Episode Transcript
Troy is going to discuss the front row seat he has to innovation at Ross video and the unique way customers are adopting new technology and new workflows. Before we get started, it's worth mentioning Recode is a top tier partner of Ross video installing systems across the country. Well, Troy, it's great to be able to talk with you today.
And there's a there's a lot to cover. And I think we ought to just dove right into it.
That was great. Thanks, Steve.
So one of the things that always confuses people is what do we mean by specific terms, for example? What do we mean by on prem? What do we mean by cloud? What really is hybrid? Can you maybe take us through some of those those definitions? What do we really mean by on prem?
So on prem stands for on premise, which essentially at its core means that the document you're using is in your physical location or physical location that you're renting or owning, you know, different from, of course, cloud, where typically that means the equipment that's doing your video processing, audio processing, etc. is not actually yours. Typically, it should be more something in A.W. as a Google now, that sort of thing.
So it becomes usually you you it's on physically with you.
So and that would that would kind of include things within the physical building itself, maybe a small campus kind of location, something that's physically adjacent to where you're working.
Well, I wouldn't say adjacent because that's because because Rémi. Right. We haven't talked about Remi versus cloud production but but it usually means that equipment is, is physically somewhere that you have access to. So, you know, trucks are a good example of one prem as well because I mean, a truck is not premised on their hand is still you have the physical equipment.
And it's interesting because if you compare it to cloud, for example, it can still be the same kind of processing power use in the cloud, you know, in x86 computer, but it's your x86 computer. You can actually someone that works for you can go touch it. I think it's easier to do that.
That's that's a great definition. If if you can get at it and mess with it, it's on prem.
Yeah. Yeah. And when we contrast it to cloud, I mean, everything but cloud is I mean you can still even have your own private cloud. Like some people build their own cloud infrastructure, which is on premise. But, you know, we're splitting hairs there, so it's easiest to not worry too much about that.
Okay. And what do we mean by a hybrid?
Sure. So hybrid is when you want to do a bit of both. So we haven't talked too much workload processing production means yet, but if you have some cloud processing and some on premise equipment and in our case, I we also say, generally speaking from light production standpoint, we are actually bridging them with video and audio signals. So you're doing a production with some equipment that's in cloud and some equipment that is on premise.
I mean, that hybrid is, I think one of the more attractive certainly and for the next number of years where that people will actually be doing their productions as opposed to either all on premise or all. In fact.
One of the one that I think kind of needs to be cleaned, cleaned up a little bit in definition wise. And you alluded to it is Remi, why isn't cloud the same as remote?
Well, sure, because because Remi usually just means the simplest definition is that the people operating the equipment or some of the equipment, because you can, again, have some remi, some not for your production, are not like adjacent adjacent to the production equipment. So a good example of Remi is and you don't have to be cloud to be doing that.
So you can have all of your production processing power in the cloud, in which case you are by definition. Remi Like if you're doing cloud, you're right because your operators are obviously not sitting in a basement wherever AWB is hosting their servers. And of course the traditional, traditional view of production would be that you have a rack room and then a floor nearby typically or even next door.
Sometimes you have your control room and if you want them to your your control room staff could run over again and get physical assets to that equipment, whereas you can still have on premise equipment and do Remi because maybe you're all all of your physical equipment is in Washington, DC. You've got a big server firm. It's like a production rooms there, but you're actual operators.
Maybe they're maybe somewhere in Boston, somewhere in Seattle, so they're accessing the equipment remotely. Remmy stands for remote and that's a short form, but they're still accessing on prem equipment. So because you're doing Remi doesn't mean you're doing on cloud doesn't mean doing cloud production. Remi Just means the people who are doing your production are not in the same physical location as the equipment that's processing the video, audio metadata.
Great definitions. I think that's that's one of the things that people get kind of confused about. And and having not that cleaned up, I think is, is fantastic. Take us through what what cloud processing really means. Is it are we taking out physical gear that we're used to having in our facilities, in our truck? And are we putting that in a data center or are we doing something different there?
Sure. So let's talk about cloud in a couple of ways. I mean, cloud cloud in broadcasting has been used for quite a number of years. This is not like just this last couple of years. For example, newsroom control systems, asset management systems, those kinds of products have been in the cloud for quite some time, you know, five, eight more years, I'm sure, with some vendors.
