Star Citizen: Around the Verse - Upgrading Ships to Item 2.0

Transcript — every row opens the video at that moment

1. 00:00:15

   Hello and welcome to Around the Verse, our weekly look at the development of Star Citizen. I'm Sandy Gardner. And I'm Eric Karen Davis. While there are a lot of features planned for the upcoming 3.0 0 release. One that's been affecting just about every department is item 2.0. Yeah, that's right. And migrating every ship to this new system is an enormous task. So today, we're going to dive in and explore this new system and the wide-ranging effects it will have on all of our ships and vehicles. But first, let's go to Austin and Turbulent for their studio updates. Hey guys, Jake Ross here, producer in Austin. The wheels are in motion on several features here in Austin. Let's dive right in and see what's been going on this month. Right now, the Austin design team is completely focused on things related to 3 or near-term goals. The core tasks we've been working on mostly are state machines for the first

2. 00:01:03

   few NPCs we'll be implementing, state machines and organization of animation assets for the mission givers like Miles Ehart and the nav beacon system. State machines are how we not only visualize how the NPC will be behave, but also informs the animation team when and where our animations need to transition between each other. We hand off these state machines to the animators who then improve the behavior or give back with feedback. Not only does this drive the animations we need, but it also guides our NPC's behavior setup in subsumption. The nav beacon system will allow players to create their own roads throughout a given star system. These can be used both in the vastness of space as well as on planetary surfaces. Beacons are physicalized objects that are deployed from ships and give players visual markers to lock onto for quantum travel. They also point to and fly towards when

3. 00:01:52

   used on a planettoid surface. Players will be able to grant use access to others as well as hack another person's beacon. Both allows you to use someone else's nav beacon. Over time, beacons need to be serviced by the owner, restoring their energy source. Finally, because they're physical objects, you will not only be able to find them, but destroy someone else's beacon if you so desire, which should make for some fun gameplay decisions. Do I hack it and use it myself? Do I destroy it so others can't travel? Do I sell it to pirates as a potential ambush point? There are lots of opportunities with this new system. We're excited to work on it and see it progress in the future. Finally, Miles Ehart development is currently in progress. We have been organizing this mission giver assets which are being polished by the animation team in our derby office, creating his state machine, and getting his initial behavior set up and running his subsumption. Ehart will be unlocked to

4. 00:02:39

   the players by accumulating reputation with him earned by completing other available missions. Once unlocked, you can visit him for a wide variety of missions. The new mission manager will drive his selection, but you'll be able to choose from anything he currently has available. We're really looking really looking forward to getting Echart into your hands in 300 release. The PU game director, Tony Zurc, has had his hands full with several things this month, a major part of which is Subumption. As a reminder, subsumption is the datadriven and highly abstracted foundation on which all of the AI and mission logic in Star Citizen is constructed. Tony's finished conversion of the subsumption tech to Linux for integration with our backend services and completed the shopping service for game code to start hooking the new shopping tech into. He's reviewed and directed mission scenarios for 300 as well. Ship artist Josh Coons has been working on the ship lods for

5. 00:03:26

   the Cutless Black. It's a very consuming task uh since our LODs are mostly handmade and the ship he's working on is quite large with many pieces that have to be optimized. In addition to optimizing the mesh, he also reduces the material IDs as he goes down the log chain. This way, the mesh will have less draw calls from a distance and be more efficient. Our server engineers have been working very hard since our last update. We've been providing support for the shopping service, which is a non-defusion service. This service communicates with diffusion and the game systems through our diffusion gateway. The gateway allows external and non-defusion services to communicate with the game as if they were internal diffusion services. Recently, we've been focused on integration of the diffusion code into the primary game development stream. This will be deployed with 3. This was a massive integration with a

