Voices of Video
Explore the inner workings of video technology with Voices of Video: Inside the Tech. This podcast gathers industry experts and innovators to examine every facet of video technology, from decoding and encoding processes to the latest advancements in hardware versus software processing and codecs. Alongside these technical insights, we dive into practical techniques, emerging trends, and industry-shaping facts that define the future of video.
Ideal for engineers, developers, and tech enthusiasts, each episode offers hands-on advice and the in-depth knowledge you need to excel in today’s fast-evolving video landscape. Join us to master the tools, technologies, and trends driving the future of digital video.
Voices of Video
Cloud Meets Hardware: How Net Insight's Nimbra Edge Revolutionizes Video Routing
The broadcast industry stands at a technological crossroads, with traditional satellite-based workflows gradually giving way to IP-native cloud solutions. In this illuminating conversation, Net Insight's Adam Nilsson and Pierre Le Fevre join Mark Donnigan to discuss how their Nimbra Edge platform is helping bridge this divide while dramatically reducing costs.
As a Swedish company with 25 years in video transport, Net Insight brings a unique perspective on the broadcast transformation journey. Adam explains how Nimbra Edge essentially functions as "an SDI destination through the cloud with an IP native focus," allowing broadcasters to maintain existing workflows while gradually modernizing their infrastructure. This approach proves critical for organizations heavily invested in legacy systems who can't afford disruptive wholesale changes.
The fascinating concept of "The Internet Snake - Transcoding Anywhere" reveals how modern video technology can function less like traditional transcoding and more like universal adapters. This approach connects disparate systems - enabling modern HEVC cameras to seamlessly interface with H.264-only devices through intelligent conversion points positioned anywhere in the network.
The most compelling revelation comes from Net Insight's partnership with NETINT Technologies. By implementing purpose-built Video Processing Units rather than general-purpose GPUs, they've achieved approximately 67% cost reduction for cloud transcoding over three years. As Adam notes, "When we did the math compared to current cloud transcoding solutions, this is about a third of the cost." Yet these savings come alongside expanded capabilities - frame rate conversion, overlays, automated shopping, and ad insertion functionalities in a streamlined, monitored environment.
We're witnessing a fundamental shift in how broadcast technology operates. While traditional broadcasters cautiously navigate organizational silos between satellite and IP teams, newcomers like FanDuel approach content delivery from cloud-native positions, automating workflows for their 400,000 annual events with consistent output specifications regardless of input variations.
Visit Net Insight and NETINT at NAB to discover how their combined solution creates a powerful toolbox for implementing hybrid distribution models with superior service quality and operational efficiency. The future of broadcast is here - more flexible, more capable, and significantly more cost-effective than ever before.
Stay tuned for more in-depth insights on video technology, trends, and practical applications. Subscribe to Voices of Video: Inside the Tech for exclusive, hands-on knowledge from the experts. For more resources, visit Voices of Video.
Voices of Video. Voices of Video. Voices of Video.
Speaker 2:Voices of Video.
Speaker 1:Welcome to another super exciting episode of Voices of Video. Now we have been doing a series with our partners that we're going to be featuring at NAB, and today I am super excited to be with NetInsight. So I want to introduce Pierre and Adam from NetInsight. Welcome guys. Thank you, mark. Thank you, yeah, it's great to have you. You know, we our two companies have, I guess, known each other for a little while. Of course, who doesn't know NetInsight? But you know, we've known each other and you know, now we've integrated together and we'll get to talking about that later. But why don't you first just introduce yourself individually? What do you each do for the company and why are we talking?
Speaker 2:Yeah, so my name is Adam Nielsen. I work as a product manager for NetInsight, working on our cloud products. I've been there for a couple of years now and looking to evolve our product and the business we do.
Speaker 3:Yeah, and I'm Pierre. I'm a software developer for NetInsight and I've been working on the transcoding features of Nimble Edge, mainly so working with you guys at NetInsight.
