Yogi Klatt

Heya,
from what we see from the current NFM tuning there will definitely be differences. In general in atmo hovering and 6-DOF flying will only be possible when you're relatively slow (i.e. below stall speed). Once you are fast enough that the control surfaces are start to have an effect, the lateral and vertical thrusters won't do much. First of all the thrusters are a lot weaker in atmo anyways (heavily limiting the thrust output) and second the drag from lateral and vertical movement will just slow you down further. So do not expect a lot of strafing during high speed atmo flying, the forces from the aerodynamics are just too strong for that and they vastly outperform the thrusters (like in the real world). That will of course vary between ships (some ships are better at atmo flying than others, there are differences in thruster efficencies, etc.) but NFM overall requires a different approach to low flying as the model has changed.

It will definitely be more interesting though, especially considering that you can turn off all flight assists that keep you safe(ish). I really look forward to see Terada and all those other crazy fliers fully utilizing this.

So, this is a question for me, yes. So, we have been playing with heavy fighter tuning since 3.23, basically the Master Modes release, and our idea about the heavy fighter role hasn't changed much throughout the time. So basically what we think a heavy fighter is, is a heavier fighter that has lower maneuverability compared to light fighters, but increased durability and increased firepower. That's as simple as that, right? Currently, we have retuned the entire heavy fighter class, which includes ships like the F8, the new Guardian, the other Guardians, and so on and so forth, to fit a better spot. And in our opinion, the spot that the heavy fighters now actually cover is quite fine. We do notice that the heavy fighter class is slightly underperforming right now though, but we think it's related more to the relationship between cross classes rather than the heavy fighter tuning per se. So stuff, for example, like how the heavy fighter is dealing against a medium fighter, or a ship of a bigger size, like for example a Constellation or stuff like that. So we would like to dive more into the tuning for other classes first before assessing the overall heavy fighter tuning and whether that needs to be addressed once again. So currently this is the state of the heavy fighter class in general, but we have some even more updates or more ideas for the...

So the current ratio on where the fighters sit doesn't necessarily need to stay this way. We have the updated flight model coming and we're using that to basically, I wouldn't say reset, but reevaluate especially the relationships between the fighters. For example, from our perspective at the moment with the updated flight model, the heavy fighters just feel a little bit too tanky. Because they're still fighters, right? So we're currently evaluating whether or not we should make them more vulnerable. Because if you compare some numbers from, for example, a Gladius versus an F8, we're looking at like five times the health. That is quite a lot for a fighter class. So we're evaluating or reevaluating that once the new flight model, or the updated flight model, is coming online.

And when you say evaluate, that means you're still making your decisions?

Yeah, sure. Nothing is ever set in stone when it comes to that.

You're asking the question whether that five times is right. You're not sitting here saying today five times is wrong.

No. I mean, we can never really set it to a fixed thing. We can plan as much as we want, but whether our plans survive reality is a different question. Also, sometimes other things are coming, for example the upcoming armor system and so on, that will also change these dynamics and might just require some other adjustments somewhere else.

I joined in 2017, but gameplay since 2019, I think. I'm sure there are some things that are still the same, but I think details change over time a bit, right? Because sometimes you just prove that A or B doesn't really work. For example, also the idea to have a perfectly balanced game: sure, we can go for this, but it doesn't necessarily need to be that, because the relationships of players playing are so complex to grasp. The hardest case we have in our combat design is same skill, same pilot, same ship. To get a fight going the way we want in that scenario without any other external factors. When you get this one right, then you probably have something that you can really work with. But it's hard to actually get to that, because even if you have two really good pro players fighting against each other, it's hard to tell which one of these is better. I mean, you can maybe count kill times or so, but there's so many factors and that's sometimes hard to grasp. So I'm not entirely sure that we need a completely balanced game, because ships are different, like people are different, real-world airplanes are different, have pros and cons and so on, but you can never really balance them out completely, right? We'll see.