6. 00:04:14

   lot of moving parts and has required a large amount of collaboration between server engineering team and devops. The effort has taken a few weeks to get everything moved over, tested, and in a state where it can be deployed. From an internal perspective, we have been working on a service creation tool. This tool provides a simple to use UI allowing engineers to create new services, add or remove or modify components and management and resource control. The output of the tool is a basic service shell and set of source files that are customized for the new service. This is a huge timesaver and allows new engineers to create services without having to worry about any boilerplate work and thus allowing for rapid service development. We have started to add Star Citizen specific extensions to OO. For those who don't know, ooze was is a language written by lead server engineer Jason Elely. It's

7. 00:05:01

   the scripting language that drives diffusion. These extensions expose Star Citizen specific constructs to diffusion, allowing services to provide more intricate support for gameplay features, helping to move the game into a more distributed architecture. Finally, there has been work on the router mesh functionality. The router mesh distributes services over multiple router endpoints and provides redundant communication paths between other services. The mesh will use a technique to isolate high bandwidth services away from lower bandwidth or more critical services. The primary responsibility of the router mesh is to provide a high levels of service availability and performance. This month on the persistent universe animation team, we finished up the two-handed carry animations. This includes standing, crouching, and for zero g standing, we can pick up 25 cm, 50 cm, and 75 cm crates. That is the new standard metric

8. 00:05:50

   for crates in game that you can pick up. We can pick them up from the floor all the way up to a shelf that is 175 cm centimeters high. Crouching, we will be able to pick up 25 and 50 cm crates from the floor all the way to a ledge that is 100 centimeters high. In zero G, we'll be able to grab 25, 50, and 75 cm crates floating around in space. Code and tech has hooked it up so that you can retrieve cargo in zero g, eva back to your ship, stow your acquired loot in your cargo bay. We also did a last minute mocap shoot in our office when animation director Steve Bender stopped by our Austin studio for a visit. Steve put on the mocap suit and ran around like a crazy per person capturing all of our FPS starts and stops for a stocked rifle locomotion set. We also took this opportunity to get Sandy Gardener in the suit and capture some exercise motion for our female characters when they decide to do a workout in our exercise

9. 00:06:37

   usable. In this mocap shoot, we stuffed our lead animator Brian Brew in the suit as well because we just about killed Steve Bender on the first day and captured needed animations for the crouching carry animations. We are starting work on some specific NPCs so that you can go into a pub, get a drink from an NPC bartender or go into a shop and buy things from an NPC shopkeeper. Animation is working closely with design to make this experience a fun uh experience for all. On the ship animation team, we have been continuing our improvements to the cockpit experience. Working with the UK studios design and programming team, we are in the midst of updating our GeForce blend space poses utilizing a low pass filter for smoother, smarter camera motion, as well as adjusting the cockpit geometry to allow for button presses. In addition to this, we are creating the system that will allow us to make coms calls within the ships during flight. This month, the DevOps team has been busy optimizing the

10. 00:07:26

    build and publishing systems. The game builds are growing rapidly as content continues to pour in for 3. So, we're constantly tweaking and tuning to keep up with the demands of the dev team. Uh, Amed and his team have been collecting feedback on network performance from our three locations and comparing that to internal data so that we can optimize network performance wherever possible. This is an ongoing task, but we've already found some good opportunities for improvement in this area. For May, the Austin QA has been working heavily on regression of bugs, particularly on a massive sweep through our open bugs to see what items are still valid given the new systems and tech coming online for both the PU and Squad 42. This allowed us to eliminate a considerable number of bugs before they ever reached development, saving our busy busy developer cohorts time they would have spent investigating issues that no longer were occurring on the latest

11. 00:08:14

    build. Major testing items for our groups included actor serialization, multi-threaded resource containers, and network transport queue for the engineering teams. We continued testing the moons in the Stanton system for any potential issues such as collision and performance testing. New vehicles, ships, and FPS items came online throughout the month, including the Bearing P8 SCMG that we were very excited to play with. In addition to testing the continued item 2 implementation, all of which have kept our Arena Commander and Star Marine testers very busy. On the new system front, we've been working very hard testing the new procedural breathing and stamina system, as well as a new air traffic controller. Other projects we've been assisting with have included testing some updates to our current game launcher, primarily bug fixes to our players, but also a few quality of life fixes. We continue providing additional support for the animation groups here in