Speaker 1:Yeah, it's great. Yeah, you guys have done some really good work. Now, for those who don't know NetInsight, because you know it's possible, our audience, you know, is involved in a lot of different. You know industries and markets and applications, you know. Adam, do you want to give a little quick overview of the company?
Speaker 2:Sure. So yeah, netinsight is a Swedish company based in Stockholm. It's turned 25 years, like a year ago, I think. So we've been roughly in the industry for that long, working mostly on transporting solutions, so offering world-class and industry-leading technologies for high-quality content delivery. So very much focused on the tier one events sending content from the venue to different distribution places mainly on the hardware side. But then in the recent five, 10 years we've been going more and more towards the software side and focusing more on IP and solutions like that. In the recent five, 10 years we've been going more and more towards the software side and focusing more on IP and solutions like that. And we're, you know, a global company with customers all over the world. So it's been a good, you know, 25 years for us.
Speaker 1:Yeah, amazing. Well, you know, I know, as we were sketching out what we wanted to cover and talk about the, you know, transformation of broadcast is definitely something that I'm sure we'll touch on in this interview. But, yeah, so just real quick for the listeners, we are going to be at NAB, obviously, that is NetInt. Netinsight is also going to be at nab. Now, if you're listening to this after nab, well, and you didn't get to see either of our companies, you know, contact one or both of us and we'll be happy to give you an overview of what you missed. But if you're watching this before nab, you'll want to be sure and visit NetInsight at the NetInsight booth, obviously, netint.
Speaker 1:We also have a booth and it's going to be a great opportunity to see these solutions in person. So you know where I'd like to start is to talk about your product, nimbra Edge, is to talk about your product, nimbra Edge, and I think it might be helpful to give a quick overview of what Nimbra Edge is, you know. So, describe the product, talk about you know how it works and what it's designed for, and then that might be a nice segue to then discuss the integration of VPUs as we move towards talking about how broadcast is just fundamentally being transformed right now from an architectural and a technical perspective.
Speaker 2:But what's Nimbra Edge? Yeah, nimbra Edge is, you know, it's a very much simplified, it's an SDI destination through the cloud with an IP native focus, and that was sort of a big thing for NetInsight, because NetInsight had used different ways of transporting and mostly been focused on hardware solutions. So this was a software solution trying to do similar things, as our hardware products are mostly focused on the production and contribution side, but, you know, using and leveraging the arq protocols we have. So basically, it's a software you put on a server that makes it possible to transport, yeah, your content from point a to point b, c, d, e, well, as many points as you want, and, yeah, that is basically what we'll be doing with this end-to-end monitoring as well interesting, you know?
Speaker 1:um, I don't think I want to take a little detour um, why did you spin off sai? And I don't think everyone knows aboutsy uh and the fact that you know that's pretty important to Amazon and their live uh sports and you know all their live streaming efforts. But, yeah, why did you spin off Psy? Why'd you sell it?
Speaker 2:well, I guess money is always involved. It's it's before me and Pierre joined the company. But yeah, it was a product more focused on OTT and I think NetInsight understood that our strength maybe isn't on the end consumer, transporting things to the end consumer, but rather before that, so fitting into the company strategy Contribution more contribution than distribution. Exactly exactly company strategy, more contribution than distribution. Exactly exactly so you know. But it was a. It's a really cool product and it's nice seeing it being used at the absolutely highest level now yeah, exactly.
Speaker 1:Uh, I've heard, you know. I just came back from mile high video and um, traditionally I you would never really hear sai mentioned at all by video engineers at the conferences, and I'm just saying not that maybe nobody knew about it, but you just didn't hear much talk of the technology. I think no less than maybe two or three times. Somebody brought it up and I went oh okay, I've known and I know what Amazon's doing, you know. So, yeah, very interesting. Well, you know, pierre, you, along with Adam and, I believe, david Edwards and Jonathan Smith, authored a paper Really fascinating title. Do you remember the title I've got up in front of me?
Speaker 2:Ooh something with is it? Is it distributed some internet snake stuff or what?
