21923 - SCTE Broadband Aug2023 COMPLETE v1 (1)

71 Vol. 45 No. 3 - September 2023 Issue from the industry Of course SDI, ASI, MPEG-2 will be with us forever, because there is legacy stuff that’s going to continue and why not? There wasn’t a lot of resistance overall, but the speed of adaptation is not as fast as I would have believed it would be. Within distribution, terrestrial will be with us for the foreseeable future. Satellite is brilliant for distribution purposes, because it can reach places where you don’t have fibre optics. Satellites will never die, because there is a lot of legacy content, plus a lot of military and government activity. What’s around the corner for IP? SF: Ultimately content delivery as a service will go away and it will be put on top of the IP-based platform, that’s not up for debate. It’s just about how fast it will happen, and it’ll happen very, very fast with the next generation; LEO Satellites (low earth orbit) are going to emerge because by then bandwidth will be no problem. Limited bandwidth has always been the issue, but if that changes… SF: With serious bandwidth you can then fulfill crazy resolution capabilities. Then there is the need for validation equipment; if you don’t validate that it works, then how do you actually know it works? That’s only the test and measurement angle. We see that the capabilities of our system are practically endless and with the addition of different functions, we will create a system core that can become one of those invaluable boxes in production and that could ultimately allow users to get rid of other technologies. Can you give me an example? SF: Yes; I mentioned signal generation. We have now a signal generator which can generate any kind of signal, all the way from standard definition up to 4K in any frame rate, any kind of resolution with items and test tone patterns; that means that’s suddenly you don’t need that anymore, it can just use the VB440. The customer wants fewer surfaces tomaintain and to operate, but it demands a lot of hard work to make that happen, because you just can’t load functions, upon functions, upon functions. It has to be very logical, extremely accurate it has to work all the time. Five years down the line, where do you see Bridge Technologies? What are you doing? SF: Well, first of all, we never think as far ahead as five years, the most we think about is three years. Beyond three years, anything is possible and nothing is predictable. But three years-ish at least we can see fairly clearly; I suspect that we’ll have a plethora of functions. So if you want to create an outside broadcast vehicle, or a gallery, or a studio with 12 to 24 cameras, you will have probably three to five of our units or maybe even more inside that vehicle for your production. It’s possible to reduce half a rack or more, probably nearly a full rack of equipment into one of our units. We already have customers doing this. Will AI be helping you as you would go forward? SF: I’m a big skeptic. I don’t even call it AI, because I don’t think that’s a very accurate name. I call it machine learning and I can see pattern recognition and ML as an interesting part of our future, when it comes to automatic analytics of big distribution data. As it has a lot of capabilities; it can automatically look for specific traits and then give you both better warning mechanisms. It’s perfect for dissecting errors in the past as well. But overall, I’m not very impressed by what I’ve seen so far. Most people are either blown away or worried sick. Is it just early days for you? SF: In our business, customers believe that they want certain technologies, such as freeze frame detection, which we have explored already. It takes a huge amount of calibration from