12. 00:09:01

    Austin, including mocap file cleanup, supporting setup and tearown for pickup shoots, and in-game video captures for final reviews. Regular editor and engine testing is continued as well with Austin QA completing regular smokes of the subsumption editor, procedural planet tools, as well as our normal editor testing. The player relations team has been extremely busy preparing for upcoming 300 work as well. The biggest item that players will see is the new player experience that will ultimately go on the website and will match the new content coming up in the game. They'll also be adding to the Evocati ranks in the coming weeks and are excited to announce that they'll be adding headcount in Austin, Manchester, and Frankfurt. We're getting more and more excited the closer we get to 3 release. We have a ton of new content we're pumped to show off and get into your hands. Thanks for everything you do. Keep it up. We'll see you around. Hi guys. Here's your Star

13. 00:09:49

    Citizen platform update. But to start somebody you haven't seen in a while. We're going to see Benjamin Farell to talk to us about other projects here at Turbinant. Hi, I'm Benjamin Farardell and this is the Turbulent monthly update for the month of May. This month, we've been working really hard on redesigning a huge section of their RSI website. What we're designing now is a consolidated view of everything that's currently playable, what you can currently do in the game, and that will hopefully make it more understandable and more usable for any newcomers and existing players, too. Our UX and content teams were came forwards to us with a lot of examples that we took great inspirations from about websites that felt particularly immersive of games that were making it easy to get into their universe, their backstory, and get involved into what the game is without actually having played it yet.

14. 00:10:38

    And that's what we're going for now, so that people can can come and and know what Star Citizen is about even before engaging the rest of the community. What those websites has have in common is that they make high use of a clean, streamlined design and images and videos that make you feel like you're becoming into the game as you watch them. So that's the inspiration that we took from there is we're making it all about making Star Citizen shine in the website by showing and showcasing what it's accomplished so far, which is fantastic. We feel like we have more than enough material to make a compelling view of what starters is now even that far into the process. We also took a lot of inspiration from what the community has been producing. We've witnessed a lot of incredible productions from members of community videos produced in the

15. 00:11:25

    director mode but also tutorial videos and and streams uh the whole setups that people have when they broadcast about Star Citizen. And that became a a great uh input in the design that we put up for for Star Citizen. a new module we're implementing for this new release of RSI is a new player guide, uh, which is going to be a a new module where that covers all the basics you need to know, not necessarily if you want to get good at Star Citizen from the get- go, but if you just want to learn how to start playing the game. So, we're working really closely with all of the CIG teams uh in player relations, QA, and uh and marketing so that we can cover all the basis of everything that new players need to know when they when they want to start playing the game. And we're designing this as a modular interface that we can make evolve as new patches

16. 00:12:13

    come in and new features come in so that it can become a staple feature of the website when you want to point to something that's currently playable in the game. So keep your eyes open for this new and redesigned Star Citizen website coming in this summer. So in the past few weeks, we've been hard at work reviewing the first two months of live operations for our Spectrum Alpha. We've been reading all of your feedback compiling all of it. We've begun already begun some work, but in this release that we're preparing now called 035, we've got major updates and major changes to the system which I think you guys will like. The first one is we're transferring the view preference for uh threads. So instead of somebody creating a new thread and choosing which type of discussion, it's going to be fully managed by view mode which the viewer can decide. Uh so you'll be able to set

