROADCAST. You can access the full magazine here.

The transition towards IP and software-defined infrastructures has established itself as one of the industry’s main drivers of innovation. However, as important as technological development is, its ability to adapt to real market needs is equally critical. This was one of the central themes of our conversation with Erling Hedkvist, Broadcast Technology Sales & Business Development at both arkona and manifold technologies, and the latest guest in our Industry Voices section.

Thanks to his professional trajectory and position within the industry, Hedkvist offers a valuable perspective on both the evolution of the market over the past decades and the context in which initiatives such as manifold have emerged. The organisation was born from a shared strategic reflection: “If we were starting from scratch today, what would we do differently?”

A brief overview of Hedkvist’s career helps to put the origins of manifold into context. Beyond the purely technical dimension, he quickly identified his core strength: “I realised early on that my strength was bridging the gap between technology and people, understanding requirements and translating them into something practical and meaningful.”

He began his career at Net Insight in 2000, at a time of considerable uncertainty. “At the time, the company had no plans to move into the broadcast industry at all—they were aiming to compete with Cisco. I joined in February 2000, and within six months the dot-com bubble burst.”

The impact was immediate: “We went from hiring three people a week—at one point people were even sitting in hallways because there weren’t enough offices—to laying off two-thirds of the company within another six months. At that point, the company had to focus on the only segment that was still buying: broadcasters. The IT sector had essentially stopped investing, but broadcasters continued to purchase solutions.”

After more than a decade at Net Insight, Hedkvist moved to Evertz, where he deepened his understanding of the sector: “That was where I deepened my understanding of the broadcast industry. I had been hired because of my background in video transport, and also my early experience with IP.”

He later joined Lawo at a time when the company was looking to expand its video product portfolio in collaboration with arkona. “Lawo had an OEM agreement with arkona technologies. Arkona was responsible for developing the video products, while Lawo marketed them under its own brand. My role was essentially to define the products—working closely with arkona, explaining what we wanted to build—while they developed them and Lawo brought them to market.” This stage laid the foundations for a relationship that would later shape his professional path.

In 2020, a change in Lawo’s leadership marked a turning point: “Philipp Lawo had an accident and was no longer with the company, and the new leadership decided to move away from the OEM model with arkona and develop products in-house.”

After a period working independently, Hedkvist returned to the arkona ecosystem—this time directly: “The idea was to take the products that had previously been OEM and bring them to market under arkona’s own brand.” From there, manifold was born.

“Where we see the convergence between ProAV and Tier 1 broadcast is in large-scale ProAV systems. Major enterprise organisations that want to build high-end internal AV infrastructures”

Manifold: three pillars

In Hedkvist’s words, manifold represents “a rethinking of the software-defined infrastructure concept”, built around three core pillars.

The first is its software-only nature. “Unlike all the companies I had worked for previously, which developed both hardware and software, manifold does not manufacture hardware at all.”

The second pillar is virtualisation. “It’s not just software running on standard hardware—it’s a fully virtualised architecture where hardware is abstracted from the user.”

In this model, users define their needs while the system dynamically allocates the required resources. This approach breaks away from the logic of fixed systems: “It’s no longer about a fixed system—like a PC running Windows and then running applications on top. It’s about a flexible, abstracted infrastructure where resources are dynamically orchestrated.”

“Unlike all the companies I had worked for previously, which developed both hardware and software, manifold does not manufacture hardware at all”

The third pillar is based on the use of standard hardware combined with dedicated acceleration—particularly through FPGA and high-bandwidth memory. This enables a shared-memory architecture that redefines traditional processing workflows. “What this means is that, instead of building traditional broadcast pipelines—where you have inputs going through sequential stages like colour correction, frame synchronisation, delay, audio embedding and so on—everything is written into memory first. You ingest IP streams, place everything into memory, and because the memory has very high bandwidth, it can handle all incoming data in real time. Processing then happens directly on that memory.”

This approach delivers significant efficiency gains: “Instead of having, for example, eight separate colour correctors, you have a single colour correction function that operates at a much higher speed than the media itself. If your media is running at 3 or 12 gigabits per second, that processing function might run at 400 gigabits. This means you can process multiple streams almost instantaneously.”

“That’s the core of the manifold approach,” Hedkvist concludes, aligned with emerging architectures such as Media Exchange Layer (MXL), which is also based on a shared-memory architecture.

From your perspective, how do arkona and manifold complement each other?

They complement each other well, although they address different stages of the customer’s transition. Arkona Technologies follows a more traditional architecture than manifold technologies. Arkona still builds both hardware and software. Even if the platform is software-defined, we manufacture physical units—installed on-site—with long lifecycles and warranties of up to ten years. In that sense, it remains a traditional approach. With manifold, by contrast, there is no hardware. Customers use off-the-shelf infrastructure, so the conversation starts from a completely different point.