So what we're really going to talk about today and focus on today are those aspects of it. But for production, which means if I'm going to have equipment that sits in the cloud that is processing a video stream, an audio stream, obviously the sources for those, of course, are local somewhere generally. So when we talk about cloud production, we're talking about the shift.
And again, Ross, we we concentrate on live production as opposed to post production. So again, people being in post-production for quite some time. So really we're talking about live video streams and live audio streams. If you were doing a sports event, for example, and the only thing in that sports event are the cameras and you're in those cameras are going into a cloud of someone's and then in that cloud is where you're running the software that is a virtual production switch or virtual audio mixer.
And then of course, all the other sorts of features like watermark encoding for advertising, etc.. So when we talk cloud for today's conversation, we're really talking about the live production version of Cloud.
And so one of the ways that I think I interpret that is it is the virtualization on standardized equipment that enables us to to have the same functional processes as what we'd have on instantiated hardware, on premises. But with it being virtualized, it can run on common equipment in the data center, in the cloud and not have to be unique equipment.
So literally, it's we're just we're using the same services that that a cell phone service company might provide or that a data mining group might use in the cloud. Correct.
Close and this is where we start getting pedantic and splitting hairs. And since my background is software pedantry. Yes, because we ensure that's what it's like. You can, of course, take the same software that you run in the cloud and generally speaking, run that software on a piece of what you owned on prem as well, because it's software running on generally an x86 machine of some kind.
So when we say cloud processing, I wouldn't say that it's the ability to run it on. You know, we can think of it's common off the shelf platforms because you can do that on premise, but that doesn't have to be in the cloud. So when we say cloud processing, we specifically mean running software, which is, yes, the software does run on, you know, that kind of compute processing, but additionally you're not running it on equipment that is on premise, but you're running that software in a cloud provider and then not your own physical are perfect.
Well, I think that takes us right to the domain topic here. Is is hyperconverged infrastructure. What does that really mean and how does that how does that work for us?
Yeah. So today, I love the timing of this because the the recording of this this conversation was thankfully two days after we put we made a new announcement. So we we've been really working really hard and creating something which has never been seen before, which is, you know, we call our Hyperconverged platform and this is an on premise solution.
And what it really means is if you take that thought experiment and you say, well, if I take all of the processing that I do for for video and audio and the metadata of the closed captioning and such and if I put as much of that as I can into the same piece of equipment, what does that actually do?
Then on the surface, it's like, Well, okay, I guess if your power supply is maybe it's a bit smaller, but it's far bigger than that. What it really means is an insane reduction in complexity and everything that you start caring about. It's not just complexity, it's size, it's the power to run the equipment. And then one of the things that it's funny, David likes to say is after talking to a lot of people, there's nothing that engineers, engineers in television stations have to see more than cables they despise cables because cable management, you know, you've got to run thousands of these things.
And when you actually dismiss Hyperconverged, it's a massive, massive reduction in cables. And with that reduction comes a lot of really interesting things. So and I'm going to try not to send too much sales pitch. Just take one for us. I'm super excited about this. We've been working on this for years and this is not a new thing we just announced yesterday is so we just got to announce something called the Ultra SFR 12 and the 12 is means 12 vacuum.
It's a big investment roughly. So but in that it's got flexible IO, right? So it takes 2110 and takes SDI, takes fiber on inputs and outputs up to 288 squared and you HD 12 gig rates and you can mix and match any input output. So all that normal, you know, you may do a conversion externally sometimes, you know, SD 8 to 21 and less that's just built in because the system takes it from the IO converts down to video ships in around so it's independent and the system doesn't we don't care what you bring.
And then we added in a ton of the functionality that traditionally you would actually have individual components for, right? And you'd have to cable from the router into that equipment and then back to the router. And you have all of these loops everywhere. So audio, it's got a 6000 squared audio matrix in the middle of it, an audio embedded embed.
So you can take any video signal, DMB audio and then there's some processing such as gain on every and on every audio signal 66,000 and then shuffle, right? So any and any audio mono channel from any source to any mono channel on an output. And historically we've had people who buy old tricks just for the audio embed. The embed, right.
So you add that in, you put frame synchronizes on essentially every input, clean, quiet on every output, and then you go further and then you're like, okay, well what are the big things people use it. We were talking to one client. This is actually a 2110 client now, but a year and a half ago and they they were saying that for their 21 environment, just what they were doing, half of the 2110 signals they were doing were for multi viewers, comes out of the router.