17. 00:13:01

    a global setting to decide if you want to see threads as nested or chronological or we call them classic now. Or you can also per thread decide that you want to see this thread as chronological but this one is nested. And so we we're going to process your choice across sessions. And so this basically becomes a completely user preference. And so hopefully uh guys who prefer chronological threads versus nested will be able to just see chronological across the board. And hopefully you guys would like that. Uh we've also done major work in this release to try to unify what we call the red state which is if you've read a thread or a reply already. And so now we've on we've changed the UI so that we display a yellow dot for items that you have not read. So this will reflect in your sidebar on the left, but also on threads and on every reply within the

18. 00:13:49

    thread. And so this will help nested uh thread users who want to know if they've already read a reply or not through the tree. Even if it's non chronological, you'll be able to see I haven't seen this reply or I've seen it before. One major change that we're also implementing is uh related to tags. So, we're changing the uh way we've we're now surfacing uh tags on the community index so that you'll be able to jump directly from the list of channels to a tag straight up, which is basically more functional as a sub form. Uh the tags are also now me can be set to mandatory. So, when you create a channel, you can say this channel requires tags or not. Uh they're browsable from the community index and from the channel directly. They're also bookmarkable. So you can bookmark a tag within a channel if

19. 00:14:36

    that's what you're interested in. Uh and uh there also now these bookmarks uh uh which are now and this applies for every bookmark. Every bookmark are now renameable. So you can go in your manage bookmarks and change the name of your bookmarks uh to something that you want to organize your sidebar the way you want. And so this applies to tags, but also to every other type of bookmarks that you'll store in the sidebar. 035 also comes with major changes to search. We're adding new filters that will allow you to filter by author and by role. So you'll be able to search by for all staff posts for example or search for every post from this user uh within the community or all communities you have access to. So this will make search more viable. It's also now accessible on mobile which it wasn't it wasn't before. So now if you're on a mobile screen uh the search options will display and

20. 00:15:23

    you'll be able to use them. 035 also comes with a lot of work behind the scenes where we've been looking at usage patterns over the past few weeks and there is a lot of performance updates we've done so spectrum should load faster for you guys. The uh thread listing should be at least 10 times faster than it was before. We've added a lot of server side reporting so we can track track client crashes and basically try to preempt fixing stuff that you guys uh without having you guys report them. It's all automated for us. So that's really great. So this is it for 035. So, a lot of changes coming up. We're hoping that by the time you see this, it's already live. Uh, it'll be a matter of days if that's not the case. So, see you guys on 035. In terms of 036, which is our next shorter term version, we're working on the revamping the mini profiles to display more information about you guys. Uh, make

21. 00:16:11

    have a bunch of actions available on the mini profiles. We're going to more of a columnar display. We'll be able to jump from the mini profile to a user's post and take actions specifically on a user. This is where we're going to add later on, not in this release, but later on, all the functions for friendsness, friends system, blocking, stuff like will go in that in that area. The major feature for 036 is the availability of custom roles. And so we want orgs to be able to add additional roles to their org, assign roles to their members, uh, all done through the mini profile, uh, directly on Spectrum. So that's going to be a big thing for 036. There's also work going on in the editor for the forms to allow inline images, link formatting, and more formatting options. Uh, this is turning out to be a bigger project than we thought, but we're hoping that it's going to be ready in

22. 00:16:58

    036 for you guys to use straight on the live site. There's also uh group private messaging that we're trying to achieve in 036, though this is more looking like a 37 feature, but we're still working on it. And of course, all of the background uh work related to voice transmission uh and game integration is also ongoing. But, you know, this is basically what we're looking for in Spectrum for 036. It's great to see subsumption taking another step forward. The system is the bedrock for so much of the game. So, it's exciting to see it all coming online. Yeah. And the PU game director, Tony Zurovic, who is creating the tech, is in Los Angeles this week, working with the team on subsumption and many other important features. Another new system that you've heard mentioned in previous studio updates is item 2.0. It lays the groundwork for swappable ship