You've just given people their first Reddit thread: "CIG developers: I'm not sure we need a completely balanced game."

Hang on. What we're still aiming for is we try to avoid a meta game. This is something we're still trying. It will be incredibly hard. If we have a game at some point where almost all ship types have a purpose, where it makes sense, also maybe in the value that you bought that ship for in an in-game shop or something like that, then we're probably in a good place.

Yeah, and it's one of those things: some ships are meant to have progression in mind. It's not that every ship is supposed to be equal. Some ships you're supposed to progress through. A Gladius is where the light fighter journey starts. It's not where your light fighter journey is supposed to end. So just by the very concept of progression, not all ships can be balanced, nor should they. There's supposed to be a 'you start here and you move here and you move here.' Now, that has changed quite a bit over the years and will continue to change as the game continues to develop.

Okay, so at this point the quantum experience that you saw there was exactly how it was at that time in Squadron 42. That design has changed a bit since then, also with the requirements of the other flight model changes that we're currently doing. So in general there's three parts to the new flight model: the quantum experience, the updated space flight model and the control surfaces. And all of these got changed and expanded based on the requirements that we saw. And right now, what we back then described as quantum boosting has changed a little bit. So quantum boosting will be used for the slow speed range that you used to do Nav Mode with, right? So there's basically three speed ranges that you need in Star Citizen. One is to go from, let's say, one POI to another or between planets or something like this, where we're talking millions of kilometers, right? Then you need the range in which you're navigating within these spaces. So let's say you have a bubble of 300 kilometers or so in size. You need to navigate in between there, but you don't want to fly with your SCM speeds, right? So for that, quantum boosting will be used. And then there's the normal SCM speed region, so basically the thruster-based flying, which will still be limited to, I don't know, maybe 300 meters per second or so.

Caveman Crafting in the Twitch chat says, "In other words, that quantum video was marketing fluff." No, it was not. He literally just said it's exactly how it was in Squadron at the time it was recorded.

It was real gameplay. You can actually see that it was real gameplay because on stage our left stick broke, and you can see that in the video because the velocity meter on the left side is constantly twitching because the stick broke. It was all real.

So it's about providing a tuning that fits the archetype of that ship. And that's what we're working towards: giving the ship the archetype. And as I said, it's always work in progress. And that ship in particular, it's got loads of guns, it's got loads of thrusters. And as you say, not all ships have to fit the perfect progression of balance in that sense. So we're definitely looking at what the ship should be with the new flight model as well. There's going to be some more changes there. Because a big thing with the new flight model is we're going to have far more tools at our disposal to make ships different. Whereas at the moment we've got tools, but we're quite limited in that context of providing ships that are quick but sluggish at the same time. So we're kind of in a box for the current balance a little bit in that regard, but with the new flight model we've got more tools to give us maybe a ship that can be really quick, but it actually takes a little bit longer to get there. The acceleration is a little bit slower, but it can move faster.

Anything you want to add, Mattia?

So we do know that at the current state, the overall heavy fighter angular velocity has decreased a little bit, but we think it's for the better. And once again, we think the feeling and the perception that we see on the player side these days is mostly related to, once again, the relation between those ship classes, right? In our opinion, once again, what we want to reach is a balanced and intended gameplay experience for the ship per se, right? So having a super OP, mega powerful ship can be cool the first time that you have it, right? But after a while, when it becomes the meta and it becomes the forced choice, game design-wise and gameplay-wise, I don't think this is a good direction or a fun direction. So what we want...

Yeah, it's not healthy.

Now, of course, you're going to say that and Twitch is about to fill up with people saying, "But what about this ship? And what about that ship that's currently OP?" And the answer is, we'll get there.

Yeah, and that's the big thing about the balance: it's a continuous thing. It's moving, it's evolving all the time. And we can't tune 200 ships all in one go when we figure something out. So it has to come in steps, which is why we're doing a pass on the mediums as well. We've done the pass on the heavies. And that'll keep evolving, but we have to do it in pieces because we can't move 200 ships at the same time.

