Star Citizen Live: All About Alpha 3.9

It wasn't SOCS that caused it. There was a physics issue. And in fact, there was a physics proxy refactor that was done. And we had that fix ready for like 3.8.2. However, the code base between 3.8 and 3.9 had changed so much, so we deemed it too risky to actually check it in. So we checked it into 3.9. It's in 3.9. Ideally, we think that fixed the root of the problem. And then if we run into any kind of edge-based issues, we'll go in and fix those case by case. But ideally, most of that has been resolved. Now, obviously, there's other factors that play into that, whether that be server lag or even we kind of have some internal lag built in. And we're trying to work those issues out right now to reduce as much latency as possible. We know that it's there. It is a focus of ours because it is a huge quality of life issue that needs to be resolved.

We've talked about the physics refactor a couple times on this show, actually, throughout the last quarter. It's important to remember that while server-side object container streaming was a big tentpole feature of 3.8, it was not the only feature that went in. So when you see issues like that, it's often easy to go, oh, that was the biggest one, that must be it. But in this case, it was the physics refactor.

We've had a couple of factors that we've been dealing with. Obviously, we really saw it on, I think, 3.8.1; that's kind of where it reared its ugly head. And then from there, we tried to do some hotfixes for it. So if it's rearing its ugly head again, then send those things in and we'll get it fixed, because it should just be a hotfix versus having to roll out a whole new patch.

And of course, when he says send it in, we're referring to the Issue Council, which is the thing on RobertsSpaceIndustries.com where you can submit your own bug reports in much the same way our QA team does. And then you can verify the bug reports that have been submitted by other backers and help us find these things and track these things down even faster. Because again, as we've discussed several times, the same resultant bug can be caused by a multitude of underlying features. So just because you see it submitted already doesn't necessarily mean you shouldn't submit yours, because it may be happening for you for one reason and it happened for that person for another reason. And the more reports we get and the more you can verify them, the more it allows us to track down the multiple underlying causes of these things.

Or it gives us another repro case to actually get to it.

No. And in fact, I talked to Rob Johnson, our lead programmer in the UK, about this, and not player facing. Now, there's been a lot of work done that's on basically how we edit and how we work on things. Think of it as you work on a file, you export it, and then now you have an OC. Now, a lot of times, let's say a space station is made up of multiple OCs, planets are made up of multiple OCs. What we've been doing is trying to allow the designers and the artists to go into a high-level OC, crack it open and then be able to edit and adjust things. And so it increases the workflow versus having to go: okay, I have my big OC that has everything, and then I go to one of those smaller OCs and open up a different editor window, and then basically just keep on stacking editor windows open until I find the one that I want. So this is a way for us to kind of crack things open and almost Russian doll it down to what we need to do to fix it. So we're working on things that increase our workflow and increase our iteration time.

Definitely. On ISC this week, we showcased the particle layer separation tool that was made for VFX. Oftentimes, it's just as important to be developing internal-facing tools to help us track these things down and isolate where the issues are, like with the Russian nesting doll thing you were talking about.

So equivalent: if something takes 20, 30 minutes to check on and see if you got it right, you can only do that so many times in a day. But obviously if we can get that down to a minute to check on it, then you're going to be able to get more iterations on it, make sure it's right and then basically move on to the next task. We want to tighten that loop as much as possible.

It is. No, it's not. Sorry. It is a pain point. We know that. It's not going to be in 3.9, basically. We know that we need to refactor the mobiGlas and actually get it into Building Blocks, and then that allows us to scale. So from my standpoint, that is one of my highest priorities, and I can kind of lay out the highest priorities from a UI standpoint. So on our EU Vehicle Feature Team, we actually have people that are working on the ship HUD and basically moving towards what we need to do, and you start seeing elements of it being pushed out with each release. So basically the center part of the screen versus the upper corner or the MFDs; we kind of just work through each section, and it's the same thing with the player visor. So basically right now we have the bottom left-hand corner, which is the actor status, and then from there we'll be doing the mini map and we'll be doing the weapon section and we'll be doing mission notifications and tutorials. So basically we've got part of the EU Vehicle Feature Team working on the ships, and then we've got part of the Actor Feature Team working on the visor, as well as some of the UI Feature Team to kind of balance between the visor if they need help on those, and then the next thing would be the mobiGlas.

And we've touched on this before. We try to minimize temporary work as much as possible. The mobiGlas right now is made entirely out of Flash, and we want to convert it to our new Building Blocks tech, which we first discussed on ISC a couple quarters ago. If we built the technology into it now, if we did the transfers now, we'd have to redo it later, because we'd be building it for Flash. We'd have to redo it for Building Blocks. So rather than spend our time doing work that we know we'll have to throw away in the near-term future, it just has to wait. Again, it's a matter of priorities. But we're very aware. We were actually doing an internal playtest here just with the video team last week. And one of us had a lot of money and none of the other ones did. And it was very much like, I would like to just be able to give you money.