It's the multi viewer and of course, the multi viewers. So you need one signal for every pivot, right? And it got lots of these things through your facility. That's a lot of signals. And so what we did is we put multiple users inside the same box. So up to 48 multi viewers, which is know for most facilities, 48 multi viewers and 100 heads.
Yeah. And up to 100 pips per multi viewer. I wouldn't do that. They're kind of small.
Wow.
But I usually, you know, go from 40 or 25 whatever makes sense for user in all of the routing for that is independent in the box from all the inputs and outputs. So if you look at that, that's 48 times 100, that's 4800 cables and they're all gone now. And the weight of them and the cost of them and then and then we took our production switchers, which are not big consumer of sources, and we embed the production switcher into the router.
So, you know, you can have an acuity switcher, which is our large one that's used often and like very large sports productions or the carbonate switches, we can actually put up to eight carbonate switches in the same box and every switcher has access to every input and output. And again, independent flavors, all of a sudden you get SDI in fiber and 2110 for all of that.
And once you've done that, so I have a cheat sheet here, guys did an analysis and what they did is they said, Well, let's pretend somebody used every feature inside of the set for 12. Unlikely, right? Nobody uses every single, but they used every single feature. What is the equivalent, you know, system? If you didn't use all tricks, you did it in a more traditional manner.
And it's interesting that traditional manner, by the way, would also apply to whether you're using, you know, standard off the shelf processing, because a lot of the big thing you want to reduce is sending video from one computer to another computer yet, you know, an x86 based one or B, you know, the more custom hardware that many of us make, right?
So the comparison was guilt. The old tricks of our 12 takes 14. Are you just seen as 12 or 14? 12 are you the processing box? And two, are you for the power? So 14 are equivalent in individual components. And we tried to be fair about this, right? We weren't trying to make it. You know, something works and we look and go, Well, you just cheated with 549, are you?
And that's racks and racks. And right now and you know, we all want to be green, right? It's one of the big things in the industry. So the effort 12 that location is 2400 watts and the equivalent in in non hyperconverged solutions would be 52,000 watts. It's a massive decrease, 576 cables versus 11,000 cables in £128. And this matters actually, because a lot of people also travel with are equipped.
Right. Maybe it's a truck. And we do have the occasional person who actually flies equipment around. So £128 versus £6,000. That's a.
Tremendous weight. I know from from past experience, some 30% of the weight of a truck used to be the amount of cabling that they had to carry around. And and George Hoover, I think I'm sure you know, George, former CTO of of NEP and retired now he used to he used to say, you know, you manufacturers, you you all make switchers, you make routers, you make all this modular vacuum walls of yours.
Can't you take two or three of those and combine them together and and give me one box? And it looks like you guys have I've got the whole thing altogether and.
Yeah. Oh, the funny thing also is I always forget something as we also have an audio mixer built in as well, 128 by 64 and there's lots of fancy stuff, what you can subdivide it, etc. But the you know what? Because a lot of us I mean, I used to have hairdresser in the industry.
So.
And when I started, I mean the first thing you'd see in any studio, you'd go into a matter what kind of studio you'd see on in engineering. There'd be a big, you know, used to be AutoCAD drawing, right? Shows all the equipment, the interconnects and it was very big and complex and a lot of people are still building facilities that way.
And when you look at something like Hyperconverged, it changes the paradigm from I have to interconnect and send all this video. And this, by the way, doesn't matter if you're talking to SGI or 2110, it's the exact same issue. It just happens in 2110 that you don't have an SGI router, you've got a router from, you know, pick your favorite i.t manufacturer.
And if you can reduce that, you reduce massively, you know, costs, weight, complexity, and you also reduce points of failure. Like you don't have to worry about how somebody cuts the cables because cables aren't there. But right. I think weight is frankly one of the biggest things. And Rackspace, because when we talk about cloud, which we talk too much about, one of the big things about cloud, of course, is, you know, you can get more services and they don't take up more physical.
Rackspace matter. You know, if you're downtown New York, that that's basis critical, right?
Yeah, absolutely. There are buildings there I've dealt with where you couldn't physically get all the equipment under an old design on a on a single floor and you were going eight, nine, ten floors and you're starting to get into into the maximum lengths of cable runs at three gig cycles. And it's just crushing at 4K if you're trying to do 12 gig.
So yeah, it's it becomes a real issue.
Yeah. So, so here at Ross, we are, we're actually pushing heavily forward in both directions simultaneously. So we really believe that with hyper converged, it's very compelling for a lot of people, solves a lot of problems and some of the issues people would have had before, you know, like, well, I need to go a cloud because I don't want to rent another building.
