Star Citizen: Reverse the Verse LIVE - Ship Combat Experience

Yeah, I guess all of those things. When we think of a weapon being overpowered, I tend to think, okay, that's the DPS, that's the time to kill of all the ships. The smaller ships usually, if I sink, so your weapon is overpowered. That's how I test it anyway.

Give it a little go on those small ships. Step one is what should this weapon be doing? If it's a cannon and it's overpowered, it's probably either doing too much damage or it's firing too fast. So an overpowered cannon is something different to an overpowered repeater. They'll probably have very different behaviours that are causing that overpowered weapon.

The easiest thing is checking the damage, checking the rate of fire, and then the spread, I suppose. It's the accuracy, the range. So how easy is it to hit with this weapon, and how much does it do damage-wise over a short period of time?

I mean, you can get other areas that aren't the natural statistics that cause the feeling of a weapon being overpowered. Some things you don't tend to think of, like the VFX or the audio. Some VFX might not be playing as you intended. They might be less intrusive, which gives the feeling that it's more powerful than it is. And the same for the audio. Quite often, people perceive weapon strength to the sound it's making. It sounds silly, but it is something. If you have a big sniper rifle in the game and it just goes [weak sound], then people tend to think, well, that's a rubbish sniper rifle. It could be doing all the damage in the world. But that audio really sells it, and if that's not on point as well, that can sometimes factor into it.

It all ties it all together. And it's like quite a lot of games, actually, if you dig into the details or you have access behind the scenes, some of the weapons are clones of each other. And it's just the visuals and the audio that change. You'll have people arguing that Weapon X is better than Weapon Y and it's not, it's the same, it's just the whole package that makes it feel that way.

Where do we start? Start small. The time to kill, I know obviously we said this, prefaced it a lot, that things aren't finished and it's still ongoing balance. So we tend to try and get things right on the small level, the small scale with damage, rate of fire and all those things. And then with the shield items, the armor items and the ship health. I wouldn't say that we have an exact time to kill in mind.

So you start at the small level and you do one ship against another ship. You get that rough ballpark time that we're interested in. And then we go, oh, what's the next level up? Does it sort of scale with that? And then they all sort of build onto each other. Then the wildcard is obviously people can change their loadouts. So when we do our time to kill goals or estimates, we do it off the base loadout. Because you can't really factor in every single combination of weapon and missile and shield and armor combinations that they can do. So we go on the defaults, and the default loadouts for every ship sort of contain most of our weapons. So when you're balancing one ship against another, you're already factoring in those weapons. So hopefully it doesn't get skewed massively if someone changes all the weapons from one to another, because it's factored in somewhere. But as we build up the ships, it does get longer.

We tend to play test these things a lot with smaller ships actually to get the feel right for those. I think I'd be honest to say that we haven't spent as much time with the larger ships and that is still work in progress. There's a lot more to do there.

There's a lot of systems still to come that will make those ships more enjoyable to behave.

Yes. Hopefully people noticed a quite significant change in how the ship combat was between 3.0 and 3.0.1.

It's quite interesting you have to talk about that patch as a little hotfix, which it kind of was. It was taking the state of the game in 3.0 and the things that we weren't happy with and quickly isolating things that we could do and fix quickly. So it was a very quick turnaround on, I suppose, everyone being back in the office and assessing the state of 3.0 and then what we can do just to get balance feeling much more enjoyable. A quick fix was, say, we brought down the health on shields and the regen on the shields. That's going to be something that's going to change a lot, even from now, because we won't have it right for a long time, I think. But yeah, that patch was a quick turnaround thing.

Made to quickly isolate what we could fix very short term. And then what we wanted to fix for 3.1. And then what we wanted to do beyond that.

It kind of highlights that balance, it's never done, but we can quickly assess and make changes just based on the current state of the game.

