Star Citizen Live: All About Development

6 September 201901:03:481,478 linesWatch on YouTube ↗

Transcript — every row opens the video at that moment

00:00:06
[Music] hi everybody welcome to another edition of star citizen live all about development I'm your host Jared Huckaby and on today's show we are going to be answering questions about a broad spectrum of topics related to star citizens overall development our reefs are newly stat our new staggered development initiative and of course changes to the roadmap joining us on the show this week for the first time in quite a while the three amigos are back together again Todd and Brian how are you guys doing good mi Chevy Chase or Steve Martin I prefer Steve Martin since I'm great I think with our dynamic you are Steve Martin and Brian is definitely Chevy Chase and that leaves me to be Martin Short which is not ideal but it makes sense no but
00:00:55
in this situation I like the 4c I think of myself as a see Martin but in this dynamic I'm the Martin shirt so if you if you haven't watched star season live before this is where we take questions from our community and answer them live on Twitch and if you it's been a while since we've seen Todd and Brian so before we dump jump right into the questions let's take a few minutes tell folks who you are and what you do for star citizen what sir my name is Brian chambers I am the vice president of development I work with Chris and the directors and really closely with Aaron Roberts just make sure we're on track set priorities where we're going and make sure that we we deliver what we need when we need i'm
00:01:45
todd papy star citizen live game director so that means i focus on the releases and then I also work with Chris and the other directors Brian and Erin and make sure that's the vision and that's we set is basically passed along to the team and make sure that everything is at the quality level that we want to hit now I should say that later on in the show we're going to be joined by a third member Chad McKinney we had a couple questions come in over the week about server-side OCS and we want to be able to address some of those and Chad is our person to do that he is actually in a meeting right now with chris roberts and others about server-side OCS so when he's done with that he's going to come he's going to join us here next to the table there so we're gonna put those questions off
00:02:33
until he can join us and just jump right into it right off the bat why did we decide to switch to stagger development now well obviously for three oh we were very feature driven and what that what happened was things just kind of kept on coming in and the date just kind of kept in getting pushback obviously when we switched over to 3 1 3 2 then we from there we were going through and setting up a certain cadence you know that we're we're we were supposed to be more date driven and that was working for us for 3 1 through 3 4 however some of the features that were coming in were still
00:03:21
rough and Rufford that then we would like obviously for 3 5 3 6 we didn't release on time and a lot of the features just kept on coming in later and later and later so with 3 7 3 8 we've really started to move over to this where for depending on the size of the future but for most of our features it takes longer than basically three Sprint's to get it working so that's six weeks that were we're talking about and getting it to be the quality level that we want it to be so in this case for example let's say mining you know our first iteration of mining that took us you know it took
00:04:10
that team six months to basically deliver everything that needed to be done however you know other teams that are smaller or or depending on again the size of feature they'll be able to do those in the smaller time frame how we were just noticing that and we were feeling an internally of this cycle that we were getting into as well as the product that we were putting out was it hitting our mark and basically the teams were frustrated we were frustrated and so we started talking about it internally how we can adjust things and and try this out and see how it works now the goal for it obviously is that we have what we call a go/no-go meeting
00:05:00
kind of in the middle of the quarter and at that point in time that is the directors including Chris and Erin and Brian taking a look at the features and saying yeah this is good enough this hits the quality bar and what should should be happening is basically all future work is done on that thing it's just minor bugs at at that point time then from there that gets dropped into the stream and it's good it's ready for release before we weren't having that some teams were some teams weren't so therefore some of the quality level you know was was was there and some of it wasn't and but so this allows us to actually you know make sure that there's that gate
00:05:50
you know of quality level and this also gets the teams a chance to say okay yeah you know for some reason whether this whether there was a vacation or somebody got sick or this feature was more complicated you know there's a lot of times when we put things together and we have our best estimate of hey this is how long it will take and every now and then we hit that and every now and then it goes a little bit longer but this allows us to look at the future and make sure that it's hitting that quality level and then gives us basically six weeks to make sure that we're hitting in evocati we're hitting basically wave one at a good time or making sure that's the quality of light bugs that we have are
00:06:38
getting worked on and getting fixed and I you know features aren't getting dropped in like three weeks before we go live yeah I guess an important part of that one too for me is that it's it's then you know our releases are solid on being date driven yeah that if a feature is not ready it's not ready we're not gonna hold up that that quarter release because of a feature that kind of does everyone a disservice so we we have these meetings where we judge we look as Todd says we go you know what this is ready awesome let's go or no it's not it needs more time that unfortunately will miss that release but we think it's more important just to keep the cadence go under the releases it means the team immediately is addressing those issues
00:07:26
without working on bugs while they're in live and so on so it reduces some complexity in our end as well yeah I think one of the hallmarks of star citizen's development is the constant iteration and re-evaluation of our processes I mean we want to we want to get better at making the game just as we want the game itself to get better so these kind of changes are in our DNA and have this very beginning yeah but it's I mean the other thing that I think people need to understand is normal game development it's you just run you run until the very end and then you get to alpha and then you look at it and you're like oh I created a huge mess now I got to clean it up and I gotta clean up before beta and in when beta it happens then you're like okay now I got to