So the big thing for us, I think, with the mobiGlas is basically getting that home screen done right. And then from the home screen, it will become basically which prioritized apps. Now, S42 has some apps that they need. And then obviously, the PU has the existing ones that we have, as well as other ones that we've thought about: the transfers, even how we are going to clean up the journals and the logs so that it's easier to use and track. And obviously the contract manager really needs to be cleaned up as well. So there's other apps, but as soon as we get the home screen, then we'll be building in templates to actually get the other stuff as well.

Yeah, I mean, this will be our first iteration on it. And then from there, I definitely think that there'll be more things planned. Right now, I think this is a pretty good, what we consider tier zero, a minimum viable product for it. The one going in, not the one that we have right now. So from there, I'm sure there will be some UI issues or maybe some backend services or some stuff on Spectrum that we'll need to clean up. But I think this one's been thought about from a player standpoint. This is moving forward well.

It is cool in the test. I mean, it's a simple thing. But in the test, you can bring Spectrum up on one screen and the game client up on another screen. You can add a friend and it immediately pops up on the other, and vice versa, and stuff like that. So it's an important first step.

It's cleaning up a lot of the issues that we had. I mean, the friend stuff before wasn't very intuitive. And then even on the back end, there were almost like four different lists that I had to keep track of. Now it's one unified list. So there's been a lot of work from a technology standpoint to go in and make this work, again, working on the website as well as working on the client. So I think it's a big improvement.

Yeah, it's not a flashy feature like a new planet, but it's so important to an online multiplayer game to make the process of finding friends and playing with your friends as easy and efficient as possible.

No. Right now, I think we talked about it last time, the server framerate still isn't running the way that we want it to run. We have some code fixes in there. It's not a death by a thousand cuts, but basically there's no silver bullet for it right now. So we're going through, and this is another thing that I was talking to Rob about, basically what the plans are to hit this hard. And we have some code that's coming in that we feel will allow us to evaluate those things better and basically see the outlier and see if we're getting stalled on any thread, and then figure out how we can work that out and reduce that stall. So obviously that will have an impact on the server, and then ideally we get the server frame rate up first before we start adding new people in.

You'll see the visor, and obviously the center of the screen on the ships. It's been reworked and is in Building Blocks now. And then there's other HUD elements, like certain kiosks and other things, that have been moved over from Flash to Building Blocks. The group system. So there are elements on the FPS visor. If you kind of break it down into different screens, basically, you've got your bottom left-hand side; that's Building Blocks. Everything else right now is Flash. And then we're kind of working our way through the quadrants to switch everything over. And it's the same thing that we're doing with ships.

So it's not going to be a rip-the-band-aid-off thing, where one day you come and there's a brand new, top-to-bottom, fully different HUD.

We're replacing the elements. We're doing it in patches and we're doing it in pieces. So if we could get all four done in that time, fantastic. But right now we've got the bottom left-hand side screen and then obviously the center for the ships. And from there, it's just a process of switching everything over.

It's actually company-wide. It's not one small team. It's company-wide that we work towards fixing this stuff. So internally, we have issues that, let's say, QA bring up, or the Issue Council, or other factors: somebody playing the game just doesn't like it. In JIRA, we have a label called QAGP Review, which basically just means QA Gameplay Review, which then comes directly to me. I triage it, and then I add what we would consider a gameplay experience label on it. And so out of our bucket of bugs, we have what we consider gameplay experience bugs, and all of those gameplay experience bugs have directly to do with quality of life. When we're actually doing bug fixing and release support throughout the second half of the quarter, those quality of life, or basically the gameplay experience bugs, get risen up to the top and get resolved. Because in the past, what was happening was we were actually building up a lot of bugs. We have bugs broken down into five different categories: basically blockers, criticals, highs, moderates and trivials. Now, a lot of the polish stuff and a lot of things that just affect the quality of life are really highs, moderates and trivials. So any of the blockers or criticals basically got resolved, but then that left a lot of highs, moderates and trivials. So this is our way of saying, no, I view this as very important, and so, team, you need to work on this. And it's something that throughout the quarter I'm always chasing down, trying to make sure that the feature teams as well as the content teams have very clear focuses: all right, for the ships, I want to make sure that we're going through and we're hitting the manufacturer really hard and squashing as many bugs as we possibly can for that. For locations, it'll be location by location. Obviously, for feature teams, I just want to make sure that they're going through all of their features and making sure that we have a small amount of issues on them. Because we are a live product and we need to have it running not only for the PU but also for S42, these quality of life issues also affect us when we're developing the game as well.