I'll try and briefly go over it all. Like you said, we balance for the game as it is now. How we do stuff is how we can do it. The long term goal is still going to be the whole component-driven behavior. So rather than ships having an artificial health pool and armor as a flat reduction, it will be physically based damage. So the parts of the ship will have strength based on their materials. Projectiles will have mass and velocity, and then that will be how the damage is calculated; penetration will fall out of that. Energy weapons, obviously, well, lasers don't have any mass or very, very little mass, but they have velocity, so they'll be great at some things. Ballistics have mass, less velocity, and that's shields. The lore of the shields is that they slow down mass, so that's why energy weapons are great against them, and they sort of eat it all up and deplete the shields. Ballistics go through them because the mass, it slows the mass down. It doesn't stop the mass, it slows it down. And they've got enough mass to go through and then hit the hull underneath, whereas lasers and energy weapons don't have that. So we're still going towards that long-term goal of doing it that way. Very simulation driven. We've been talking about it for a long time. We started looking into it when we were first talking about it, and we sort of realized that there was a lot of pieces of the puzzle that we had to get in place first, mostly to do with getting everything to Item 2.0. So for 3.0, most of the ship components got done for Item 2.0. At the moment, we're finishing converting the weapons, which is naturally a very big part of this whole new damage behavior that we'll be going towards. Everything has to be Item 2.0 to work nicely with each other. You could sort of do it before, but it wouldn't have been efficient and would have resulted in a lot of wasted time. So all the pieces of the puzzle had to be Item 2.0: ships, weapons, ammo, and sort of all of that. And then you've got to actually come up with the rule sets for it all, which we're working on at the moment. So whilst, yes, the ships have a health pool now and the armor does X and shields do Y, that is simulating to the best of our abilities future behavior with what we have. It's not one-to-one. It's not going to stay like this forever. It's going towards that system.

It's one of the unique challenges of having to run a live environment in the middle of game development. Because we are running a live environment in the middle of our game development, one of the difficult things when talking about stuff, whether it's you guys talking about weapon balance yesterday on ATV, or developers talking about piracy gameplay systems, there's always this disconnect between talking about what is going on right now and what we're capable of doing now with the amount of game systems that are online and in place and implemented and working, versus our long-term design goals. We've said for a long time that we want a systemic damage system. It's basically you set all the materials, you set all the features of all the materials, all the components of the ships, and basically it works itself out. It's all energy-based, whether the energy is coming from a laser bolt or it's the kinetic energy involved in a ballistic. We all want to move to the systemic damage system, and it'll happen just like everything else. But where we are right now is what we were really talking about in the segment yesterday.

One of the really cool, interesting things that's going to come out of it is at the moment we talk about time to kill because it's what we have in the game at the moment. It's a very sort of linear concept of ships have health. You get the health to zero, ship goes bang. You don't have that nice way of disabling ships. Whereas when we move to that system, it's less about destroying the ship, more about disabling the components, which in turn disables the ship. And ships are only going to explode when you do critical damage to power plants or huge ammo boxes. And even then, they might not kill the ship. It will cause an explosion, and depending on the tensile strength or the fracture index or the strength of the part and the location of it, that bit might snap off. We've got the ability now to have multiple physics grids in ships. We saw that at Gamescom with the Idris, where it snapped apart, and you couldn't quite see it in the livestream, but we had fun with it in testing in that it snaps apart and it didn't kill the people inside. You could still wander around. You could board the hulk of the Idris hangar that was floating around with all the power out, because the power plants were in the bit that had snapped off. But it was a living bit of ship that was just floating through space. And that sort of gameplay is going to fall out of it naturally.

So I'll tell a little story about beam weapons. When I first started here, beam weapons weren't in the lore. And then a designer, who shall remain anonymous, did a little prototype and showed it to Chris and the writers, I believe it was the Squadron guys, of this turret with beam weapons. And the moment they saw that, they were like, that looks awesome, we should do that. And whilst we don't have the code in place to do it, we do have them on the schedule to come. And they do provide a very different gameplay. For each of the weapon types, ballistic and energy and other types, we sort of try and have symbiotic ones. So the ballistic gatlings are the very fast-firing ballistic weapons, whereas energy beam ones are their equivalent. So it's a continuous stream of matter. So yeah, they are on the list.

Yeah, it's strange, isn't it? Because the door's open for like an energy gatling as well. There's nothing to stop it.

It doesn't work like that.

So what's the answer about beam weapons in Star Citizen? Just one line.

Yes. Maybe. Yes, at some point.

