Join Jamie and Paul Cox, Director of Software Engineering at Broadcom in episode 11 of The RDK Podcast! They discuss:
🎙 Broadcom Inc.’s role in the RDK community
🎙 Their contributions to RDK
🎙 How they’ve had to adapt as the technology has adapted
Hello, and welcome to the RDK podcast. I’m your host, Jamie Walker. And today I’m joined by the fantastic Paul Cox, Director of software engineering at Broadcom.
Hi, good morning, Jamie.
Good morning Paul! So shall we just jump straight in? I guess to start off with, tell us a little bit about yourself and your background in video tech and how you got to join Broadcom today.
Sure, thank you, Jamie. Yeah, so I’ve done video engineering, video technologies, all my professional career. So which is now kind of scary. 26 years. I started off in Philips Research quite a long time ago. Moving to ST Microelectronics. ST were really I think, one of the main delivery companies for silicon in the set top box space in their early phase in the industry, based in Bristol here, and I was responsible for security in Chemical Transport development, if you like at that point. So 2005 came along. And I was offered the opportunity to start up a team in Bristol for Broadcom, effectively as Broadcom was growing into the set top box market in Europe, and kind of coincided with the migration to HD platforms. So yeah, I stepped team up in 2005. We’ve got a team here. And we’ve been here ever since.
That’s it, and obviously, being quite local to to ourselves, right? It’s great. It’s great to have you on the show, Paul. But I think for everybody who’s not so familiar. How does Broadcom contribute to the video and broadband space specifically?
Yeah, so Broadcom’s, I would argue, absolutely centre and central to the broadband and video technology, you know, we are delivering the silicon into pretty much all of the RDK products that you see on the market today and are in development. And, and really since the beginning of, of RDK, both broadband and video we’ve been doing so. So you know, in really simple terms. We develop silicon, we deliver an SDK on top, we map that into the RDK interfaces that RDK LLC come up with, and then we support customers through to production with it.
That’s it. So obviously Broadcom been a household name, you play a key role in the delivery of this function. What steps did Broadcom take and your team take that took you in the direction of RDK then initially?
Got you… Well, I guess to some extent we joined we get involved on the back of our relationship with Comcast, you know, we have delivered semiconductors into Comcast for many years and certainly way before RDK became popular. So I first projects were with pace back in 2010. To give you some idea, so really before RDK became named as RDK. So, you know, initial development was was was able to Comcast team, and then really as that evolved through into a separate organisation, we then looked at effectively following and delivering that software into the into the community that’s now formed.
That’s it, so obviously, again, a global scale for Broadcom and your team. Can you tell us a little bit about some of the RDK projects that you’ve been a part of that maybe stood out to you during your involvement?
Yeah, so as you rightly say, I mean, RDL has now deployed globally. My remit and my team’s remit have primarily focused on India. So platforms that have deployed in this region. And so, you know, I think there are now a good backlog of devices that have gone out with RDK here, whether that’s with the Liberty organisation, and yeah, so I think all those RDK devices have gone out with us, I think have been important and valuable. I think certainly the first generations where the Liberty team were delivering I think we’re really important to us. But really, all in my mind a recall, I think, you know, it’s, it’s about from our point of view, it’s about getting the delivery of drivers, getting stability behind those drivers out to customers, you know, some of the early RDK projects, effectively RDK was taken by a middleware company integrated into their product and then delivered to market. But if I look at the last, say, five years or so that’s now changed more. So the fact that an operator is taking an RDK package from us building a product. So you know, that say that it’s definitely a development path that’s migrating and evolving in the last couple of years.
That’s it, I guess the, the exciting part of the journey is working with those clients, right to overcome the obstacles and deliver that final product. But I guess, I guess for us, it’s been said many times, especially on our podcast, that RDK is very different to any other type of software on the market. What do you personally think that makes it stand out amongst those other key players within the industry? And why it’s taking off the way it is?
Yeah, I mean, what what for me, is the standout is the is the fact that it is an operator driven development. And, you know, because most of the set top box products that I’ve been involved with in those 20 years, you know, you were taking proprietary middleware solution, proprietary CA solution; you’re building a product and then delivering it to an operator against a spec. And, and it was all about testing against that spec filled, trialling certifying delivering, and, you know, the actual implementation of the software, to some extent was was irrelevant, it was does it meet my functional requirements. And once you change to a model, where the operator is aware, is building the software is aware of the capability, it naturally brings them deeper into understanding the product.