00:08:15
really polish and tune the game and you know how this iteration aspect obviously having the live product and having the live product since almost the very beginning of the this this product creates challenges that you know sometimes you're used to sometimes you're not and you know it's it is a learning process for all of us I mean he's been making games for a long time I've been making games for a long time this is a very unique beast you know of a game challenges and opportunities it's definitely a double-edged sword it's there there there are things that we have to account for but there's also a lot of benefits to running that live environment the ability to get back that instant tangible feedback you know during the development process that can
00:09:02
guide us you know each step along the way so I would completely completely agree with that you know it but at the same time it does sometimes it does feel like you're building a railroad track while you're in the Train you know on it and but but at the same time you know like you said getting that feedback does help us out a lot absolutely all right next question can we focus in the words of our backers can we focus less on ships and more on profession mechanics and gameplay also any updates on future professions so with ships well I mean we have ships that we are building you know and that's we've basically done concept
00:09:52
sales for so and so forth so that pipeline is just running as far as what we would consider careers or professions I view those and I think all of our system designers view it as a collection of different activities and what we're doing is we're really building activities and then once those activities come together then you have something that would be considered a profession so from a PU standpoint we we have a little we have our priorities that that we need to do and then we have our what we consider as 42 priorities and from our development standpoint as 42 is the number one priority so most of
00:10:43
the team and most of the resources are focused towards that and then basically we have what we would consider solely as 42 and then we have kind of as 42 pu and then we have pu only so if those professions or things are not needed in s 42 they're not going to get as much love so for right now yeah yeah and the media you know until s 42 is done and ready to go then at that point time you know the floodgates will open up more on the PU and then we'll be able to to push harder on those things but we do have resources that are solely focused on PU and solely you know working on aspects of that meaning you know for example mining or
00:11:31
basically what we consider resource resource gathering so whether that be mining salvaging harvesting we have a team that's focused on that stuff then we have you know teams that have been focused more towards like cargo and those things so and but that team has been also working on scanning and other aspects that we consider very integral in to both projects so the priority is for them to finish that first before they move back on to like cargo right next question from the chat is why sorry I think the last thing was what are the professions that we're looking at status of professions well I mean mining is in progress and so we we've done basically
00:12:23
obviously prospector money we we will need to do a Ryan mining once once that starts getting ready we've done or we're working on FPS money from there we've got obviously Salvage but again all of those to me are kind of resource gatherers then we have very early stages for kind of trading I wouldn't really consider hauling butt but basically trading and then there's piracy there's security or basically mercenary and bounty hunting and those those would be the kind of the first five that I would be focusing on and then there would be kind of secondary ones which would be like refueling rearming that kind of stuff okay from the chat why are long-standing bugs not being addressed
00:13:13
in every patch some some bugs we've been waiting for over year to fix now so yeah I think it really depends on the bug you know for example I know that there's one in there that drives everybody nuts it drives me nuts as well which is the HUD in particular when you look at certain bright backgrounds right and that is something that's just had it been prioritized because we know that we're redoing the HUD you know we know that we're a Justin it we have had initial talks regarding how to fix that thing and then I at least for me it's it it's what I consider a quality of life experience that that will need to get fixed so ideally I would like it to be
00:14:02
fixed sooner than later however when we look at these kind of the staggered development again if we're building features and minoo and we need to release and let's say this time frame and you're building features up until right here then that gives you less time to actually build or to fix versus if you turn off your features here then that gives you more time to fix these bugs and and work through the quality of life so that's been a point for Aaron and myself and QA to go through and look at the bugs and say ok you know we feel like this is a big one same with working with problem player relations and those get prioritized in what we would consider game play experience aspect and whether that be a moderate or
00:14:51
trivial or order that's just a minor bug versus only focusing on criticals and blockers so we've changed our pipeline to actually get those incorporated into getting fixed throughout the quarter as well so you will start seeing those quality of life issues be slowly resolved pushed with with you and the other guys would probably good three five six weeks now yeah for me it started it started longer yeah about us actually getting him in production talking it through with QA and so on we're really going okay well it's so easy as we kind of go back to the previous question you know building Trax while you're on the train we're moving so fast moving forward and looking at these future things we really need to stop and it was everybody taking
00:15:39
a time and Todd taking initiative to go well there's some bugs that have been there for a while and and you know what we we don't like them either so actually let's take the time take a breath put them on the roadmap or you know put them on people's schedules actually address these things so as Todd said with releases we're slowly whittling away at these things that people have seen for a while but I think with the quality of life stuff when we when we take these in these bugs into account we know internally kind of roughly when we think we're going to be let's say for example the UI we know that we're redoing that ship UI right now so if we have MF debug then we might be looking at saying okay is that going to be fixed based off of
00:16:26
