Okay, let’s start with the obvious–the link to the latest review video:
The past couple of months have been insane really. Getting campaign and mission management up and running, along with the AI basics, was a LOT to tackle. It all really had to be done at the same time, too. Mission management needs to know what the AI is doing, the AI needs to know what the mission expects of them at any given moment, and then of course the AI has to “talk” back and forth with the ship systems so they can properly fly the ship. It had to be a coordinated undertaking. The worst is over though–these things are in and working so now it’s a matter of building on and refining them.
So, what are we looking at in the vid? Let’s start with the AI:
What is occurring now is what I call “Fighter Pilot Aiming”–this solely allows the AI to determine what inputs need to be fed into the flight control system so that the ship aims at the target. I expanded on this a bit so that the AI also maintains a range to the current target. This is what allows them to circle you like a shark without getting pulled away from you because of variations in your orbital trajectory. No other maneuvers are being performed.
The next step is to refine the steering logic (can get caught in some terrible spins at the moment), as well as add logic so they can perform longer translations while managing their orbit. In the prototype orbital mechanics were not applied to ships; so I wasn’t an issue then.
Please keep in mind that there are two kinds of AI brains–a single-crew brain as well as a multi-crew brain. Starting with the single-crew is somewhat easier, I think, because one person is responsible for everything. There doesn’t have to be communication back and forth from Captain, to pilot, to navigator, etc. Obviously the multi-crew brain is what will allow us to have multi-crewed ships later on in MP, with the players taking the place of the AI logic for each of their respective positions.
Meanwhile, as the AI is doing their thing, information is being passed back and forth with the mission and campaign managers. If a ship can no longer perform a mission the manager needs to know about it so another flight can be re-routed (or created). If a task is complete then the mission manager updates the ships in the flight; and then the individual pilots start selecting new targets, etc.
Next is what happens when a ship gets hit with a physical ammo round, such as a rail gun slug (by the way, disclaimer, ammo rounds are not yet being affected by gravity from the nearby celestial body. That is on my to-do list for this week). When a collision is detected we know what type of object was hit (ship, asteroid, etc). If it’s a ship I can also tell what COMPONENT was hit. A general armor strike will get checked. If the armor was compromised at that location then damage will occur to the system/s in that location. However, a direct hit to a weapon, let’s say, will cause damage to that specific component.
Also, at the point of impact the relative velocity of the struck ship and the round are compared, along with the impact angle (glancing blow as opposed to a direct hit), and the mass of the round. All this info is used to determine the impact force that gets applied to the ship, thus displacing it. Finally, thanks to Michael, we were able to get dynamic hull impact sounds put in that change based on the location of the hit as well as the impact force. If you wear headphones you’ll have a very good idea where your ship just got tagged, and how hard, just from the sound.
Lastly, there’s the new Viewport Head’s-Up Display. Currently this only consists of the trace/track icons and the Lead-Computed Sight. More information will be added in the near future to better inform you of your current orbit, etc. As you know, right now we only have the optical scanner, which scans space looking for a return. Each return gets displayed as a trace. If you select a track (the green-dashed box) you can then try for a lock. Since the current ship only has the one scanner, which has to be focused, when the target is locked all other traces are gone. When we have multiple scanners it will be possible to lock a target while having trace data still available.
Once locked, we then get information about our current closing velocity with respect to the target. Information coming will include an estimate of the target’s current apoapsis and periapsis. Likewise, the lead-computed sight is project, including the TTS (Time To Strike), which shows you how long the round will take to strike the target if fired at that moment. A large “X” over the sight means the shot is blocked, while a “NO SHOT” means there is no possible chance the round can hit the target. Something that would cause this is if the ship were moving away from you at a speed greater than the velocity of the fired round.
I found that the lead-computed sight also made rendezvousing with the station reliably possible at ranges of about 500km. Any greater than that and a proper orbital intercept really needs to be calculated. That said, my best rendezvous range from eye-balling it has been 850-ish km.
With that, let’s talk about what’s left before the first early-access release. I need to set up the three must have missions (start-up and undock, dock and shut-down, and basic 1vs1 combat). These will need tutorial pop-up text to help you learn the proper sequencing. Also, there’s some more front-end UI work to do. Finally, there is damage application, reaction and the Caution&Warning system. It’s a tall order, but I think a mid to late-March release is still do-able. My goal is to have the bulk of the work finished within the next two weeks; so I imagine I won’t be posting much during that time.
With that. I know it’s been a long haul; but we’re almost there. As always, thanks for your patience and support!