As Jared said, if you go read the Shipyard posts, then you'll see it. It's a sort of convenient way of comparing. It applies to ships and it also applies to items. So we have the ship classifications, military, civilian, transport or industrial and whatever, and then grades A to D. So civilian C is an item that performs at 100% of its capacity. So ships come with typically civilian grade C items by default. That's bog standard behavior. And then every other sort of combination on that matrix increases various stats various ways. So, for example, shields: a civilian shield, grade C, is a shield. A military one, still grade C, has much greater shield output, regen, but consumes more power and gives out more signal. A stealth grade C shield is one that has less output, probably a slower regen, but it's significantly reduced. So you can tailor the behavior, and then you factor in the grades, they go up naturally and down. But each item has, I think, up to eight statistics that are factored into that grade. So depending on the item, they all go up and down, and that ties into their performance, balancing falls out of it, and then it factors into the economy as well. Value, definitely.

It's the same when it comes to the career roles, right? They all have, say, the transporter has industrial items that fit onto it to fulfill that role of the ship. The items are supposed to help us create these identities.

Transport ships are supposed to be out there a long time, long haul. So the items they carry are ones that have high durability.

It's probably better answered by the scanning guys. For now, the scanning will have multiple stages. So a very, very high-level initial scan gives you the rough information. So when you're trying to lock onto it, it will do a brief scan. It might say Hornet Ghost or another stealthy ship. It won't give you all the information of the ship. You have to keep scanning it. And naturally you're going to learn. If you don't own these ships or know nothing about it, you're going to hopefully learn by feeling that every time I encounter one of these ships, it takes longer to do this. Whereas if I find another Hornet, then it's, oh, that was much quicker. So then something is up with the Ghost. If you own these ships yourself, then naturally it's all explained to you in the lore on your equipment terminals. It tells you all about this.

There should be signature readouts of these ships once you scan and target them. The ship's infrared levels will be there.

You'll notice that it is, from your experience, less than other ships you encounter.

Yeah, it was supposed to be like the signature of the ship makes you harder to hit or easier to hit. You've got loads of the items that are overpowered, maybe overclocked, giving off high signatures. The difficulty of showing that to players, I suppose we used to have little readouts that would just say numbers. That's coming back. It's coming back, good, because that makes a big difference.

I saw it on the UI guys' screen today.

It's something that I think is really open to us at the moment. We have been discussing this this week, actually. I think some of us are more expecting that we'll go down the road of shields and armor that can be switched in and out that will have different resistances to different damage types. We have different damage types in the weapons already. We need to be making sure that the defensive side can be alternated in the same way. You can change your loadout for different weapons to do different damage types; you'd be able to do the same to counter on the defensive side.

It's no good just stacking all the goodies on the offensive side without providing at least a fair set of tools on the defensive side.

Yeah, that's the gameplay we want. I think everyone would like to see it that way. So yeah, we've been talking about that a lot.

And at the moment on all the shields, you do have some ability to sort of tune it. So you can move the faces around and that does have a significant effect. You can do shield hardening, which, as the name implies, does harden the shields and really does reduce the amount of damage coming through. So if you get into a combat situation and you go, I'm seriously outgunned here, I need to get out of here, you can just start flying in the opposite direction, punt all your shields to your back face, harden it, and it will significantly increase the amount of time it takes for those shots to come through your shields, to give you the time to quantum out of there or do other things.

Current state, where it is: we know there are issues with it. And I actually have, because I saw that question, some notes from the engineers that are working on it handed to me. I'm a designer, not an engineer, so I apologize to whoever gave me these notes if I misinterpret it. ESP. People were complaining that it was broken or not reliable enough. There is an issue where the pips were, because it uses the pips to work out the ESP, and there is an issue where it was picking the wrong pip or no pips to do the ESP on. So you get wildly different behavior to what you were expecting and inconsistent behavior. That's been fixed, and is with QA now in a QA test request that they're testing at the moment. I believe you had a look at it.

Yes. Just a quick look at the code shelf myself, and it seemed fixed to me. I tested it on a static ship. We had some issues, obviously, with UI preventing QA from getting a full test on it. But the changes that John Pritchett had made seemed to have resolved this problem. I mean, we recognized in 3.0 that there was an issue. We didn't have enough time for it to come together for the 3.0.1 hotfix or interim patch, whatever we want to call it.

Yeah, it took a lot of investigation because it wasn't a 100% reproducible issue. You had to have a bit of luck to reproduce it and then catch it in Visual Studio. So you had to spend a lot of time investigating. And they had a vague idea of what was going on fairly early on, but it took a lot of iteration to really nail it down.