us doing the new M of these or you know or can we live with it for a little while so there's that kind of juggle that we do and so put in a fix that's gonna be overwritten by the new UI ingest we don't want to waste time yeah we want to be as efficient as we possibly can and in it it's just us looking at it saying okay do we really think that let's say in the next six weeks you know do we feel like this is going to get resolved or in the next three months yeah okay then then we won't do it but that's that's kind of the juggling act that we go through every quarter and that that's a big part you touched on this a bit but it's a big part of why this new staggered development initiative is taking place
00:17:13
it's that desserts that desire to have more time to spend addressing bugs and addressing these quality-of-life issues so you know if if all things go as we hope they will this will lead to a overall and improved experience and many of those long-standing bugs being addressed I mean I I've again I've never seen a game developer get it right the first time it's it's through constant iteration and and and so getting allowing us to to work on it more giving in a more polished experience than that that means that just in its nature you're gonna have less bugs you know associated with that feature next question let's see here how much is the actually no we're
00:18:03
gonna wait for a chat on that one so so this one is actually for me it says why was the second pillar talk video not about pillars as this specific term was coined by chris Roberts himself outlined in the citizencon 29:48 road to release presentation is this an indicator that those pillars have been abandoned and replaced by other goals absolutely not when we when we make a pillar talk obviously we've only done two episodes of pillar talks so we are still exploring format in all the different areas that we can touch on with that particular format for the second episode we had Aaron here he was he was visiting the studio at the time and we thought we knew we were moving toward we had already moved towards this staggered development initiative and we thought this was a good opportunity to have him talk about that very candidly with John and Eric and attempt to shed some light
00:18:54
about why we're doing a lot of things that we're covering here today in the future pillar talks will continue to it to address you know pillar issues you know they'll have the pillar directors it's just if it was only the show was only going to be about stuff that was specific to a patch we probably would have called it patch talk so we were two episodes in which were we're exploring the format of this and I think you'll continue to see it develop and iterate just like you see anything else mister I said or something sorry do you hear is my pop yeah I keep hearing your microphone keep hearing my mic pop yeah oh no but I think we're setting up for Chad Chad's just walking in we're hearing Chad at the same time okay okay next question this is an oldie we used to get real design documents in the past that gave us at least a moderate
00:19:42
glimpse of what a mechanic was and how CID is planning on it affecting gameplay can we please go back to getting these deep dive documents into and then they list a variety of professions so we thought I mean we've we've internally changed the way that we two things so a lot of times with those those documents they went into a very very deep detail and I don't think that we've hit all of those details once you know on on what so in internally we have what we consider a tier zero Tier one - you're too kind of thing and also I'm getting chad's I'm getting
00:20:31
feedback come on in Chad good and so that's pretty much where we're at on server server side oh yes thank you for spending your time going through that lengthy a pleasure it makes so much sense huh hi guys we're back obviously some unfortunate technical difficulties that crept in halfway through the show but we think we've got them resolved now so all right we've got Chad here we know he's got to get back so we're gonna do the server-side osseous then we'll go back to the design question so Chad what is the current status of server-side Josias okay yeah so I'm gonna give a long-winded answer here like I said I
00:21:20
wanted to talk about kind of the past and the present in the future as far as this as well as kind of server mushing to start with the engine that we had CryEngine not lumberyard it's it's a level loading based engine mm-hm there's no entity streaming in in the engine like stock you know something made for like you know whatever deathmatch games we years ago now made a large sort of changes to the engine to create something called ugly containers which is our own grouping of entities as well as level data like brushes and things like this planning years in advance where we want to get to both now and in the future so that was a huge push that was a foundational sort of changes we built on top of that with something called net
00:22:07
bind calling in serialize variables which allows us to at be able to replicate and unbind entities from clients and then we deployed last year client-side streaming which now everyone's is familiar with which is taking the object containers at runtime and streaming them in it out so basically unbinding and binding those object containers so that was multi-year effort there in the loan now we're at the point where we're looking at trying to do the same we are definitely going to do the same thing on the server and this is itself a very large effort and it started actually kind of in in full force last year and this is like kind of the present part where we actually had a big summit where we got the programming
00:22:57
disciplines together over in Germany and got a week-long kind of really deep dive on the real specific implementation details for the effort across the entire game including the core engine game code as well as the backend services in order to execute this this is something that we've been working on ever since then so for you know about a year now we've been pushing really hard on server streaming we've made meaningful progress on a very meaningful in fact I just got out of meeting out of a meeting about our progress with Chris Roberts and I would say that the people is pretty positive actually coming out of it so some notable things that have happened so far include we've done a large set of changes to object containers as far as how they load entities we've now
00:23:44
separated things out into what are called dynamic entities and Static entities which gives us a certain amount of information that we need about what is streamable in the universe and what is not so that's that's a big set of changes another thing that we did is we've made something called a star hash service or star engine service rather and star gin services are game services that are written in a way that's similar to game code but can be deployed as a cloud service then we've taken this framework and we've built something using it called the star hash service and star hashes are our version of kind of a coordinate system across the universe similar to geohashes if anyone's familiar with that concept and it gives us a way to partition the universe so that we know how it like accession essentially what sections of