I'm not a physical space like, well, that's true. But then you get hyper converged and also need to racks and racks space back. So we're pushing really hard on that. And, you know, we're very proud of it. We think that it's actually going to be a real game changer, especially now that we have the larger rack frame. But the same token, we also believe there's lots and lots of benefits to cloud.
There's lots of reasons people go to cloud and and we believe strongly in those as well. And we have another team that's working heavily on cloud processing of everything, to be honest.
So so you've got the physical gear hyper converged into a very compact package. 12 Are you for that for the gear? Two Are you additional are you for the power supplies? What is the the cloud ing pack going to be? What happens next?
Yeah, well, cloud is interesting because you go to cloud for different reasons. I think. So. Cloud, of course, has a lot of like I mean, one of the biggest things cloud gives you is obviously flexibility, right? Like you can get, you know, a new virtual production switcher very quickly in the cloud. You don't have to wait for someone to deliver it.
You don't have to rack it. You can go on to your cloud provider, start the service as long as you're licensed and the way you go and cloud is also a great place. Like if you want to do if you want to experiment with a new show and you want to do an hour a week just to try it out, see what people like, see how things are.
I mean, you can do that with cloud without having to go out and buy a large set of equipment, install it, wait for it to arrive. You can just start, you can you can dabble in it. And I think the other really nice things about cloud is the ability to flex, right? When you buy hardware, you know, you get a certain capacity and if you want more capacity, we have to buy more hardware and that's fine.
And, you know, it seems like an effort. Well, it's very cost effective, but it's still not, you know, I guess, you know, hardware for a short term.
But it's still it's still big CapEx expense for somebody as a as opposed to a minimized optics.
Yeah. So so good examples of hybrid are, you know, if we talk about something like television station news, right sorry elections like. Oh, okay, yeah, yeah. Election season comes every four years and it states and for a lot of people that means they really want to do a much more involved, show a lot more graphics, if nothing else.
So you may want for three, four months a lot more graphics engines. And if you have cloud and the ability to just say for the next four months, I want to add a number of channels of graphics and then I'll stop using them again. I mean, that flexibility has a ton of value and physical equipment isn't as flexible, obviously, by its nature.
And so that's one of the big advantages of cloud, of course, is that flexibility. And that's where the hybrid comes full circle is, you know, you can have the hyper efficient hyperconverged solutions on prem, but then you can flex in the cloud either to experiment or Olympics. Olympics We do have a two year trend. I can do the Olympics and I don't have to keep it.
I don't have to buy it and keep it in the warehouse, frankly, for the next Olympics. And to be honest, sometimes if you buy equipment for something like that, you may find the next time you want to use it. Now, that's 2 hours at a day. You don't want to do something new. I did the last Olympics, you know, in three gig.
But this year I like to do it in the company I bought two years ago. It's still three gig and it doesn't work. So that flexibility that the cloud gives you is great and hybrid solution is kind of like take the pro because everything has pros and cons. So I can say, here's my pros and cons of all parameters.
The pros and cons of cloud. What if I don't take the cons? What if I don't accept the cons? What if I just say I want the pros from on prem and the pros from cloud and use those? And then the cons. I'm going to deal with those because I'm going to use one from the other. So that's what I mean.
That's what hybrid is about. And that's why we're pushing so hard for both solutions, is we really want to be able to give people all of the pros. And if we were doing just on prem, obviously, there's a lot of you know, there's a lot of people that we're just not going be able to make happy, right? You're making people happy.
So what we want to do, but it's actually one of the big drivers here is is making our customers happy. And if I can't provide them cloud solutions, I'm not making them happy because of some of the downsides to physical equipment and vice versa. The other way, if I if I went all cloud and didn't take advantage of what things like hyperconverged can give, then then there's another set of customers.
I'm not making that right.
And then you can you can address a broad swath that that needs the hybrid solution because they've got local productions they've got to be doing. And it's just easier to to have the gear on hand and in your pocket. But then they're doing regional productions or nationwide or worldwide productions, and you got to be able to use all those remote sources so that you've got people, wherever they are, working on working on the project.
And remember, of course, that Remi can be done with both models and you can go to Remi with a hybrid solution as well, frankly. I mean, you could have operators who are accessing some on prem equipment, as well as accessing some cloud solutions simultaneously from wherever they are. As long as the on prem equipment they're accessing gives the ability for it to be controlled remotely.