They also address different needs. Even though the industry is moving towards native IP, more than 50% of current requirements still involve SDI. The only practical way to support SDI is to develop dedicated hardware, as there are no truly cost-effective off-the-shelf SDI-to-IP solutions. That’s why arkona continues to play a key role. It provides interfaces for SDI and other legacy signals—Word Clock, Blackburst, tri-level sync—everything traditional broadcast infrastructures still rely on. Manifold, on the other hand, is entirely IT-based and IP-native. Any signal involving SDI, MADI, AES or similar formats is handled by arkona, converting it into IP and back again when needed.

Arkona’s platform is also very mature. Development began around 2015, with the first generation launched in 2017 and the second in 2022. It has proven itself in demanding environments such as the Olympic Games and FIFA World Cups, as well as in continuous operation with broadcasters worldwide.

Manifold, by contrast, is much newer and focused on high-density IP processing. If your content is already in IP and needs processing—conversion, monitoring, timing adjustment, mixing or compression—then manifold is the right tool. This is largely due to the use of off-the-shelf acceleration cards, similar to GPUs, installed in servers, which are well suited to intensive media processing tasks.

In practice, the solutions we deliver typically combine both arkona and manifold. We focus on what we call core infrastructure—the elements that tie everything together within a broadcast facility: routers, multiviewers, up/down/cross converters, colour correction and audio routing. We provide these functions in a more compact, modern and flexible way.

“Many decision-makers have built their careers over 20 or 30 years working with SDI. It’s what they know. So when it comes to investing in a new system, there can be hesitation”

What are the main strategic objectives for manifold in the coming years?

Manifold Technologies is a much newer product, so there are still many things that haven’t been developed yet. There’s a much longer to-do list. Compared with arkona technologies, which is more mature and focused on refinement, manifold is still expanding its capabilities. We have a long list of customer requests that we haven’t yet had time to address.

Over the next few years—particularly over the next five—the focus will be on expanding functionality. There are many features we want to introduce that are not yet implemented. In parallel, we will continue refining the platform: improving the user interface, optimising workflows and increasing efficiency. These are areas where arkona is already in a fine-tuning phase, whereas with manifold we are still, in many cases, building the first version.

In essence, the main objective for manifold over the next five years is to significantly expand its feature set and overall functionality.

If you had to highlight just one additional functionality, what would it be?

I would point to something already on our roadmap for this year. Internally, we refer to these as “product services” in manifold, because that’s how customers purchase them. You don’t buy a product—you buy a service. It can be licensed, even as a perpetual licence, but it still operates as a service because it runs on hardware rather than being tied to a physical device. One of the services we are planning to introduce is called Manifold Bridge.

Manifold Bridge enables the conversion of uncompressed video into compressed formats for distribution—using protocols such as SRT, NDI or RIST—and also the reverse process. For example, you can take multiple 1080p streams, compress them using codecs such as MP4 or HEVC, and distribute them. Equally, you can take compressed streams and convert them back into uncompressed video. This actually works in both directions: from uncompressed to compressed, and back again.

This is the next major feature we want to introduce. So far, everything we do is uncompressed, which works well, but there is clear value in supporting compressed workflows—especially as modern computing platforms, and particularly GPUs, are highly efficient at compression and decompression. In practice, this will become our next product—although it is delivered as a service, which is how everything in manifold is structured.

“Even though the industry is moving towards native IP, more than 50% of current requirements still involve SDI”

Convergence between ProAV and Broadcast

Amid the ongoing convergence between broadcast and ProAV, we asked Hedkvist whether his customer profile has changed—and how the company is navigating this shift. His answer is clear: their focus remains firmly on Tier 1 broadcast.

“I always say that, as a company, you have to work with the capabilities you already have. I can say, ‘we want to build electric cars’ or ‘we want to build space rockets’, but that only makes sense if you actually have the people who know how to do it. And if you look at it the other way around, we don’t have venture capital backing to suddenly go out and hire rocket scientists. In essence, the real question is: what are the people in the company good at? At both arkona technologies and manifold technologies, our teams are very strong in Tier 1 broadcast. That’s our core expertise. So naturally, the most straightforward thing for us is to build Tier 1 broadcast solutions.”

From this foundation, however, a new opportunity space is emerging. As Hedkvist explains, the convergence between ProAV and broadcast becomes particularly visible in large-scale corporate environments.