23. 00:17:46

    components and so much more. Once fully implemented, item 2.0 will provide the framework for many cool features to come. Yeah. And although converting ships to this technology is well underway, we thought it'd be good to explain the facets of this system and highlight just how important item 2.0 will be to improving the overall Star Citizen experience. Hi, I'm Kirk Toé and I'm the lead technical designer here at Glenoparium Games. My name is Mark Ain. I'm a senior gameplay programmer here and I'm also known as the bug smasherish. Yes, it's that tacky, but it's great. Hi, my name is Ashram Kaine. I'm a producer here in Los Angeles working primarily with the engineering and tech

24. 00:18:33

    design team. So part of my job is working with engineering and tech design to get all of our ships converted over to a new framework, our item 2.0 framework, which is going to support all of the multitude of features that we want to have in our ships moving forward in the game. The reason we had to do this conversion to get all these ships into this new framework was to support the kind of features we wanted in the long term. In the case of ships that already existed, uh we do need to update uh various components, the seats, the way we enter the ships um and the way we interact with the items and the various systems on the ships so we can um make them work with the new interaction system. In the cases of new systems, um we need to implement them in in

25. 00:19:23

    accordance to that new architecture. It's easy enough to make a spaceship and put it in the game and not that easy. But compared to what we're trying to do, it's a walk in the park. But when you want to have a spaceship with modular components that's upgradable, swappable, changeable, damageable, when you want to have ships that die because components fail inside or ships that explode because of damage done to important internal pieces, that's not something you can do with a out ofthe-box system. That's something you have to build from scratch. Honestly, this conversion is involving every single team in the entire studio. There is not a team that is not part of this and I mean everybody from marketing all the way to the physics guys in Germany who are having to be part of this because this conversion touches everything. It's primarily being driven by LA engineering and LA tech design through Mark Avent

26. 00:20:13

    and Kirk Toé. So as you guys know now um the current release we're using the 1.0 0 infrastructure and we've used that since basically we released the dog fight module way back in during the dinosaur era. That system worked really really great for our single sheeter ships where you know you used to have a guy sitting here maybe one co-pilot but when you started having these much more big complex ships and you're just adding a bunch of items onto there it became a big gigantic behemoth to maintain our item infrastructure. Uh the biggest issue we had is when we have our ship, he knows of all of the items that are attached onto here. And when you have a guy right here, he gets all these call backs and events saying, "Hey, this was added, this was added, and you have control, you have control." And we have

27. 00:21:01

    another seat, since that event system was basically on the main route of the ship, he would get those same events. So if you had five um seats, they would get events from everybody else and then every other seat had to figure out who had what control and where. And then you had a more complex thing because then you had the visor or the HUD or the UI, you had the AI, you had the different modules all listening for these events trying to figure out who had control when, how, and then when you add in multiplayer where this seat could come before the vehicle and then this item could come and then we had to reconnect it. connect this here. Anything could come in in any order. It just became pure chaos trying to make sure everything worked in a nice coherent manner. As you can tell, we had a lot of fun bugs and it's also not maintainable,

28. 00:21:49

    especially since um each of those items were their very own special cupcakes where we had all this logic built in for like the weapons. And if we want to use a single part of it, like say we had another type of weapon that didn't shoot projectiles, it shot some we want to shoot M50s out. You had to basically copy pasta that logic into that other item. And then you have another set of bugs cuz you have to maintain two sets of logics blocks and it this was the whole item 1.0 infrastructure. So it just became a mess. Going back to that big logical block, we want to take each of those bits of logic and turn them into these bits of lo um we call them components. This thing handles geometry. We take it out, stick it in a component that handles geometry. This thing handle

29. 00:22:37

    physics. Take that out, stick it into a component has physics. The idea was we shrink this item into nothing but its bare bones and it's just a list of logical blocks or components that define what the item actually did. So in theory, I could take this piece, put it here, take this piece, put it here, and now I have a whole new weapon or I have a whole new um logical item that we wanted to do. And that was the very bare bones item 2.0 or infrastructure. So the item 2.0 system comes as a lot of different components to make ships uh fit within our new architecture. The items that we're improving uh the gameplay functionality on are systems like power uh cooling um the shield system utilizing our new room system uh