And I think that's pretty standard for equipment from almost everybody today is that, you know, it's IP control typically, right? Right. The only thing you've got to deal with is a little bit more latency. If someone is in Seattle and controlling equipment in Tokyo, it's still doable. It just means there's been more latency on the control manual. And we know people who did that obviously.
Yeah, well, you know, that that that brings up a good point. There are challenges to both and we touched on some of them. The amount of cabling, the complexity, the physical size that traditional gear takes, moving to the cloud, obviously there's the latency issue and bandwidth accessibility, if you will, for for doing LAN productions. What are some of the other challenges that that that you've seen and what have you guys done to overcome them?
So one of the big challenges Cloud has, let's be frank, is it's just newer. And, you know, for the on prem equipment we've had 30, 40 plus years of figuring out the workflows and really optimizing what that means. And cloud is for live production, right? Cloud for game like video and audio. And now it's just newer. So some of those issues that we've resolved for the on prem equipment aren't resolved yet.
So you mentioned latency, obviously, that's one of the issues. On top of that, though, there's bandwidth is a big one video in the cloud, right. So obviously one of the big pushes people liked about 21 tens. If you look at 2110, it was, hey, I want to get the advantages that IPD is for video, but almost everybody who went 2110, almost exclusively went uncompressed, right?
So the second I go to cloud, you just don't have the data pipes. You just don't. I mean, if you look at it at some productions have terabits of uncompressed video data, you actually look at it, you know, because a 12 gig signal is obviously 12 gigabytes per second. So and if you've got lots of cameras, lots of video servers, that adds up really fast.
If you want to get into the cloud, some of those sources move to be needed in the cloud. Right. Because I'm doing video servers, I'm not going to do them on prem and shift the video up all instantiating video servers in the cloud. So mostly we're talking cameras there on that are physically need to go up but you still need to generally compress them.
I've I've spoken with somebody nobody yet has told me they're actually managing to get uncompressed to the cloud. So at this point it's compressed. Typical things we see are you know, SRT I think is one of the the more common ones by far to get video you know the cloud right but that compression adds latency. And for some people, if you compare if you care with the fact that you're compressing your video, you can make some argument about quality being not as good as uncompressed.
I actually think that for the most part, the quality is that as long as you choose a decent compression, you're actually pretty good in the layperson, I'll never notice. You know, people who know it really well might, but I don't think people at home will actually notice that. And obviously we all know that at the end of the day, the time that someone sees at home, it's compressed, right.
Have a self-interest. Yeah. But one of the challenges here, though, is, you know, with SDI, I used to be able to say, if you give me video, it's one of the big six. You know, there's 720 to the different frequencies, etc., that standardization isn't necessarily there into the cloud like which transport are using. SRT, a lot of people like to use MCI nowadays, right, as we know well, but they're not the only ones.
And then even within SRT, which codec you're using station six for the DC six five, you know, pick your favorite codec vendor. And even within those there's various settings. And for interoperability, you really want to get video from anybody and receive it. And because there's so many different ways you could do that today, you know, you got to be a bit more careful about that.
So that interoperability is something on video is still working it out. And then there's you know, if we look at 21, I think 21 hands, a good place to compare to cloud with 2110. I can use, you know, ISO for four to tell me tell me all the video signals that exist where is the equivalent for ISO for for video the use in the cloud and the reality is is I'm really isn't one for broadcast video.
Yeah so one of the things that we've done so far video we've actually put proposals in front of ANWAR on iOS four, which is the discovery of video signals and equipment to extend it actually to include things like, well, could I discover like an SRT stream somehow? What does that look like? So we're talking with ANWAR about term extension status for and the next one is switching, you know, again in 2110, there's iOS five used as a five to switch video signals.
Where's the equivalent that says, I can switch? I know that if I've got a control system, maybe independently from someone, that they can switch all the video sources that they know about in a 2110 2005. So we're talking about extensions to ISO five to be able to again switch all the kind of video that you would find natively in the cloud in the including things like KDR and CDI is, you know, Amazon's way of uncompressed video, typically with between cloud instances.
So we're pushing that, that ability to do that. And then there's control and in orchestration because it used to be equipment was pretty plug and play. Historically, no video video flowed but also there's control and of equipment. And again, you know, as an operator, do you really want to sit in front of a control surfaces like, okay, this is the limits controlling the video server.