And also it's more important for us to move those pieces step by step because we understand the dynamics that we generate inside the game a bit better as well. So we get more knowledge about it, we understand how players prefer to use certain types of ships rather than others. So moving those Lego bricks in tiny blocks is easier than just moving one, seeing how it performs and then trying to scale everything according to that. So that's what we're going to try to do, what we have tried to do, and what we are actually doing as well with the current monthly patches. So we always try to deliver something to address maybe a specific issue or a specific area of ship flight in the game.

So it's really complex. So we start off with a ship concept, and the teams we work with, the vehicle team and the drone crew, and we work with them on identifying the identity of that ship and what it should be and what it will be in the game. And then according to that, we're building a certain character that we think suits that ship. It could be really fast forwards, it could have really good handling, or it's really slow and clunky. It depends on the ship. So we try and apply that. And then it falls to the loadout of the ship; it can depend on the weapons, or if it doesn't have the weapons to defend itself, what does that mean as well? Does that mean make it really tanky, so it can't fight back, but it can sustain? But all those different things that we have, all the tools we have at our disposal, create what that ship is in the game. And then from that, the players themselves have choices within that ship as well. And we try and provide as much as we can to let the players move the ship a little bit more to that place or that place, depending on what they want. But the ship always has a kind of window that it will operate in, and it might creep into other ships if you have different components, stuff like that, but it's always got that core identity.

And sometimes that's not going to align with each individual player and what they want it to be, and that's not a balance issue, that's a thing that happens when you've got a game with 200 ships.

I mean, there is a good analogy. There was a German World War II ace called Adolf Galland, and he flew the Focke-Wulf 190 and the Me 109. And although one of them was better, like the Focke-Wulf was a lot better in terms of the stats, he actually preferred the Me 109 because it was just better in handling. And it just speaks to the characteristics of the ship. The ship has its characteristics and might not be the right ship for you. Maybe you just focused so much on one ship that your muscle memory is all aligned with how this or that ship should handle, right? But this is why we have different ships, right? They give you different access to the universe. Maybe the handling is different, but then you have a big [garbled] somewhere else.

No, so I'd say it's part of the journey. And let's say we throw away everything tomorrow, we still have what we've learned along the way. And that's really important because we're constantly learning. It goes back to what you said earlier: I've been here seven years and my understanding of the game and the game mechanics has fundamentally changed in that time, as we've built more systems and added more to the game, and we do know the game's not complete. But it's kind of exciting because we're building towards that and we're always moving in that direction and learning more and getting better, hopefully, at what we do. So we change things along the way, and then we have other systems that come in and they might throw a curveball into that system that we've made. But we're learning along the way to build what we think is the best game.

I think I can add something on that from a different perspective, because I joined the team roughly around two years ago, so not as long as these other two veterans, right? But it's very important, as a game designer in my case, but in general I think as a developer, to have that knowledge and to be able to have that experience in tuning and getting accustomed to the game, getting to know how the game is working currently. Also to understand, first of all, what you can do and what you can plan to improve the game in the future, in the long run, and also be prepared for that when the moment comes. You know what to do, you know what to expect, you know what to manipulate in order to create the right experience. It's never a waste of time. And also, I guess you guys want to play the game as well from time to time, so playing a game that is in a better state is probably better than a game that is not balanced at all.

We know what we're doing and there is a plan, but those things change. We know what we're doing, but we learn more along the way, and those things that we learn change our perceptions.

I don't think that the vision has changed a lot though, right? I mean, the one thing you can't underestimate is engineering and design and development. It just takes time. I mean, seriously, 90% of all the things that players propose that we should do better, we usually thought about these already, right? Because this is what we're doing for a living. But it just takes time to implement these. If somebody tells me, "Oh, please add an altimeter" or something like this, then there's 15 different problems we need to solve. It's a lot.