30. 00:23:26

    when you turn power off to a particular room all the lights that are in there uh will automatically turn off or when you uh open up a door that's connected to that room entity uh we can suck the atmosphere out. Being able to do things like set a door lock in between those two rooms so that we can simply uh connect those two rooms utilizing that airlock and it knows the states of the atmospheric conditions in within those rooms and can react accordingly when you open that airlock. Implementing new um ways in which atmospheric flight affects your fuel consumption. how those thrusters actually utilize power and fuel to fly. Better systems for functionality such as quantum drives. We're able to balance the fuel consumption that ships will take to uh go from one um nav point to another.

31. 00:24:16

    Therefore, we can make uh some of the ships that are larger and perhaps have a larger quantum fuel tank be able to traverse make those trips uh without running out of fuel where some of the smaller ships maybe they can't and they must refuel before they can take that next jump. These are all parameters that we're updating and making it um a little bit uh more transparent for the player to understand what's going on with them. So, now that you have this control, it's cool. And the problem is how do you interact with this control? And that's where we bring in this other set of technologies that we have which is the interaction system. We had an old interaction system that hit a button and did something. But now we have much more contextual system where you we could register these things called interactions on interaction points. And what that allows us to do is we could

32. 00:25:04

    build a list of these things and then the code gets some callbacks. And if I look at say a door, I could get two interactions maybe open and close and then code could get those and figure out what to do with them using this thing with the control thing is basically we have this thing called a seat access with seats. Before uh we have this ship called a a constellation and we have these two turrets and the thing about the turrets is when one came down it had to block the other one but we couldn't do that on the old one. So, we kind of had a um you could actually do this now where the existing system if you use the top turret, you could use the bottom turret and you can see the animations messing with each other because these two seats don't know about each other. But now we have this thing called a seat access that is aware of these two. It's basically the state machine for these two things. So, if you want to get into

33. 00:25:52

    either one, it will block the other one. And we do that by using the interaction system. Uh the top one has an interaction to enter exit. The bottom one has an interaction to enter exit. and there's a invisible like physics box, but you'll just see a glowing thing saying I can enter this turret or enter this turret. Walk up to it and you see you see actual floating text saying I want to enter the top, I want to enter the bottom. So when I enter the top, the code will get the call back saying I should begin to move this one down and the seat axis will be all right since this guy is moving down I could block this one. So now if someone walks up as you're entering this guy, he won't get any interactions because he can't enter this one because he's blocked until this guy goes up to the top and then now that interaction will pop up saying, "Hey, you can enter now." This isn't simply an engineering task or a design task. We're

34. 00:26:40

    having to revisit animations. We're trying to revisit entire structures inside how ships are built from the ground up, which means we're having to reconsider how doors, simple doors are implemented on the interior and exterior of the ship. And that's art time. That's time that the artists and modelers have to be involved in. And part of that means we're also having to do rework on some of the damage materials, the UV2s that we use for our damage states. Literally, every team, one of the biggest impacts is things you wouldn't think about like the prop team, because we're having to do things like build our power components and our modular power supplies. That means we need a prop for that. And that prop needs to be built by somebody. So all of the sudden things that you never really saw on the ships were kind of there. I mean, they were a little bit hidden. they're suddenly going to be there. And as we start rolling out this technology more and more and implementing it through later

35. 00:27:28

    iterations in the ships, you're going to be able to see these components inside the ships. And that means we need to have those components to show. One awesome example of moving um one of the items to this new whole item 2.0 centric system on the vehicles is we have this uh these lights on their spaceship. And with our um 1.0 know our legacy ships. We had an item that attached to an item port and it would create these things called um basically render nodes all along the ship and it would look cool. I could see the the lights on. Problem is if I get out of the ship or um the wing blows off, the lights are technically still there. We did cheat and listen for when something blew off, but power doesn't do anything. Um you can't control them. They're always kind of on.