It's tricky. I mean, you played the game and you felt something was awry, and the testers felt the same, and all the community feedback was telling us this. And you get loads of different feedback: oh, it's working differently when I'm using a joystick or a gamepad or mouse. And you start to go, okay, there's something weird. How do we pin this down? It's fine for me. It's fine for me, but maybe they don't understand what ESP is, because it's quite a complicated thing to even explain. For those that don't know, it's enhanced stick precision, which, the closer your movements are to the target's movements, the more I think it helps adjust you to the target. I think that's the best, easiest way to put it.

It sort of desensitizes your movement to enable you to be more accurate.

If you're good, it'll make you better. That's the kind of design philosophy behind it. It's not an auto-aim in the strictest sense.

You still have to do input to do it. It doesn't just move your cursor near it.

Yeah, and I was playing it this week to see how it was working again. There was perhaps a sense that it was a little bit too sticky or corrected a bit too much.

So that's one thing that came out of the QATR that we've got to work on now. It is design-tunable, but the values we had given for it were with the previous iteration of it. And with these fixes, it's changed it, so we now need to go back and do some more iteration on it to make it not as gluey.

I think with changes to rotational velocities on the ships in 3.0, that probably made it out of sync with each other. The algorithm isn't able to keep up with these ships and their turning rates. So yeah, we need to look at that for 3.1, but the fix is in QA testing. Future of ESP? I mean, it's so important to have something there, I think. When you're targeting these tiny little ships sometimes, some kind of correction to get you on course. When it's not there, people notice it, I think, eventually.

Yeah, and it's like all of our assist systems, it's toggleable, so if you don't commit, tick it off. Advanced players can easily switch it off. And about IFCS, this is a really engineering one, so if people want more of an in-depth explanation, I'm sure Jared or someone can find an engineer to talk about it. Before you could fix ESP, we had to fix IFCS. So a big problem people were seeing was the control attitude sort of separating position- and rotation-wise. That was caused by changes between 2.6 and 3.0 that sort of went under the radar.

It did to us as designers. We would never have known about this.

It was part of the whole optimization for 3.0, moving lots of things to a batch update system so it would be more efficient. IFCS worked on a fixed time step, and with the batch update change, it now works on a variable time step. So things were sort of getting out of sync, and that's what people were seeing with the two reticles becoming out of sync. So the control methods that IFCS uses have been improved to account for the new optimizations more thoroughly. And lots of improvements and just iterations on that. And that, again, is with QA now being QATR'd.

Before we came on air, because I knew you were going to ask that, I went and had a quick look at it. They are actually both axes first order, not zero order. It's just that the values on them are very different. And long term, we want to make it so you can have it on zero order on both or first order on both and give people the preference, because as with everything, people tend to go, I like this or I like that. You can never get the two camps agreeing on what they want from it. So at the end of the day, it's what feels right for you.

I mean, just to point out what the problem was that was highlighted with the question, I think, from the community, was that pitch and yaw on turrets were feeling very different.

Yeah, when you pitched it was more instant than yawing. And yeah, it's come down to the values that are in there, and we're going to be looking at that and always changing and improving stuff.

But it's one of those things, I think, where we weren't aware there was a problem. But we took a little look and we seem to have a way to fix that.

Yeah, I had a quick look at it, and the changes can be done that make it closer to zero order. It is actually first order, but the dead zone is very small on one axis compared to the other. And the other factor, if you think of our turrets, they're not gimballed weapons. So for pitch, the guns just have to go up and down, whereas for yaw, the turret has to rotate to match it. Perhaps if the turrets had gimballed weapons on them, they could yaw to hit your point of aim quicker. But there's lots of things to think about and improve on. We know they're not as happy as they should be. There has been work and there is work going on to make them better. Because everyone always comes to us going, how are you going to make multi-crew ships with turrets competitive against single-seaters? Why would I ever bring a multi-crew ship instead of three single-seaters? We're aware of that and working on it.

The one great thing about interdiction is that it's a completely new feature. It's a layer on top of the game that can be patched outside of the game. As people will hopefully have noticed after 3.0, various issues like asteroids turning up outside Olisar, interdictions going mad: those things, the people working on interdiction can quickly iterate on behind the scenes without having to push patches. We had whole new things turn up for 3.0.1 that those guys have diligently worked on behind the scenes to fix.

