Is there anything you can tell us about the antennas, the bridge shutters, the stuff that's still outstanding for the Carrack?

The jump antennas will be there when we have jump points. The shutters, I think when we did the model of it, we left them off because we weren't quite sure how we wanted to do them. We modeled them out, proved it out in concept, but we never transitioned that to actual in-game. And we have been doing some performance work on the Carrack, which maybe Ben can talk about, what and why we were doing it.

I think part of what we were talking about before, we have times between ships, and we decided to put a bit more of a focus onto looking at our ships and our processes, working with the graphics team and the rendering team to see what performance costs we can save. In our game, like John's mentioned, there can be points where you could have a large number of the same ship turning up, or you might have a number of completely varied ships turning up into one player space. And what we're finding is that whilst our ships were OK when you were up close and personal with them, when they were off in the distance, we were hitting some not great performance. So we just took a step back, took a look at it, worked out what we could and couldn't do. And just through a better understanding of the way our engine works, we made significant improvements on the Carrack. The Carrack is not by any means our worst performing ship, but it was middle of the road. And we've got a lot of ships that, off the back of that, we've looked into and gone, oh, actually, do you know what? They're working pretty good, they're set up well, and we're quite happy with this. But a lot of it as well is, I could probably talk about it for an hour, so I won't. A lot of our initial budgets for our team, we have the rendering cost budgets that we set out really early on in the project. And then over the years, more features and more complexities have come into it. We have to go back and revisit those and go, actually, is it achievable to deliver the ships to the visual standard that we're aiming for, that support all the functionality that we require, and still be within those original budgets? The short answer is no. Even just things like the dashboard interaction system, separating off all those individual elements into their own interactive buttons and things like that adds cost. If you look at what we've done with the Gladius recently, adding in all the ship component panels and the access panels, that's all work that we've got marked on to do for our other ships, but that all impacts cost. So what we need to do is, we don't want to just be piling in, piling in feature work, and then suddenly be at a point where we've completely tanked the performance. So what we're trying to do is, every time we're adding something in, we're revisiting the way we create the assets and seeing where we can make savings. And with the Carrack stuff, like I say, a lot of it was just really digging in. We know where we can save performance, but it's really spending the time and getting into the details. And a lot of it's to do with our lighting setup, our VisArea setup, the way we do our LODs when you pull really far away from our ships, and a lot of the details. When you're working up close on a ship and you're manually doing these LODs, you can get a little bit tunnel vision on what you're working on. So it's actually looking at things and how they're going to be used in the game. So I'm rambling. We're making some great performance wins on the Carrack. That was our first one. And we'll be rolling on to more as we chip through the backlog.

Hopefully players don't even notice a visible difference on it either.

If anything, we've been improving the visual pop and the LODs by just spending a bit more time thinking about exactly what the player sees.

And to give a rough ballpark number, you can now have two Carracks on screen for the cost of one of them previously. And most of that's gone into 3.13.1. I think there might be a little bit more coming in 3.14.

There are one or two things you might notice that we didn't quite get time to close out in time for 3.13.1, but it was better that we put it in with the performance gains that we did achieve. And there might be one or two things that we just have to go back and tweak. There's a little window that doesn't quite look right. Just minor details. But mostly it went in.