00:24:32
the universe are streaming at any given time in a really efficient way as far as well as storing and looking them up efficiently so we built a service that kind of keeps track of all of what we call entity aggregates inators which was kind of the roots of trees of entity trees so that exists and the star hash service now also in addition has been hooked up to what's called the streaming manager previously the streaming manager was was created and it governs the current client-side binding and unbinding so the client-side streaming of object containers but now we've we've done the work to hook it up to this new star hash service to drive straighting at streaming events to indicate to it that certain aggregates are streamed out or to collect certain aggregate groups and stream them in so that work has been done and now we have kind of a still a
00:25:24
bunch of work to go but what we've done is we've implemented something that is a prototype in-memory persistence cache for server-side streaming to work we basically have to implement global persistence which is obviously a huge set of changes for the game it completely changes the way that the game runs and the way that we spawn and load entities for testing purposes we made this prototype in-memory persistence model which allows us to write out - just like a local cache this state of any one of these like star hashes and sectors whenever we stream them out that is so that we can test game code as such as the mission system or the transit system or service beacons whatever to see what breaks whenever we start
00:26:12
streaming things in and out on the game server obviously we're changing the way the game works quite significantly so we have to revisit the existing game features that we have to make sure that they are broke in and fix them in places that we are so we have done that the next big things for us to do are to implement the the real back-end for persistence which requires if I we are rewriting the way that the game writes out to persistence and the backend the cloud services that handle that data so that they're much better at handling much higher loads you can imagine that the amount of data that we have in the game right now is large ish like we've persist out players and ships items this kind of thing and yeah you make changes to those all the time even your fuel when you're flying around
00:27:00
we're pushing updates to the backend periodically for that when you break something off a ship right all those changes are persisted but now we're looking at the whole universe for all the game servers and the whole game right so there's a much higher volume so we're having to do a really big set of changes on the backend side to be able to handle that load that work is in progress have been in progress and is going to continue after that then the last kind of really big thing and there's a lot of little details in there these are like the big-ticket items the last big thing is is that game code changes that need to happen so we need all the game feature teams to go and revisit their features and make sure that they're not broken which is a big effort it's and it's one of those things where until we really start testing
00:27:48
heavily we won't even know all the work that's going to happen so there's a lot of unknown unknowns and there however so that's good I think so like I said we came out of this meeting and I think we made a lot of good progress this year however one really positive thing that I think we've figured out is that our prototype in-memory persistence model seems to be working fairly well so we're considering phasing out servers streaming Ashley so that it doesn't need the full persistence back in the first time we ship it so we can ship it in such a way that if a server goes down it's state is lost player state is not lost notably so the way the game works right now won't change from the player's perspective it just means that we can add in some of the streaming benefits without yet
00:28:37
having the full persistence benefits which will come once we finished that work so that means that we might see it on a more accelerated release schedule than then we may be anticipated I am NOT seven yes a part of it yeah I'm not giving dates but it's we feel positive about it effectively yeah and so that's servers streaming we've done what server meshing a chance of the last part of the question server meshing is something again there's been a long time in the making I mentioned at the very beginning net bind calling that was itself a prerequisite for server meshing and serialize variables as well we have similar to server streaming began significant planning and work on the
00:29:27
future however server streaming is a harder requirement for server meshing so what that means is that even while we can make progress on it because most of the work for server meshing falls in the network team with some support from the core engine and game code teams where as server streaming is mainly core engine game code back-end services with support by the network team we can do some amount in parallel but we can't deploy it in full until service streaming is done because it's a prerequisite it's how is how it even fundamentally works so it will come after server streaming for sure alright maybe got anyone ahead [Applause] thank you for answering that that was all right let's do a couple follow-ups
00:30:17
that are ten generally are linked to server side OCS before we let you go it's been said that this it's been said before by my Chris and by a Shawn I believe in the past that the servers are kind of at capacity now and we can't add any more larger locations until we had server-side OCS we currently see Microtech on the 3.8 map but not server site OS yes why is that and why is that we'll start there I I can okay I don't know what I can and can't say but we don't want to talk mainly the server streaming we don't want to commit to release that's quite yet for service trimming because like I said things are still in flux there are we can or nothing that we will but we can
00:31:05
consider maybe content swaps if we wanted to add in a new area and take something out to account for them that said no decisions have been made on that if we did end up getting server streaming out sooner than we anticipated even then I'm not sure what the result would be because we have to measure like anytime you implement an optimization is something I say all my programmers on my team you don't know if you've done the job until you've profiled you don't know until you take a measurement and so even whenever we get server streaming done that's not a that's not a guarantee that anything is possible that we can keep increasing the content and we may find that actually the gains aren't as much as we like or maybe we have to push in some other ways so I guess what I would