“Where we see the convergence between ProAV and Tier 1 broadcast is in large-scale ProAV systems. For example, major enterprise organisations—companies like Apple or Bloomberg—that want to build high-end internal AV infrastructures. They’re not looking for something basic. They want professional distribution, cloud integration, internal access for employees, and often a full studio setup with cameras and lighting. It’s not just someone filming on an iPhone—it’s something designed for executive-level communication. We’ve worked on a number of these types of projects, and that’s where we see a strong fit.”

Transition to IP

What do you see as the main barriers to full adoption of IP today?

I would say this has changed over time. A few years ago, I would have said the main barrier was cost. Today, I don’t think that’s the case anymore. Now, I would say the biggest barrier is—honestly—fear. There is a perception among many buyers that IP is complex. And to be fair, that perception isn’t entirely unfounded. There is a learning curve. But I do think the fear is somewhat misplaced. In reality, once you understand it, IP is not necessarily more difficult—if anything, it can be simpler. Like anything new, it requires some learning, but the curve is not as steep as many people assume.

What we see is that many broadcasters are very comfortable with what they already know. Moving to something new can feel risky. There is also, to some extent, a generational factor. Many decision-makers have built their careers over 20 or 30 years working with SDI. It’s what they know. So, when it comes to investing in a new system, there can be hesitation. If you’re not familiar with IT environments, you may feel less confident adopting an IP-based approach.

In practice, this is reflected in most projects today. In around 80% of cases, customers are still comparing SDI and IP when planning a new facility or upgrade. The question always comes up: should we go with SDI or IP? Even when IP is clearly an option, there is often a tendency to lean towards SDI—while recognising it may not be the most future-proof choice. It’s a bit like buying a car you expect to keep for 20 years. Do you go for a combustion engine, or invest in something electric because it represents the future?

So there is still hesitation around IP, and it requires some effort on our side to guide customers. But once they make the transition, things tend to fall into place. To support this, we have developed a dedicated solution combining arkona technologies and manifold technologies, which we call ECIP, designed to make the transition to IP as straightforward as possible.

“In around 80% of cases, customers are still comparing SDI and IP when planning a new facility or upgrade. The question always comes up: should we go with SDI or IP? Even when IP is clearly an option, there is often a tendency to lean towards SDI—while recognising it may not be the most future-proof choice”

What do you think will be the main topic of debate in the industry this year, from your area of activity?

From a forward-looking perspective, many people are already talking about what comes next—namely MXL. That will certainly be a hot topic, but in reality, most broadcasters are still focused on completing the transition to IP. In many ways, it is the next step beyond ST 2110, while many organisations haven’t fully made that transition yet. So yes, there will be conversations around MXL and similar concepts, particularly among more advanced or specialised segments of the industry. And that’s important—it’s something we should be working towards.

But in practical terms, I believe the real discussions inside broadcasters are still centred on SDI versus IP. When organisations plan new investments, there is often a clear intention to move towards IP—but also hesitation. There are concerns about having the right people or the necessary expertise, and hiring new talent is not always an option.

That means existing teams need to be trained, which can feel like a significant challenge. So, while the industry is talking about future architectures, many organisations are still trying to figure out how to take that first step. Management may say, “we want to modernise,” but from an operational perspective, the response can be: “we don’t fully understand this yet.” In that situation, the default is often to fall back on what is already known. Therefore I think that tension between ambition and practical readiness is where most of the real conversations will happen this year.

“I believe the real discussions inside broadcasters are still centred on SDI versus IP”

Would you like to add anything else to be included in the interview?

I would just add that this is a very small industry, and that’s something I really like. Over time, you build relationships that last for decades. You end up working with the same people throughout your career—the business cards change, but the people remain. That creates a strong sense of familiarity and trust, and from that perspective, it’s a very enjoyable industry to be part of.

At the same time, because it’s such a specialised field, much of what we do is consultative. We often guide our customers. It’s very different from buying a consumer product. If I want to buy a monitor, I might watch a few YouTube videos or read some Reddit posts. No one from Apple is going to call me and ask what I need. In our industry, it’s the opposite. We have direct conversations with customers, and they rely on vendors like us to help them decide what the right approach is.

It’s a small industry—you work with the same people over decades. That builds trust, and that trust really matters.

Ultimately, our role is to guide customers through change. If you get that wrong, they won’t come back. If you get it right, you build relationships that last a career.

Our hope for this year and the next few is to help guide customers towards IP-based solutions and reduce the fear around them. There’s no reason to be afraid—this is the direction the industry is moving in, and it will enable us to do better things.

“It’s a small industry—you work with the same people over decades. That builds trust, and that trust really matters. Ultimately, our role is to guide customers through change. If you get that wrong, they won’t come back. If you get it right, you build relationships that last a career”