Speaker 1:Internet snake yeah, exactly, yeah, yeah, yeah, exactly, yeah, but by the way, not a trick question. And and hey, just as an aside if, if, if anyone asks me like so, mark, I loved your last blog post. That article was awesome. What was the title? Again, I'd have a total blank stare. So anyway, yeah, network Distribution and the Internet Snake exclamation mark transcoding anywhere. Really fascinating. Maybe would you mind, pierre, sharing sort of like an abstract or just a high level of what is it that you were communicating in this paper.
Speaker 3:Yeah, for sure. It's sort of the concept of being able to connect anything to anything and make this sort of transformation that is required sometimes for the connection to happen, being able to do that anywhere, so you know, connecting different middle boxes that have different spec sheets, and all of this in the cloud or on-premise. So we sort of dive into that and discuss pricing and what sort of cards and hardware supports these features.
Speaker 1:Yeah, yeah for sure. Do you think and I believe this is what comes through even in the paper is the role of the transcoder transitioning, transforming to be? You know, it's interesting, you call Nimboreg a router. You know, is a transcoder even starting to become? Yes, it still needs to take a file in or a bitstream in. You know we're talking mostly about live, I think. So you know, it takes a bitstream in and then it, you know, gives you a higher rendition and a medium resolution and maybe a couple different bit rates, and so there's still that function, right. But do you think that the role of the transcoder is being rewritten?
Speaker 3:The transcoder as the file transcoder, I think, is still alive and well, but what we're doing is more like an adapter right where we can plug in your European plug to the US plug, right where you might have a super modern camera sending HEVC, but you have some middle box somewhere that can only take HE64, for example.
Speaker 1:Interesting. That can only take H.264, for example. Interesting and certainly, as you said. Adam made the comment that you work primarily in the broadcast domain, with broadcasters. There is a lot of legacy standards still out there. Hello, MPEG-2 and interlaced.
Speaker 2:For sure. Just to comment on that, that and that's what we're seeing. Right, we need to adapt to these legacy standards. But you know, you talk to customers or users and they're saying you know, we, we want to, we want to do this, uh, but they've so heavily invested into a previous technology so, in order for them to move to something new, that step might be very big. So, in order to find a way of making that transition a little bit smoother by, you know, using this concept of, like Pierre says, like with the adapter, it simplifies their way of moving to some, you know, towards something that's a little bit more modern and new, and they don't need to go full step and change everything that they did just to adjust it to the new things. They can take it one step at a time, and that's what we're sort of trying to solve with our approach to this.
Speaker 1:Yeah Well, I'd like to understand. The product even has Edge in the name, so Nimbra Edge, and yet in the paper you reference that the cloud is clearly a platform that a lot of broadcasters have migrated to or are in the middle of migrating to, or have been there. You know they've been there for years now. So help us understand. And for those listening saying wait, a second Ed sort of implies like a box that's sitting somewhere. Where does this architecturally go? Where does this system go and how does it connect, you know, to the common platforms that are in use today? Do you want to, do, you want to take this?
Speaker 3:pierre should I yeah, sure, uh, no, I mean, you're definitely right. Right, like the term edge you would expect like a box at an isp room, um, and and like edge can totally do that right. We have the concept of video regions, which means that you can create a region of appliances so whether that's boxes or servers, that's up to you and these can be in different geographical regions. So you might actually have one at a stadium, and then the cloud region, which is your routing region, and then maybe you have some ISP somewhere that wants MPEG-2 audio and H.262. And so you might do the transcoding at their sites at the edge. So it is edge, but it's also everything in between the edges.