00:31:53
say is that those details will become more clear as the plan for service trimming gets filled out and as we do more testing and we can actually say with more confidence what the performance behavior is also Microtech and new babitch are still being built you know so no matter what the content team is still working on that still finishing that that's aspect off until we get the LEDs and we get the AI and all the other you know kind of downstream dependencies in there then that will give us an accurate profile that chats talking about you know for us to to say okay yeah it it does really mess up the server you know or you know we need to evaluate and taking something out or everything's okay right and give a little bit more detail to like why we
00:32:42
can't know that quite part of the benefit of server streaming is that it allows us to stream out areas where no players are but there's this kind of fundament unknown about how much that impacts like the actual universe because if your most expensive assets and content areas are the popular lighting zones depending on player habits and where the players go in the game we may find that actually those areas are constantly streamed in their pen effectively to always be streamed in and while yes we do stream out most of the universe it doesn't impact performance that much because most of our performance issues are at these large landing areas what we are more confident that server streaming will do is make it to where detail in the universe can be
00:33:30
added much more confidently so we could do things like atom which we're out poser and what it more tracks dobbs or whatever these locations that even if all 50 players went to a different one if you had a thousand of them well it doesn't matter only 50 of them will get loaded in so there's it gives kind of a upward maximum on kind of the maximum distribution of what can be loaded in but it's these kind of real spiking from a performance point of view Eric like areas in the game where you know what is you know what really happens you know maybe half one gets trimmed out most of the time and that's good enough we just have to measure anything you want to add done no I mean we know that our major landing zones are basically the biggest performance hits you know obviously area
00:34:20
18 and there's a lot of entity count that's there just because of the city planet the way it's built it's it's very different verses like Hurston or other or Microtech so it's we I think we know our problem areas it really is just going through and saying okay you know we have this tech now whether that be an MPC manager as well as you know the other stuff that Chad's been talking about then I think that will really help our our gains on us being able to just push cut into content in and as Ford just specifically to bring it back to why server-side OCS isn't specifically on the roadmap the roadmap we've said this number times before the roadmap is
00:35:06
not a list of things that are in development the roadmap is a list of things of when we think we can deliver them when we feel confident that we can deliver them we all know that that oh it doesn't always work out that way but with something as big and as robust a server-side OCS it touches as many teams as it does it touches as many disap as it does we want it we want to take a little more time be a little more confident be a little more a certain of when we can deliver that before we add it to the roadmap and when we're at that point you will see it at it it's definitely coming I think we proved that we're working on it and then the last related question how much is the implementation of various gameplay loops hindered by the lack of server-side improvements such as server-side OCS server meshing the light will we need to
00:35:55
wait to get those system implemented before we can expect to see more expensive gameplay yeah so there's there's not really gameplay features that rely on server streaming and it goes more the other way where service trimming relies on the gameplay features being written in a way that work with it but it's it's not like you know something needs to be streamed out to to work correctly I mean really what it gives you is just a way to scale better so that we can put more content and more features in but not so much that there's some added behavior or feature that this adds in that now the feature can do something it could not before there I mean there are some interesting things that the tech that we write to create server streaming can enable but it's not serve assuming that does it it's more
00:36:43
that the foundational technologies that we make in order to create server streaming themselves could lend to some interesting future development and this is especially true to give it a perfect example actually global persistence service streaming requires global persistence to work we have to string out an area and store it out yeah but obviously global persistence is something that we really want in the game global persistence doesn't require service learning in fact we can have it without it technically performance wouldn't be good but it's something that we're making that you could think of as a feature that we made along the way in support of service driven yeah it can't be understand it understate it how important it is if we're gonna pull things out and put things back we have to be able to remember where they were
00:37:30
at otherwise the whole the whole thing doesn't work I mean it the only thing like Chad said early the only thing that the game teams will be responsible for is making sure that their game code works with that so obviously if they're doing that then they're not future development you know there they are but they're not you know so it's it's it's it's a necessary evil you know that we we we know that the future teams have to go through in order to get us to this you know the funnel journey and we had the same concept to just with the initial implementation of the object containers yes there was the unknowns were unknown at that point you know so it's a bit of a roll the dice but we got it sorted since we're talking about persistence this next question
00:38:20
would like some clarification on persistence and the the notion of ending wipes whoa whoa what we'll persistent they based want to know if persistence will be the end of wipes no yeah about this I agree with Todd no we so there are some really fundamental reasons why we need to keep wiping the persistence database right now largely because the game is so much in flux right now every release every patch even we're changing the format of the data pretty dramatically at times so you know trying to to kind of ferry this data when it's
00:39:08