Here's the one that's controlling, especially from different vendors. So if there was an easy, concerted way to have a control system or even individual products see each other controller to talk to each other. And we've got a proposal that we've been talking about internally called Catena, which stands for a series of interconnected links. And we've been looking at an organization called the OSA, the Open Service Alliance, which was built around making cloud microservices work better to be able to control, you know, in a similar way, control equipment from any vendor because again, usually you can get control between equipment within one vendor.
But Multi-Vendor solutions, of course, it's really push and also it's what customers want. But let's be honest, I mean, customers want flexibility to be able to buy equipment from multiple people, not have to be locked into one vendor. And this interop is is really what allows that. So we've been working hard on all of those factors, but it's not done yet.
And there's there's work to do still, but we're pushing forward very hard on it all.
What do you think the possibilities are as a result of of getting some of these these bugs worked out? You know, I was just thinking, if if we can have auto detection of signal types and compression rates and those kinds of things, and we we we get better control of of disparate manufacturers, of solutions. Are there some advanced techniques that maybe can jump to by can we can we apply some AI here and help the TD to accomplish more, have a higher quality production?
Well, air is a fascinating topic in itself, and we're only talking two things for an hour today. So to to jump into it, I'm yeah, I mean, there's there's you know, we've already released some A.I. products and there's a lot of A.I. actually in broadcast already known as a light production area. But live captioning, for example. Right, is an area where air is quite prevalent and media asset management and A.I. for certainly you know annotating clips with metadata is is quite commonly used today where we're using A.I. right now within our robotics product lines, camera motion systems, as we call it, to be able to actually steer the robot so that if, you know, if
you shift like I tend to you probably notice a tendency for and I apologize to your editor if you're trying to frame which just my nature. But if you've got someone like me and I'm doing this, we've got to so that it can actually steer the camera to keep you framed correctly with some hysteresis. So it's not, you know, making your cameras all look terrible.
But A.I., going forward for a lot of stuff has a lot of interesting possibilities. And I think we're only touching the surface of it. But to be honest, I, I know lots of people are looking into what more we can do in the future, including ourselves.
That that's great. I see it as is anything we can do to enable the creatives to remain creative and avoid that technical domain that the better quality production we're going to get and that the tools that we're talking about here, the tools that you guys have developed, I think are just targeted right at that. Simplifying the overall operation console in the equipment and the functionality, reducing weight, reducing complexity.
All those things go to a solution that makes it easier to build workflows to to change those workflows on demand and to have a much more elegant solution, if you will, so you spend more time on the creative and quality aspects of the production rather than than trying to make sure all that all those connections are working correctly and everything's behaving right.
So yeah, one of the things that I've always believed in is let's let people do the things that people are great at and then less like computers do the things that computers are great at. And there's no if a computer can do something really well for you, there's no reason to waste a human's time doing it. Let's let the human do the more interesting things.
And was one of the reasons we you know, we've developed automation systems for many years and I think it's worth mentioning the cloud stuff. You know, there's a lot of discussion here when you talk about cloud productions, people talk about putting in production switches in the cloud, which is great. You know, we've got a production switcher that we're we're quite proud of in the cloud as well.
But it's more than that. You really want to talk about the entire infrastructure to do a production. If you want to talk about cloud production and it is not just taken from the cloud, my server in the cloud is kind of put my automation systems, my newsroom control systems, my asset management systems. I'm going to do all of that in the cloud.
And that's that's the direction we're looking at it from, which is not, you know, can I put it in what does it look like to put a switch in the cloud or what's it look like? Put playout server McLoud We're looking at it more from let's talk about putting a production system in the cloud and and that really involves a large number of systems.
You it's got graphics, it's got playout, it's got the switcher audio processing as well, etc.. So I think that's the more interesting way to look at the cloud as well is not just a moving one piece of kit from here to here. It's moving actually the ability to do the entire production of the studio up into the cloud.
Now. Well, this has been a absolutely fascinating conversation. I think we could go on for a while.
I was like, oh, we're over like.
Yeah, it's going to incredibly fast. Well, Troy, I really appreciate you spending some time with us today. Tremendous talent. And again, kudos to roster for your development here and greatly appreciate your time with us today.
Thanks for watching. Broadcast Apologies. Make sure to subscribe to the podcast to receive future episodes. Follow Decode Media on LinkedIn, Twitter, Facebook or Instagram to receive news on additional AP broadcast and post-production technology content. See you next time folks.