Speaker 1:Interesting, yeah, interesting, yeah. So, adam, what are you saying? You're talking to customers. How quickly are broadcasters modernizing? And that's my term, but I think that's what you're inferring here is that there is a modernization effort that has to happen. It's always fascinating. Fascinating to me because, you know, we sort of have a foot in Well, there's more than two camps, but for this illustration, since I only have two feet, you know, in kind of the streaming world and increasingly a little more of the traditional, you know the broadcast world, and I always sort of chuckle a little bit inside when I'm talking to maybe someone's a little more traditional and they'll spend five minutes telling you how critical it is that they match Netflix quality, that Netflix is their competitor, and then they'll say, but we don't do OTT, we're not OTT, and and it's like, well, and the modernization that's happening is the standards now are homogenizing around three codecs, you know, which has been the case for a while.
Speaker 1:So you know, but H.264, hevc and now AV1 is, you know, coming out more strongly. I think it's lagging a little more in the true broadcast side, but for OTT, av1 is absolutely coming. You know, you have HLS or dash in terms of your if you're going to be distributing over the open internet. You know those are largely the, you know the formats that you're going to distribute the content in. And then, of course, we can go and look at everything else, right, and say, okay, yep, that's being used in modern broadcast, yep, that's what meta uses when they're streaming Facebook lives. You know, and that's what YouTube uses, and you know that's what Netflix uses. And so what are you seeing? You know, in terms of just the pace, you know I'm curious of, of willing to be disrupted and even disrupt themselves, you know, for your customers, um, are they?
Speaker 1:yeah, they're doing a good job. Are they running into this? I don't know if we're ready yet it's it's.
Speaker 2:It's interesting, I would say, like the traditional broadcasters is you know they're. They're more the last you know scenario. They're more the last scenario. They're more like we don't really know if we want to go into this. I don't think that's a bad approach per se. I think there are some valid reasons for not diving into the deep end, and especially because I think, if you don't understand how to leverage the new technology and the different pieces of it, it's super easy to go to AWS and buy everything from their marketplace and bolt these things together. Good, you will get a fine solution. Problem is you will get a lock-in like no one else. You're going to buy them and you're not going to get out, so you can do that. It's fine, you're going to be happy, but it's probably going to cost you a bunch of money in the end because you're going to get lock-in effects and they won't be able to get out and switch these pieces and stuff like that. So we have that, but we do see that broadcasters do want to leverage what's out there, but it's you know, going from satellite to this more cloud-based solutions, it takes time because they're used to doing things in certain ways. It might be even different teams that have set up the satellite links towards, compared to the ones that are now working on the more IP-native cloud type of solutions. So having those conversations is just not about talking to the people you used to talk to. It's about talking to the new ones that have joined the company and bridging them as well. And I think that is sort of the biggest challenge for us is that usually when we talk to people, we're just talking to a small, you know part of the company, but you need to bridge the gap in between. You know all these sites and that that's the difficult part. But then then you have I'm in Los Angeles right now for the video services forum and yesterday there was a company that got up that have been it's a new company FanDuel on the stage, betting and all that. And the way they do things is very different. They're basically cloud native from the start because they have no legacy to think about.
Speaker 2:And when they told me they do 400,000 events a year and they need to spin things up immediately. But what? What they told us and this sort of leads into what we're trying to solve is that you know they have these events that are coming from everywhere all over the world. So they're getting a bunch of inputs or a bunch of sources that you know might be from europe, asia and then us, and they have a certain frame rate. But when they send it out, they want it. You know they want to perfectly set up.
Speaker 2:So they get an input and then they basically want to automate a workflow that tells them we're receiving this, but we want to send it out. As you know, we're receiving this frame rate, we want to send it out as this frame rate and do these things and build that automation. And that's where we're seeing that the industry is going, that in the end, you just want to automate these workflows and have less hardware and just be able to spin things up and down quickly, because you know that's the way people are looking at things now. It's, it's not. We're seeing less and less engagement into watching full live events. They're watching highlights or clips, or you know it's, it's yeah, it's a huge difference, so yeah, it's.
Speaker 1:It's interesting, but it's difficult, it's a difficult thing to, and and I guess that's a perfect segue to how we started working together because you know so yeah, I mean I won't take the words out of your mouth what you know, why don't you? You know both, pierre. You know Adam you can give you know the journey here of how you discovered NetEnt, and then you know what we're doing for you and the role that's playing in being able to deliver a solution that does what you just described.