So what we have working right now for volatile cargo is based off of the Prospector and the Mole. So when you're mining, let's say your ship gets jostled, or based off of what you mine, there's a time associated with it, and it will basically become volatile. And so it can destroy your ship, or you would have to jettison it. So we have it working for the Prospector and the Mole, because the Prospector and Mole kind of have their own internal cargo grids. So I don't want people thinking that it will work with the cargo that you carry around by hand right now. That's kind of the next step that we want to do, but we need iCache first before we do that. Then we can do it for the hands and for the other cargo boxes. But for 3.9, when you're mining, we've added a button on the dashboard. So basically, when you're flying around, it will start blinking. And then when it becomes too hot, or basically too jostled, it will blow up. And so you need to jettison it before it does.

Yeah. The design team does know. Basically, we're still going through and trying to play it right now. Right now we have a crash where the QA department will be able to get up to the Idris mission, or the Idris aspect, and then that crashes the game. So we haven't been able to do a lot of internal cycles on it yet. But I can guarantee you that those balance issues are going to come up internally, and then obviously once we release it out to you, it will definitely come up that way.

Yeah, I'm aware of the interest, but we almost had the segment go out yesterday with no footage whatsoever because we kept crashing every time. That's what happens. Things work, and then they don't work, and then they work again, and we were on that downslope up until late Wednesday.

We're still having issues with it. In the meeting I got out of just right before this, we're still having issues with it.

I don't think it is.

I think they're referring to a card on the public roadmap about PvP bounty improvements. I think it refers to the synced assassination missions and stuff like that. Well, those are PvE, though. I don't know what it means for PvP.

I mean, the improvements are... It's not a big sweeping change. With bounty, the way that we want it to go, there's a fair amount of work there, because right now you don't hunt. You just show up to a place and it's like, okay, there they are. There's no hunting. I want to feel like the Mandalorian or Boba Fett. I want to go hunting.

The card, I think, allows players to hunt other players with bounties on their heads. I'm not sure. We'd have to look at it. I made a note here. I'm going to follow up afterwards, and we'll see if we can get some information on that.

Okay, so consuming takes as long as the animation does. Now, what we've talked about in the past: right now a burger would be eaten in one bite. Obviously we know that's not the case. In the future, the idea would be that you take three or four bites out of it and then it's consumed. Now, it depends on the size of the burger. One of the things that we do have right now is that the food basically has kind of a consumption over time, or sorry, absorption over time. It's the same thing with alcohol, with drinks. So basically there's a certain time that it takes to break down and kind of move through your system. Now, what you'll see is basically you've been sustained, and it takes you up to that level and kind of resets the clock. And then, as it's being absorbed, that bar still hasn't been reset and started counting down. It just stays there until the absorption has stopped, and then the bar starts counting down, if I remember correctly on that one.

Chris has been known to say for a while: you take things to the point of realism, then you bring them back to the point of fun. The conversation now is about how long it should take to eat a burger. We can go for that fidelity, where you're forced to stop gameplay for a good five minutes while you eat a burger, but that might not be the most fun experience. You take it to the point of realism, you bring it back to the point of fun.

Yeah, basically we've got certain attachments that you can add, and it'll just bring up the weapon in front of your face, and then at that point in time you can choose: all right, do I change out the sights, or do I add something on the barrel or under barrel? So really, we want to make sure that we're showing off the weapon, as well as that the player gets that tactile experience of switching things out.

Yeah, when I was talking about when it looked cool, I realized that I was actually talking about the weapon attachment UI. Not that this doesn't look cool, with the radial and stuff like that, but pulling up the weapon and then watching where the attachments would go and swapping them out is really cool.

Yeah, well, I think the weapon team has done a fantastic job over the years building in a lot of detail and a lot of things that the player a lot of times just won't experience when it's only on this side of the screen. But obviously now you can actually look at it and you can really inspect it. I mean, you could inspect it in stores before, but now you can actually show it off outside of stores.

Basically, to actually make piracy more legit. Because right now, if you look at piracy mission rewards versus good behavior mission rewards, there was no disincentive to not be a good boy or a good girl, kind of thing. So this is a way for us to really focus on that and really show that there is a separation between right and wrong. And then obviously this feeds into other careers, like bounty hunting, piracy, saboteur, smuggling, so on and so forth. So this is a way for us to really start enforcing right and wrong in the system.

Yeah, we talked a bit with Luke Pressley and the team when we did this segment earlier this quarter. Prisons seem like a punishment. It seems like it's a bad thing for the bad players. But what it actually is: you have to have the risk before we can provide greater rewards. Once we add the risk, we can then create greater rewards for that lifestyle. So it's actually one of those net positive things for the people that want to play on the ill-reputed side of the law.

And I guess if you look at it, this does help. Like I said, anything that really is on two sides of the law, whether that be law enforcement, bounty hunting, assassination, piracy, it helps those career paths. And there's gameplay associated with it. So it's not just like other games where, okay, you go to jail, you pay the fine and instantly respawn, and now you don't have anybody looking after you. There's time associated with it. And then from there, that creates new, unique gameplay: do I want to try to break out of jail, or do I want to serve my time? So that's kind of the way that we're thinking about it, and we view this as a foundational element to actually build those careers upon.