it's in such tremendous flocks is I think a fool's errand really our goal is to get to a place where our game features are a little bit more stable a little bit less tumultuous before we flip the switch on top of that we also need a strategy about what we do when we end up changing the formats of data which is absolutely going to happen it's not like we'll stop we just want to get to a point where we're doing it less often and I can say that actually this is not something that has been kind of like forgotten about it's not something that we don't think about we had a few meetings very recently actually to talk about this very specific things that something we care a lot about and I think we have a really good plan about what to do about long term stopping the wipes about how we can go about migrating data and
00:39:59
how we can give importantly the game features so the gameplay programmer is a methodology to do that ferrying of the data whenever they change their infamous however to give a little bit of hope we are also exploring I'm not making any processes promises but we are also exploring potentially ways to persist things more permanently sooner and this would be important thing so for example maybe your ship purchases in the game or your UEC so like things that you think of as like important to your progression in the game we're exploring possibly storing a much more course level data about them across wipes so that we can kind of restore you back to the you know
00:40:49
maybe your ship wasn't in the exact same place but you still have it right you know maybe your gun isn't where you dropped it last but you still have it your you'ii see that you worked for two months on three guns on it's still there and I think that would be a pretty tremendous like higher priority features I mean yes prints listing of everything is important but but you know the the kind of higher priority things that that negatively affect gameplay it sucks when you know the server's crash or when you come back and your ships are gone it's like that being able to bring elements of that online in an iterative nature and I'm specific talking about across white so that like when the next patch comes out you know everybody you know is every patch we're gonna wipe and so they're just there's almost a nihilism right to like what you're doing in the game because at a certain point you know it's all going away
00:41:37
right we want to get away from that so I think it starts with the kind of basics and and seeing what we can do about that and then long-term having a strategy for you know stopping it entirely because I mean obviously with our ship prices in-game and everything like that those are very very long-term goals you know so us you know I I don't want to wipe every every you know quarter but I want us to be able to wipe when we need to wipe you know and so exactly exactly but I do want the players to be able to work towards these longer-term goals you know versus trying to hit it in three months you know so then at that point time we are really testing an economy better to understand you know do we have
00:42:26
these things priced in the in the timeframe that we think you know they should be in you know it should it be a three month earn or six month grind or you know or longer you know and it really depends on you know the telemetry that we're getting back off these things and right now we just don't know we have a we have a best estimate and we have a best guess of you know hey we think it it does this but until we're actually seeing it you know constantly then then we'll adjust from there Thank You Jay you're welcome to hang out so Todd back to where we back to where we left off before Chad joined us we used to we in the past in the past I'm gonna restate
00:43:16
the question a little bit and the past as the project was just starting as there was less game to play less game to build an a quarterly basis we released a series of design documents as the game as development progressed is we were able to move more into content phase and start cranking things out faster we pulled back from releasing those design documents yeah a very common question is will we are will we go back to releasing those design documents publicly and if we're not why III don't think so like I said we have changed the way that we develop things and we're also in a different stage in development you know for this project so a lot of times when we're
00:44:04
let's say talking about law or we're talking about you know features that we're working on right now mining salvaging so and so forth like we we have an internal document of what we want to hit you know and and it I would cut I would call it almost like a one-pager and that has the very high level beats of what we think we can hit for tr0 and what we need to hit for tr0 and the previous documents used to go through like Tier one through zero through four and so there was a lot of things that would happen when we start getting to tier zero and then we release it and then it would change and then that would make one through four completely mute moot and then you know it would just kind of get into this weird cycle so you know I'm more than happy to talk about
00:44:53
you know what we're working on and how we're moving towards and and you know having that back-and-forth with the community and and saying hey here the here are the ideas here the designs that we're going towards but I don't think you know doing these long-long design documents that Tony or myself or or other designers that I worked with you know doing these things I don't think it's worth us putting those out anymore yeah I can add a little bit of programmers perspective I started working at a time years ago when we did you know the more monolithic designs and it's it's an interesting change that we've gone through since then and I think one of one of the big things about when you're when you're as a programmer
00:45:41
throw it onto a feature that has you look at the design document and it's like a you know for volume encyclopedia you have a lot of questions right and and it it was we were more siloed back then so you know design would happen and but then as a programmer you'd be thrown onto something and there'd be kind of a sink up there might be technical considerations that hadn't been fully realized until we started implementing it but then also design questions as we implement it and put it out there and I think as we've moved now into this phase where it's more Future teams and and programming is working much more closely with design and we're more willing to be a little bit more agile with with the design it's to the benefit of everyone I think so both on the programming side it gives us a little
00:46:29