Speaker 2:Yeah, so I can give you a little bit of background. So me and a colleague based here in the US, we've been talking for years. You know what can we do in order to evolve our business and the way we do things. You know we have this platform, but cloud contribution is it's not a huge market and it it's. It solves certain problems, but it doesn't solve all of them. Far from them, from far from it. So you know, we started talking about how can we do things and we started looking at we probably want to go into more distribution types of workflows. So doing that, we I don't know how I find that in, but I remember walking over last year at nab to the booth, uh, meeting with you guys, and I saw someone had told me that your quality was really, really high. Uh, so I walked over and we started you know the conversations and and we put I well, I put the card in in Pierre's hands, like two weeks later and you know?
Speaker 2:yeah, pierre, it's just I gave it to you and said you know, fix it, or I want to solve this.
Speaker 1:I want to do these things, check it out yeah.
Speaker 2:Exactly. Basically, that's what happened. Yeah, Pierre took over the project, basically trying to drive it forward. Yeah.
Speaker 3:Yeah.
Speaker 1:And it was like christmas because I I'd been spending um I don't know how long getting uh, xilinx, u30 uh to work. So it's an interesting.
Speaker 3:You know, we've heard, we've heard about that card and we've heard sentiment, archaeology and uh, you know, building your own images lots of fun things that take weeks. It was great, honestly. We were still testing what cards we want to support and testing the workflow. It was really great to have a card that sort of worked.
Speaker 1:So I'm curious you had reached it sounds to me like you had definitely already reached the conclusion you need to adopt hardware, right? Because you were testing the other guy's card, you discovered ours. I assume you looked at GPU as well, or did you bypass GPU for cost or other reasons?
Speaker 3:No, so we're doing GPU as well. But it's you know it's always a bit annoying to have to buy the entire GPU die and then you only use the tiny end-lens encoders on it.
Speaker 1:So that is, I think the market is, I'll say, finally waking up to that. But again, you know that's kind of the biggest, the biggest, one of the biggest rubs, you know, with GPU architecture. You know, look we, you know we respect all of our competitors. So you know me personally and our general attitude as a company is you know we don't dismiss or you know you won't see us put out a lot of negative or really anything about anybody. You know NVIDIA, some of those cards do a pretty good job, you know they're, you know they, you know they do in some cases a very adequate job, you know. So you know there's no way that anybody could really say, oh well, the quality is, you know, is not sufficient.
Speaker 1:But the big problem is is that it is a GPU, the. You know the video IP is roughly, it depends on the product, but it's about 15% of the chip. So you know you think you're paying a hundred percent, but you really only you can never get more than 15% of the value. You know, and so that right there, just economically, it just sort of hits a wall for a lot of people. And you know it's why, when we looked at building this whole, designing the and building the VPU category, the video processing unit category.
Speaker 1:It was that revelation where it all started. It's like, wait a second, if you are in video and you're using this for video decoding, video encoding and other processing, scaling functions and advanced color space conversion and different things that we need to do you know scaling functions and you know advanced color space conversion and you know different things that we need to do. Right, shouldn't you have 100% access to the chip? So it seemed obvious. But yeah, the market's really really caught on there, so this is great. Why don't you give a quick roundup as to you know, if you can? I don't know, you know. I'm not asking you to pre-announce anything from you know, from NAB or for NAB, but what are people going to see you know at NAB when they come to your booth and and what are the kinds of conversations that you hope you're going to have, you know, with the market, with the industry, with those who come, you know, to see you?