And those are the easy parts as well, because a lot of the time players have very interesting insights and ideas about very complex systems that we can potentially put in place, but those touch such a vast amount of other features in the game, and organizing that type of work and getting everyone on board from completely different teams, especially in a company the size of CIG, is very difficult, right? But in general, it's very difficult in video game making, so it's not a unique problem for us.

Yeah, we've got a very complex game, but that's very exciting because we are doing it differently. All our systems are real systems. They function, they work, and they do real things based on real physics mechanics. Most times. There's a few exceptions here and there. But most of the time, that works really well for us, because it's a really complex problem to solve. But actually, it makes Star Citizen, I think, really unique. And it makes it really fun to work on.

That 90% is probably a bit high. There are still plenty of times, being completely honest, where an idea has come from the community and we've been like, "Oh, that's a really good idea."

No, but I mean, we've all seen similar media. We've all seen similar films. We play Star Citizen because we probably liked Wing Commander or Star Wars in the past and so on, right? So there is a common understanding. So I think even for some things where we would say, "Oh, we had this idea," it probably just came from somewhere else. We incorporate a lot of community ideas in the designs as well.

So following the Master Modes release, we read the community feedback and so on. We had our own plans with this and basically decided what the next step should be. And there's three things, which are not just the flight model but the whole flight experience, right? So I mentioned before we've got the space flight model, which is how ships use the thrusters. We have the atmospheric flight model, which is the control surfaces, which we also demoed in the past but which got quite a big update now. And there is the quantum experience. So all of these things combined, they're moving the game towards the traversal vision that CR wants. Because the reason why we have Nav Mode at the moment, where you can go into this mode where you can reach 1,400 meters per second, is just there to navigate in these big spaces that we need to traverse, because we don't have the quantum code for that yet.

So in terms of the thrusters, thruster flying: this is not a groundbreaking "oh, we do everything else", it's just an evolution of the previous stuff. It's just building on what we already have. So the main thing for combat flyers out there is that we're going to have a non-spherical velocity space. This means we can go forward faster than we can go backwards or to the sides. This has the benefit that these typical six-degrees-of-freedom problems that we have in combat, like back strafing or kiting, are much, much reduced. From our perspective, this just feels better; the fights are closer and so on. Also we added jerk. So jerk is the change of acceleration. So instead of going within a frame from 0G to 9G, we're ramping this up now over a short amount of time, at least for the fighters. And this just makes sure that you have to plan a little bit more ahead, and the rotations feel smoother. What else is there? Oh yeah, and then we built in a lot of different steps for how to customize IFCS for you. So what you have in the game right now is more like an IFCS on training wheels, right? Where at the moment you can turn off a couple of assists like gravity compensation and G-safe. And these things in the future will actually always mean a little bit of a buff in terms of what your thrusters can do. So what you have at the moment in our internal builds: you can turn off gravity compensation, you can turn off the stability. So the stability, for example, changes the behavior of IFCS and gives your lateral thrust more importance than the longitudinal thrust, which means when you're in fights, it's easier for you to evade shots. But at the same time, it makes the flying a little bit more dangerous because it doesn't keep the TVI stable anymore. Well, that's more technical details.

I was looking at iCore102, who was like, "This doesn't sound like it's going to solve the problem when you're four minutes into a 20-minute exploration."

That's fine. Honestly, don't believe me. When it's available, play it and then you can judge it properly. The other thing, which is a pretty big chunk, or also makes a big difference, is that we don't have top speeds anymore. And now before you gasp, "Oh god, unrestricted speeds": no, there are still restricted speeds, but in the game data we don't have any top speeds anymore. The top speed of a ship is what your thrusters can achieve in a fixed amount of time that we define. But the good thing is when you have your power assignment and you put more pips into your engines, your thrusters will actually produce more thrust. And more thrust then means more speed.