And you know, that that for me is changed. It’s changed the nature of the relationship between us, if you like at the bottom of the stack, and the end customer, the operator at the top, I think it also changes the operators opinion on what what’s important to them. And truthfully, a lot of that software doesn’t differentiate their product. And so what, RDK, from my point of view, what RDK is done and said, Look, if the software doesn’t differentiate our product, let’s share it. Let’s make it available. It’s nothing, you know, there’s no proprietariness to it. And so RDK, on the back of that has really built into both a combination of open source components, but also now development pieces of of work, I mean, WPE browser would be a great example, or the Firebolt framework, your significant chunks of technology that an operator is saying, look, it doesn’t help me to make this unique to me. So let’s share it to the industry. And then we will focus on what differentiates our product. And you know, so there is still a closed aspect to it at the very top if you like, and which is absolutely right and should be, but then share more the code. So it’s safe. So for me, what makes it case so important is is that the contributions and the sharing of that code is done on a in a way that we’ve not seen in the industry before.
Yeah, that’s it, I think, I think you’re right in terms of taking away from a standard spec for a product and allowing individual contributions, the flow of innovation, and where that’s going to take the software going forward. I mean, everybody can contribute their own ideas. Right? And this is pretty much where where Broadcom have been, you know, innovation is a big part of a Broadcom’s background and growth over the over the years, as you’ve seen. With the industry moving as fast as it is now, how have Broadcom adapted to this?
Yeah, you’re absolutely right. I mean, things are moving very quickly. I think that for us, always key in our development processes is scalability. And, you know, I think if there’s one thing we are well known for it is it is software quality and software control. To give you some idea, there are 24 different products that we support different chipsets that we support an RDK release package and to make sure that those 24 devices are maintained, managed, updated, kept up to date with the flow of changes, needs some good quality software engineering. I think that’s the core of it really is, is having robustness, consistency, stability in your code, people.
When you’re building these kinds of devices, it’s like, in many respects, like building a house; you need solid foundations beneath you to go build those fancy houses. And that’s really what Broadcom is providing it’s silicon that’s stable, that’s consistent, robust. It’s then drivers that do the same. So that really I think is is, you know, keeping up with the speed of innovation is is great, but you must ensure that the basis, your your foundations are robust and solid. So I think that that’s number one for us scalability. And so that’s done through our own internal testing, you know, release procedures that go with it. So I think that that probably is the most important thing. I think for us.
I think the second thing is that you’ve got to be able to track the changes that the operators are looking for. What innovations are they looking for? What’s driving that robot for because you talk about going fast, it does go fast. And what’s the best way to track that is to follow it. I mean, you’ve got to keep up at the same speed. And, and for us, that really means really strong relationships with our customers, our partners, we track their specifications, we discuss the specifications, you know, we don’t just, you know, it’s not like the next generation OTT app appears on the desk. And it’s okay, now we need to support it, we need to talk to them about how it works, you know, because there’s a whole pipeline.
You know, if you want to move from OpenGL to Vulcan in a in a graphics sense, you know, that’s a different piece of silicon that needs to get developed, delivered into chip, the chip needs to go into production. So we’ve got to be 2/3/4 years ahead in our thinking about where that where the markets going, where the the applications need to be. So yeah, from from our point of view, it’s about really thinking ahead, understanding where what the requirements are going to be. And because of the lead time in developing silicon and delivering it, we really got to think multiple years ahead.
Definitely always, always staying one step ahead of everybody else. T
Two or three steps in our case, yes, yes. *laughter*
No, that’s it. Well, Paul obviously, you’ve given us a very good sort of insight into Broadcom’s growth, the innovation, the team communication. What’s what’s the next step on the journey for for Broadcom, then? And how do you see it moving forward?
Yeah, I think the key thing for us is to continue to offer a set of devices and different cost points. You know, we have a mixture of operators who are in different stages of their, their operations. And I think and the products they’re trying to build, there’s no one size fits all in the video space. So from our point of view, it’s about delivering a different set of devices. So devices that have absolute leading edge innovation in them. So maybe that is operators that want to take advantage of fibre voice tape, technologies, or machine learning capabilities, even in the future, you know, 8k video, for example.
So we want to make sure we’ve got silicon that will support that kind of roadmap at the same time as silicon that will support operators who are trying to reduce their their capex, and obviously just drive into a wider customer base. So if you’d like, step one for us is to make sure that we continue to offer a broad roadmap of devices that’s obviously critical. I think the second thing for us then it’s a is is sustainability. It’s about making sure that we can keep those devices keep the support and the development in place to deliver them and keep them going forward. And it’s about say, it’s about being sustainable. How we do that.
I say I think sustainability is having a big part to play in all future plans as well, especially in video as well in the turnover of devices. But Paul, it’s been great chatting with you today. Really thank you for joining us.