Speaker 2:So you know, what they're going to see is is they're going to see how it's possible to not just transport content but to manipulate that content to fit whoever is receiving it or whoever's sending it, so that it works. You can build multiple workflows from one single stream to fit whoever is going to receive your content in a very simple and end-to-end type of way. It's an orchestration where you get all the monitoring and all that. So that is what they're going to see and how you can take these features as where you know, we're talking about frame rate conversion, we're talking about overlays and we're talking about auto shopping all the things you can do in a very simple way for a low cost and I think we haven't really mentioned that, but this is a. I think when we did the maths on this, compared to current cloud transcoding solutions, this is about a third of a cost when you look at a three-year period. So that's what they're going to see and all the things you can also build on top of this, because these are features that you can in a very nice way. You can put them together to build some really interesting workflows and build. You know, you can do ad insertion and things like that. So that is what customers are going to see at our booth and it's going to be.
Speaker 2:You know, what I'm hoping to get out of it is a conversation where we're not just talking about, you know, cost is always important and people want to keep that cost low, but what I want to talk about is how you can actually, you know, evolve the way you do things in a simplified manner without having to actually put that much effort into it, because this is basically something you can get it straight out of the box. It's not a big thing for you to get going with this. So, you know, I want to hear how. You know, what approach do our potential you know, customers have to, you know, having this hybrid distribution model, and how can they combine these things in order to improve efficiency and be cost efficient and have high service quality. So those are the type of conversations I want to have, because this combined solution with your product and our product gives the customer you know this, it's a toolbox, it's a toolbox and they have everything. You know it's, it's, it's yeah, they can do a bunch of stuff, and that's what I'm hoping to talk about.
Speaker 1:Yeah, pierre, I'll let you answer, but I just want to put an exclamation point on it. Is true, you know, cost is front and center. On one hand we like to say, you know well, it shouldn't drive all decisions. The reality is we know that at the end of the day it may not literally drive a decision, because something can be inexpensive but if it doesn't work, doesn't work right, isn't reliable, then of course it's not going to be used. But when you can cut the cost by 67%, when your OPEX goes down greater than 50% and that's not theoretical, that's you know.
Speaker 1:By the way, we have some, you know some more on the OTT, a little bit more simplified workflows than what you're involved in. The number is 80%. Yeah, 8-0% cut in OPEX is just amazing. But I love the point that you're making where not only are we reducing costs, which in and of itself should be like where do I sign, are we reducing costs which in and of itself should be like where do I sign, but you're actually in some cases bringing functionality, bringing capability, bringing performance that isn't even available or would be even that much more costly you know to do in sort of the traditional architectures. So, yeah, you know, we really see what you're doing as an enabler, and that always puts a smile on our face because you know it's great.
Speaker 1:There's a lot of uses for our products and for anybody's products, right? But it's really fun. It's cool when you see your product getting used, getting sold and it's allowing something that just wasn't possible before you know, or cost effective, you know. So that's the word enabler for me. Pierre, you know you're going to be there. Presumably. You'll be talking to a lot of engineers. You're going to be probably touching base with customers who are in the middle of deployments or starting deployments. What are you excited about? What conversations do you anticipate you'll be having at NAB?
Speaker 3:Yeah, I think it's going to be great to see some excitement. You know we've been cooking some great demos and some great software in the lab, so hearing what people want to do with it and you know where we've done a good job of sort of building the functionality that's needed and where you know which areas we want to. You know, dive deeper into this coming year.
Speaker 1:Awesome, Awesome. Yeah, it's so valuable to be in front of, you know, of users, of customers, to hear firsthand what they care about, what they don't care about. You know it's super valuable. As an engineer, my formal education is computer science, but I never actually went down the full engineering path, so I went to the dark side, as some would say, sales and marketing. But anyway, yeah, so that's great. Well, guys, thank you for coming on and speaking with me. It was a wonderful conversation and you know, if it's not clear to the listener, we, as in NetIn, are very excited about what we're doing with NetInsight, the two net companies. You know we're very excited and we hope you'll come see both of us. But definitely make sure you visit NetInsight at NAB and I guess, until next time, happy encoding.
Speaker 2:This episode of Voices of Video is brought to you by NetInt Technologies. If you are looking for cutting-edge video encoding solutions, check out NetInt's products at netintcom.