It's not some artificial speed limit that the universe has applied to you.

Well, we need speed limits. We're a game after all, right? But this basically means, because it's based on physics, what will happen is: you have your Gladius and you fire a missile. The overall weight of the Gladius is reduced now. So your thrusters have the same thrust output, but your ship is now lighter. So you actually get more Gs. We had this before, but now you also get more speed based on this, right? And looking further ahead, this is outside of the new flight model, but when we have Maelstrom and so on, we're also going to have more realistic ship masses. At the moment, the Gladius weighs 55 tons. In Squadron it was up at 180 tons, which is close to a 747, right? But now the Squadron Gladius is around 20 tons, which means the weight of every missile and every component actually matters. So if you, for example, have your Gladius or any other ship and you just rip out the most important components, you will just get much, much faster.

This is going to be very relevant for smaller craft, maybe for racing as well. Having lighter components or removing weapons will give you that extra window of performance that you can exploit throughout the race.

That is basically just the space part of this. On the control surfaces, we have a fully licensed pilot on the control surfaces for now. That's Karim. He is working on the control surface support. What we already have ready is the unassisted aerodynamic flight, which means how the ship feels when you directly affect the deflection angles of the control surfaces. So you can glide. Your ship can be entirely off and shut off, and you still have control over your control surfaces, which means if you're getting distorted in atmosphere, you can actually glide down, you can flare and you can crash land your ship in a controlled crash landing. So for the core on/off thing which I just mentioned, we introduced a concept which is called the IFCS core. What you have right now in the game, where the ship is just constantly stabilizing, compensating for your rotation, means the core is on. So IFCS is actively working for you. But IFCS in the future can be slapped down or even turned off. And in this case, you have pure Newtonian flight, also on the rotation. And this also happens when you, for example, crash into something. For example, on the Gladius, let's say we know that the Gladius cannot go above, let's say, 20 Gs. Then we might have a limit in the IFCS code that says when that ship goes over 25 Gs, the core is slapped off for a couple of seconds and then it needs to reboot. So during this time, you still have control over your ship. You can still use the thrusters and so on, but it will not auto-stabilize. So it's more or less like flying in Kerbal Space Program. Which means you will tumble after every collision, and the ship will not recover as fast. And whether the core is on or off also plays a role in atmospheric flight, because in atmospheric flight when the core is on, we will have a fly-by-wire system that keeps the TVI stable, but when the core is off, you control the control surfaces directly. A lot, sorry.

That 20 and 25 thing: those were numbers off the top of his head to illustrate a point. Don't go and update the wiki saying it's only going to have a 20G tolerance.

These are just examples, right?

Quantum experience. Yeah, the quantum experience will change because at the moment, it's very clunky how we travel through the game. So I think the best example is: Nav Mode will die. F's in chat for Nav Mode, everybody. The reason is more that it's very clunky to traverse the 'verse. So the way it's currently planned, and I'm saying currently, it can change, right? Of course it can change, right? If we play it and it's not good, then we'll change it again. But the overall goal is to make the traversal through the 'verse very, very fluent. So the thing that will define whether you can use a quantum drive or not is just whether it's powered or not. And this is a resource network problem. So if you set up your resources so that your quantum drive is not powered, you cannot use quantum. If it's powered, you can. And then it will be as easy as looking somewhere, holding a button and off you go, like we had in the past. So there won't be any mode switching anymore. There is no quantum drag that slows you down suddenly with 70 Gs or so. So that will all be more fluent. And I'm not just saying "it will be": we have most of this already in Squadron. It's actually on my to-do list this week to rip Nav Mode out.

The PU?