bit earlier view into the design process to give feedback as well as they have a much better source of information as far as what is possible and what is reasonable within a certain amount of time and there's also no kind of there's less expectations about you're looking at that kind of tier-4 and be like ok I have to write this system hitting this really long-term goal that we're not even confident we're gonna do anyways which might color the way that you do the implementation so I think it's a pretty good like set of changes that we've made and yeah the design documents worked against in agile development and it worked against one of the the biggest benefits of star systems that we do have a live game environment where we can put out these iterative things see how they work get feedback folks who watched ISC yesterday may have seen the segment with Dave
00:47:17
Coulson about the changes to proximity with proximity assist to the hover mode that just went out in the Pratt fire in the in the big quarterly patch prior being able to deliver these tr0 things get your feedback adjust what tier one is gonna be put that out get your feedback at just what tier two is gonna be is is incredibly valuable and a long those long detailed design documents would either lock us into something that would that would deny us the feedback of the community or would be what actually happened just most of them were just wildly inaccurate so that pretty much cover it done all right let's see wrote my question what happened to fuel a gameplay feature
00:48:08
that was just removed from from from the roadmap is it coming back yeah we've we just had a meeting with Chris about you know refueling and kind of the the two different aspects that basically John's pillar so the one of the vehicle feature teams will be focusing on in soon with vehicle content and in what we what's basically my pillar basically the the resource gathering would be focusing on so there's the refueling the the fuel scoops you know those those types of things and then my pillar would be focusing on like quantum fuel mining or
00:48:57
harvesting and you know and then that gameplay loop of going through and refining it and then being able to sell it or being able to transfer it to somebody so it will come back on I think you'll see glimpses you know once once we get through some of the the core things that we want those teams to go that they're currently working on because frankly for us 42 it's it's not needed you know so again we have those teams focusing on things that are needed for us 42 if the pu team you know can can work on it then we have those four they have we have them work on it and there's a good chance to plug that going forward with with future major additions to or changes to the roadmap maybe not
00:49:45
with with some of the smaller ones at what's not we're going to start adding a commlink posts and more information about why each each successive change on the roadmap occurs when they do so you to look forward to that in the near future so from our amazing community team alright let's see another road my quote oh why do we not update some features on the roadmap for weeks maybe this is something Brian can speak to its do the complexity honestly and our desire to have everything as accurate as possible it kind of bites us a little publishing the roadmap obviously it's our ability to be transparent on what we're doing shorter progress what we plan to do in the future and and so on our end goal is
00:50:36
to give everyone you know the most update information but we wanted also to be accurate so when step down the road map comes in earlier late its they got a knock-on effect with something else so we take the time we verify what impact the changes had that then takes time then once everything's verified then we go through and we we give an update well we've realized and and we've have had this question come personally even from some members on the team that it seems like too too much of a gap in between so we've already discussed this eternally and and moving forward we're gonna change the frequency of the updates so I think moving forward everyone's gonna see smaller updates happen more frequently some of the backend stuff may not be as a hundred
00:51:25
percent accurate because we haven't gone through the entire process yet but it'll give a clearer glimpse of when we know something moves we should update the map to reflect that so people know what the expectations are and and the flip side if something may you know of Team a finish and we can pull something in sooner we should also update that as well one more question for you Brian I know you've got a train to catch here we're actually past the time when we were supposed to stop but we're gonna push a little bit longer because of our technical issues earlier what's the status what's the current status of our Vulcan implementation Vulcans going good quote-unquote good we actually had we had a summit a little while ago about Vulcan where engine team and everyone got together on the render
00:52:14
inside and said okay what is it that we need to do and and where we add and who's going to take on what it's basically broken down into three stages stage one we we clean up what you the use of the global state and the rendering code that's already been moving forward for probably it's been underway for the last I don't know a few months I would say roughly three four months Stage two is a refactor of the render passes take advantage of new API is current and future GPUs and so on so we can help milk those more that started a couple weeks ago after the summit and that progress is going good the third and final is basically the implementation of the actual Vulcan
00:53:05
beckoned and although on paper it's not officially kicked off yet for that portion parts of the team are already actually starting to work on that and there's some parts the the Vulcan back-end that's already been in there so so progress ultimately is going good with the windows first Tuesday and it's not an all-or-nothing I think as we get some of these changes we could have some stuff come in before the entire Vulcan back ends there I've been told dx11 will already benefit from the first two stages and we should see some performance improvements there and ultimately at the end the the plans are not to just to support you know quote-unquote Vulcan just by straight ports of the render but actually refactor all of the render code so you
00:53:56
can take advantage of everything that Vulcan has to offer so we are running right down to your wire Brian let's see one squeeze one more in then you can feel free to leave good question by the way I'm sorry it'll be mid question we'll just get off him yeah absolutely that's it I once squadron 42 I went squadron 42 is shipped will the puc the majority of developers who were working on squadron 42 move over to the P it was their main focus yes and no portion of the team will will probably roll over the PU fairly quickly I'm hoping there's you know a handful of them that actually take a few days off because they'll be working really hard
00:54:45