36. 00:28:17

    So, what we did for 2.0 is um we have this technology called an object container in the inside of our ships. It's basically a mini level that we could insert into our ships. So, we already have that. We're using that for our item 1.0. And what we want to do is since designers are placing a bunch of stuff like beds and all that stuff in there, we wanted them to use that same technology to move have lights those same lights that were defined before. So we basically tossed the old system allowed designers to create um lights and using this thing called um light groups. You could designate that these lights are basically you know the interior. These lights are the exterior you had a way of grouping them. When you export that level they create these things called item ports. And we attach these lights on the item ports. And

37. 00:29:04

    since they're on item ports the control manager now knows about them. And since they're an item and they have an item port, they could get they could get power, they could get control. So now they register to this thing called the control manager and they register to the um item ports and they could actually request to get power. They could request to get heat and if they don't get it, we could turn off the lights. We could um do anything we want. And we took it a step further. We have this thing called a a light controller. It registers to the control manager and those lights register to that. And if a user sits down in the seat and you get into control manager, I can hit a button to turn off those lights that are inside of the ship that normally don't spawn in. It's not something that we can just sit down and say, "We're going to do this

38. 00:29:51

    ship and then this ship and then this ship because it doesn't work like that. Each of these systems is so deeply interconnected that when we encounter a bug on one ship, we just can't stop and wait for that bug to be fixed. We have to move on to another ship that we can work on. We're having to work so quickly on so many different complicated pieces and track all of the bugs that the real challenge is being able to predict where we're going to be. And the whole point of scheduling isn't isn't to make a plan of how to get someplace. That's the easy part. It's being able to show where we think we're going to be based on the work that we're doing and the work that we've done. This work has never been done before. Nothing like this has ever been done before. The other difficult issue is that uh we do need to go in and do an entire retrofit of every single ship from uh from step one. Some of those steps include converting the ship

39. 00:30:40

    seats to the new system. Uh creating a new dashboard element which is the part of the ship uh when you're seated in the cockpit that you interact with so that we can have visual interactions that come up that allow you to turn power on, engines on and off, uh set the ship to flight ready. These are interactions that occurred automatically before. Now, we want to give the player a little bit more control to uh give them a more visceral feel when they're in the cockpit. This will also be true for other stations when you're seated. When we get an engineering station going, uh we'll allow you to um aim in a certain direction, hit a button, this does this does a particular task, then utilize the screen so that you can divert power, do repairing, uh swap out uh individual items on the ship, etc. So in some cases we take uh the old assets and then

40. 00:31:27

    update uh their the code that we use to implement them. In other cases we uh update the asset itself so that it will um adhere to the new code standard that we need. But in most cases we'll do both. So, we'll take the um the dashboard out of the ship geometry, make it its own item, and then set up the interaction points so that when the player does get seated in the seat, they'll have a uh an indication of where to look at to do these interactions. Coordinating all of these teams is a real challenge because we're talking about 400 people who need to be aware of what's happening. And the hardest part in coordinating all of these teams isn't necessarily tracking the tasks or the individual work on any given person, but making sure that all the teams are aware of what's happening and what's coming down the pipeline. We don't want to break something that's going to then

41. 00:32:15

    prevent one of the teams from being able to work. A good example of this I can give is the AI team. We're having to be very conscious of delivering components and pieces of this feature in time for AI have to have the opportunity to implement and integrate that with their system so that you know we can have people inside these ships and they can fly around and interact in all the different game modes. This is particularly vexing when it gets to more complicated parts of the interaction system because these ships have a lot of interactions. And so making sure that all the interactions are functioning is sort of the the key point to making sure we can interface with these ships as we change and integrate them into item system 2.0. But then there's the tertiary stuff that you don't really think about when you're considering doing all of this stuff. How does this interface with stations? How's this going to work when you have to go buy