No. Due to the time constraints we have in Squadron, we first do these things in Squadron. But everything we do on the flight model is directly moved to the code base of the PU. Basically, most of the stuff on the flight model is already in the current 4.2 code, but it's of course turned off. We will also not enable it until we have all things in place, including the ship tunings. Because in the Master Modes release, as Mattia mentioned before, we had to go through all the ships very, very fast, because we had this time window where we needed to commit to the Master Modes changes within two months. For the new flight model, the data sits all in parallel, which means, I don't want to say take the time, but we don't need to rush the implementation or the tuning of the new ships, because we don't want the ships to be as similar as they were before. We want to give them character. And we don't need to rush this out, because we can have the old and the new flight model data side by side without destroying each other. Unless you do an integration wrong like I did. There was a little buff to the Gladius which was due to an integration which I messed up, and I apologize for that.

Okay, so I know what you want to say. So the vertical thing, right, where ships do this: nobody wants that, right? CR doesn't want that. He had a talk with me last week, right? But this is also something we probably will only fix with the new flight model. I don't want to say, "Oh, it's going to be fine in the future," but it is the plan for this, right? So right now, when you're in atmosphere, we have something called thruster efficiency curves, and the efficiency of your thrusters goes down by about 50% in the highest case. In the future, that efficiency will be much, much lower. So what you can expect, if you have a ship, let's say, give me a ship with VTOLs, I don't know, Cutlass, right? So if you have the Cutlass, the Cutlass is a light transport, so it will not be as punishing, probably. But let's say, does the Idris have VTOLs?

Yes, but they don't work currently in game, but they will do.

Okay, fine. So let's say the Idris has landed. Then you will get a message on the heads-up display that tells you what your thrust-to-weight ratio against gravity is. And if that ratio is below one, you will not be able to take off. And this is then just taking into account the mass of the ship and how much is loaded on it. So if you have your cargo full of lead cubes, let's say, then it is possible that you will not be able to take off.

It's possible, not a guarantee. You can overload.

You can overload, and then you will not be able to take off. I mean, you may try to boost, because boost gives you, vertically, I think something like 35% more thrust. You can put more power into your engines so you get this natural resource-network-based buff in your engines. You will have a hard time just hovering like this. And at the same time, these are your strongest thrusters, but your retros are usually a lot weaker than the other thrusters. So if you do this, you will eventually just sink down. That's at least the plan. So you can expect that larger ships will be harder to fly and to land in atmosphere. We will try to give players helpers so that new players don't just crash land this thing. But still, there's still physics. And the reason why the ships can hover at the moment is because they have so much thrust that they can. Because all they need to do is be able to hover in 1G, or let's say on Earth, right? When that is not guaranteed anymore, then you automatically have a more realistic and harder situation there.

Yep, so the whole point of those turrets is that the ships can't defend themselves with guns, so we've got turrets, and we want the ships to be able to defend themselves with turrets. We are working on the general turret experience and how powerful the turrets are as well, and we've made some changes. Was it last release?

Yeah, in general, with the engineering buff, we buffed the ammo, we buffed maybe some regen rates for specific ships as well.

But there's this other thing. The thing that we're currently lacking is effective medium-range weaponry. Right now, a turret fires projectiles that are just normal gun projectiles. The thing is, we don't want to twist the projectile after we've launched it, right? So what we need is some medium-range weaponry. We have a plan for this. We would like to try out flak at some point, but it's not ready for multiplayer yet. But it can be effective beyond the range where normal gun turrets are effective. I mean, I guess the question comes mostly from, "Oh, I have this large ship and I'm getting shot at by fighters which keep their distance." But the other thing is, coming more to Torsten's work concerning the armor system and so on, there's also the question of whether smaller ships will actually be a danger to larger ships, because we are going to have penetration parameters on the bullets and we're going to have armor and so on. So a lot of these weapons that people use at the moment to damage larger ships, they might not be able to damage them in the future. It's like if you have a pistol and shoot at a tank, the tank says, "Well, okay." If the ships want to do actual damage on the larger ships, they probably need to get so close in the future that they can precisely destroy the subsystems.