but others will actually go directly on a pre-production and some probably forms a production for for chapter 2 and also important note that some some people or some disciplines actually won't be needed all the way to the end for squadron right they're working now but they'll roll off so even before squadron actually officially gets released they'll be able to roll over to PU so as we get further along will constantly evaluate where we are look at the resources see the the status of everything and most likely update the road half accordingly with with you know if that increases our progress on what we can provide for the PU while the rest of team readies themselves to jump on to chapter 2 right and for those of you in
00:55:37
chat who seem surprised that yes I was swatted for you to has been from its Assumption inception perceived as a trilogy so uh next question ship combat this sounds like a Todd question ship combat has several problems it's a Chad question combat ship combat has several issues as stated in the top voted three to six one feedback thread is there a team working on ship combat and can we expect significant tuning in three point seven or three point eight or just soon yes so basically what we did what we did previously in is we'd be working and in particular designers were not allocated programmers were not allocated just to balancing so we have
00:56:27
created a balanced team that from for all quarter they will be focused on balance issues focused on fixing you know things that that just are not working the way that we want them to work so that team just got created and in fact the the main designer on it was focusing on the Vanguard before and building that so and then from there he's he's moving over to really the tuning and polishing of these things so from his standpoint you can see where he'd be working on things and then he would only have a certain amount of time of tuning and and that's that's really how that team was working where they would be building something then they would be able to fit tuning in for about you know three or four weeks and do as
00:57:16
many changes to they possibly could and then go back to building something so they didn't get into the cadence and and as as well as get into balance changes that we want to do for a long term so you'll see stuff in three seven I think you'll see a lot more in three eight three nine kind of thing so with that team they're doing just general tuning passes they're working on basically hip and targeting problems which would include the gimble assist as well as the fixed gun problem turret behaviors making missiles its own mode kind of thing the way the encouraged kind of closer combat and penalize you know combat higher speeds you know kind of accuracy or shield degradation those types of things and
00:58:05
more you know so it we're just gonna be we're gonna be able to tune in testing and iterate on these things at a at a much faster rate than we have before so now that we have these teams just focusing on that then that's you know we filled it and we view this as a core core change and then so getting the ships to feel right getting the combat to feel right you know and having a little bit more of an intimacy you know to the combat versus it being just pixel hunter so that's that's you know our goal and our focus to you know really tighten that up because that is needed for us 42 and and obviously it's needed for PU but it you know at least the the first iteration of s 42 has to have that
00:58:54
right will there be will there be remote logging outside of spaceships with the advent of large and round cave systems as well as continued development of our planetary environments what is our intent to allow players to log out without having to constantly go back to their ship every I well I mean see ours talked about it you know and we talked about an internal lean kind of how we can do it or what we would do you know whether it be a let's say a tent or a sleeping bag or whatever but it's it has been discussed and I do think CR wants to be able have these other ways of locking out that when it just be wouldn't be allocated
00:59:42
just a ship or a landing zone or space station yeah it's something I would add to that is that we have already been working on things that will if we want to go down that route support it two things are one global persistence but then another one is actually the persistent Habs is is work actually that moves in that direction so definitely it's it's something that on the tech side we've been anticipating well most there there's just two more can we can we have the ability to lock onto ground turrets guns do not converge properly if no target is selected this didn't work answer is yes quick answer is yes basically we have it on the u.s. vehicle feature team I don't think he'll
01:00:30
be 437 but I would like to get it done sooner than later and finally before we let you go we just we just added a 4.0 to the end of our road map with it with the with the changes that we made last last week some folks think the logical patch number increase from 3 point 9 would be 3 10 as the 10th major patch for software version 3 with 4.0 on the roadmap should we expect a similarly larger leap and game play experience like the one we saw with 2.0 or with 3.0 I would like to think so I mean I I think obviously with if we get server side OCS you know most the Stanton would be there if not all um
01:01:19
so having a full system there like that's that's me is a huge huge that to me is like a vertical slice you know the this is a galaxy or solar system for us and so obviously we got more to create but you know then that would be hey here's our stake in the ground here if we have all the celestial bodies you know in in this I think that would be huge and would warrant a number change absolutely and for folks who look at the current you know 4.0 and don't see a whole lot there I'm just going to say keep your eyes and ears peeled we've got a bit of time between now and then and just a wait for it stuff what stuff will be added that's
01:02:11
it guys we did it thank you so much for for taking your time at the end of your Friday to hang here with us and be on the show this week it's it's always great when we can have a Todd and Todd and Brian on we haven't been together in a long time so a long and Chad you're always a delight you're always you always just come I agree shot you come on and you just let this this bright ray of technical light his energy like even broke the stream anything you want it you want to just hang out and just goes to show with me from now so yeah so in case you haven't been paying attention to the website or whatever a36 to went live during the
01:03:00
course of the broadcast so that's live now you can start your downloads that's got the 890 jump ready for you Todd Brian chat thank you so much thank you we'll see you Jeremy our pleasure see you choose you thanks for watching for the latest and greatest in star citizen squadron 42 you can subscribe to our Channel or you can check out some of the other shows and you can also head to our website at

Transcript from YouTube’s automatic captions — punctuation and Star Citizen jargon are approximate. The video itself is hosted by CIG on the official channel.