42. 00:33:02

    and sell and be cognizant of all of the items inside your ship inventory, your personal inventory, your player inventory, and then where do we want to go with that? How does the economy interface with all of these other pieces? So, making sure that everyone is cognizant of the long-term goals as well as present in what we're trying to achieve for 30 and 31, that's the biggest challenge. There will be some uh balance tweaks that the player will notice which will incentivize the player to uh explore upgraded items for example especially in ships because we are utilizing the new interaction system. It's the first thing that they'll notice when they walk up to a ship and when they get sat down and look at the cockpit uh there'll be a noticeable difference. Gameplay is one of the most interesting challenges in this

43. 00:33:50

    conversion because at once we need to maintain the gameplay that the community loves and the balance and hard work that the balance designers have implemented on these ships. On the other hand, we have to implement a whole new way of thinking about the ships and the ship balance. We have to implement new ways for the thrusters to function. We're rebuilding parts of IFCS and integrating IFCS into the item component system. And that means that we have to reconsider that balance as well. Suddenly, a fighter is no longer just a fighter when you have to worry about managing your shields and your power supply and your weapon groups really in real time. With item system 2.0, we're going to have a lot more capacity for multi crew ships as well. It's no longer going to be, oh, I'm riding around in your Caterpillar. It's going to be I need somebody on the shields and the power at the engineering station to manage this as I'm in a heavy

44. 00:34:39

    fight. One of the big things is probably performance. um with this componentized system each of these components um so if we have a seat it could have geometry could have physics it could have a bunch of other logic each one could have their own update but they could update on this thing called a batch update which means during a specific time in the frame we basically spawn a bunch of threads and say you're updating you're updating you're updating you're updating you're updating we may do like a couple thousand at one go and then do another thousand and then do another thousand so that um frame update went from like this to like this and we could do a lot more. One of the biggest improvements we're hoping is going to be with uh John Pritchette's um physics flight controller, the IFCS. We just actually

45. 00:35:27

    recently moved his update block from um it kind of a main thread physics call and it was stalling physics and taking that out and moving into these batch updates and it will still act the same. just now spawned off into one of those threads. So, we're pretty much utilizing multicore systems at this point. This is one of the largest things I've ever worked on and it singularly can be considered one of the largest thing that I think has ever been done in a game, particularly a multiplayer game, because this is a system that I mean, think about the multiplayer games that are out there. I mean, you might have a mount in World of Warcraft or Elder Scrolls Online or a house, but you don't fly your house around. You don't have 20 houses that you fly around and you have

46. 00:36:16

    to worry about them being damaged and the states that they're all going to be in and how other players are going to interface with them. And that's something we're having to consider when it we talk about item system. Item system is more than just um the things in a ship or the things on a space station or the things on a planet. It's an entire framework for objects and entities in the game persistence being persisted and having actual relevant information to be part of that player experience. I think players are going to start seeing an incredible versatility and power to this technology. So that was some of the big architectural changes we want to do. But of course changing something huge is going to have some big big repercussions. So stay tuned. Next time I'm going to tell you about some horrible, horrible fun that

47. 00:37:04

    we had to do when transitioning from A to B. Thanks, guys. Converting the game to item 2.0 is really no small task, but it's an essential one to building the best damn space sim ever. Yes, and it'll add a lot of versatility to ships and let players interact with them like never before. It should have a significant impact on gameplay. And that's it for today's episode. Please join us tomorrow for Happy Hour Friday at 12:00 p.m. Pacific as Mark Aen and Ashan Kaine join us to talk more about item 2.0. And thanks to all of our subscribers who make shows like ATV and Happy Hour possible. And thank you to all of our backers. We really couldn't build a game with hundreds of swappable ship components without your support. And that's all for this week. We will see you around the verse.

48. 00:37:59

    [Music] Thank you for watching. So, if you want to keep up with the latest and greatest in Star Citizen and Squadron 42's development, please follow us on our social media channels. See you soon.