Yes, in 3.0.1, I'll give you a little backstory here, guys. There was a bug that came up after 3.0.1 that was quickly fixed through hotfixes behind the scenes, where if I was on my server and I hit quantum and it rolled the dice and said, okay, you're going to get interdicted, it wasn't spawning the interdiction on my server. It was spawning the interdiction on John Crew's server. And it was spawning the interdiction on Andy's server. It was spawning the interdiction on other servers. So when 3.0.1 went out and interdictions went crazy, that's one of the things that was happening. It was the earliest version of server meshing. But it was spawning missions on other servers, and so that was quickly and promptly hotfixed by the team there in Austin.

There was another fun one as well, where they were becoming self-aware, and the ones that were being spawned through interdiction were then trying to interdict themselves. So it just got out of control.

I think that would solve a lot of problems, wouldn't it? People's issues with pip divergence. And it's something that, again, has been talked about internally. Because I know through speaking with Calix, he's had a mind to fix this kind of thing for ages. Whether it happens...

Yeah, it's one of those, in our heads, it's something we should be able to simply try. It's just trying to steal that precious engineering time away to give us a day or two of ability to set that up and test it. Because we think of lots of things. Everyone thinks of lots of things during the day and goes, that'd be great, that's going to solve everything. And quite often they don't until you actually prove it. And you've got to start working with it and see all the problems.

You've got to work through all of the obvious issues that will come up before you can even start getting these things in game.

It wouldn't be so much the rate of fire. It'd be more like dynamic convergence and velocity calculations.

We know about different weapon projectiles firing at different speeds, the problems that it has by creating the pip divergence. We've already brought them in to be similar speeds. You can bring them in further.

Another thing we're considering for a future patch is changing some of the default loadouts for ships. I was talking with Chris about it. And a lot of the default loadouts for ships haven't changed over the years. And when they were originally announced, there was a very limited set of weapons. So they all sort of have very similar loadouts, or they would have a wild mix just to make them different. And now we've got a lot more weapons in the game. It's something we can consider looking into and just going, does this ship actually benefit from having this weapon on a fixed or a gimballed mount, when perhaps actually it would be much easier and simpler for that ship to be in a different loadout? I'm not talking wild changes like size five hardpoints on the wings of Hornets and M50s, but more changing this repeater for a different one that behaves a bit better.

I'd also like to look at the weapons used across the ships at the moment. It tends to be similar stuff on the loadouts, right? If there's a majority of the loadout configurations with, let's say, these two weapons, we could look at making those two weapons behave better together. That might make a huge difference to a large number of ships just by bringing them in a bit.

Another thing we definitely want to look at is the default weapon group assignments, because there's certainly some ships out there that just have the wrong weapon groups by default. They've got fixed and gimballed weapons together in the same weapon group, and then they'll have the other weapon group with a different set, when really they should be all fixed or all gimballed, or all ones that are similar together in one group, so they have that nicely converged.

Yeah, and when they vary so much, players like to lock them in anyway. You gimbal lock them so that they don't have that problem. And that's something only advanced players really know about doing. So we've got that issue of new players, beginner players, having these divergent pips. They're having trouble hitting things. It's an issue that we still need to solve. But I guess we've got a few ideas about it.

We touched on that before, in the earlier answer about how we're balanced around the current gameplay and how we're going to move towards that component system. So yes, it is going to be like that. You will have, on the ships that have interior space, you will be knocking out those components, and someone will have to go repair, replace, reroute power to get the ship back up to function. And it's something we can't nicely do at the moment, because we have that fixed health pool and you're alive or you're dead.

There's been complaints about big ships. Sometimes they can just die really quickly. It's because of the current system we have, and we've had to work with it and try to balance around that. And sometimes it works really well for those small ships and for the large ships. And I remember the Aurora having all that superfluous health for the longest time. I remember the Aurora would just soak up damage like a sponge. And that was because of the current system.

Going forwards, mid to long term, it's going to be talking less about time to kill, more time to disable, to give you that gameplay. Because if you're blowing ships up, there's no way to board them and do anything interesting like that with it.

How are all these shipwrecks out there? Every time you get into a fight, they just blow up. I mean, obviously we've designed missions and stuff based around shipwrecks. That's the intention. That will be a long, protracted, multi-crew engagement. You need to be able to get out of one chair, run to another, change the system, defend yourself.

More great things that just fall out of that is they start generating missions themselves. So rather than it being design-placed wrecks, people can start calling for help that they're disabled, need repair, refuel, rearm. They need people to come help them do stuff, need to help them transfer their cargo. The game just sort of builds upon itself.

Yeah, you have to have that longer engagement. Time to disable.