I tried to find an answer on that several times with no luck. Uber, how does your space work? I know that you didn't do the space stuff yet, but you must have some basic design ideas already. Let's imagine a line, on one side of which is a purely real-physical world, in which all of the celestial objects move according to the Kepler's laws, and on the other the totally game-arcade world, in which planets are moving strictly by the orbits, and asteroids can move freely with some max speed. So where on the line will it be? I myself is a fan of the fair physics, although I understand that calculating an orbit of an asteroid to hit the planet in the exact spot will be challenging at least. But on the other hand, the cool gravitational slingshot which we saw in the kickstarter teaser will be pointless without, well, gravity.
That's the most interesting stuff actually I'd love to hear about the process at some livestream perhaps (and I think not me alone, but I'm too humble to say "we"). What you've tried, how it worked, or didn't work, and why. It will make a great material for your documentary anyway.
as would I! What ever you will disclose will get discussed / flamed to death in a multipage thread of course but that doesn't mean you shouldn't do it! In fact, if you just posted a few examples of things you've tried / would like to try without mentioning which you'd consider for actual implementation you might even contain aforementioned strong expressions of personal opinion to a minimum... Might want to do it anyway as some sort of social experiment ;-)
To my understandning alpha starts with focus away from celestial mechanics. Will the experimenting of differrent celestial mechanics going to be part of alpha/beta or is the general direction going to be picked beforehand?
This is definitely one of those features that will take more time than others to develop... while I'm excited to see them and do agree with you on this, I'd be much happier with Uber putting this sort of thing off until after the alpha releases (earlier alpha is fine with me!!) so that we can at least focus on the fundamentals in testing and improve those first before we get complex. Still, I'm really keen to see where this goes!
Actually, no. The development itself should be rather easy. The scientific approach requires only a couple of forces, while the arcade doesn't need even those. To make it fun on the other hand would be quite difficult indeed. Take a pole as a simple example (by "pole" I mean the north/south pole, the farthest point from the ecliptic). How are you going to drop an asteroid on it? To drop it in a cool way, the asteroid should move perpendicular to ecliptic, which is not an easy task from the design point of view.
One question that could be answered, I hope, is how the perspective is planned to be handled. Planetary rotation doesn't look like it is in yet based on the May 3rd livestream. This means that the view of the planet could be rotated and zoomed as if it was a standard map pasted on a sphere. What happens when the planet is rotating? Does your perspective slowly (quickly?) slide around the globe? Does your perspective stick to the globe and the rest of space rotate in the background relative to your current view? (My guess is this) Will the rotation of your perspective relative to the planet be static, i.e. the same speed relative to the ground or will it be faster in the direction of rotation vs anti-rotational?
Most modern games have at least two coordinate system spaces for objects in the game; object relative space and "world" relative space. There's also usually a UI space, and some concept of camera space but that's really just calculated on the fly by inverting the camera's world space matrix rather than something that "exists" as constant data. The main reason to do this is for floating point accuracy. The way floating point numbers work on computers the bigger (or smaller) a number is less accuracy you have and you always try to use the least accurate representation of a number you can get away with using for performance reasons. If you store the models for objects in game as if they're "at zero" then they get the most benefit of floating point accuracy. You then also store that model's position in world space which if it's far away from zero might be less accurate, but when rendered from the player's point of view won't loose accuracy in it's detail when the player is close to it. So, why am I talking about different spaces? We continue this abstraction such that each planet exists around their own zero, and units on their surface or in their orbit are kept track of relative to that planet space. Planets locations are stored in a "solar" (planetary) system space (likely with a sun at its center) so that the movement of planets have no effect on the accuracy of units on the surface of any of those planets. At the moment when you're looking at a planet you're moving the camera around in planet space, so the movement of the planet has no effect on the camera. And before you ask about the camera going from planet to planetary system space, we have to move stuff between different spaces all of the time. I mentioned the camera relative space because to render anything on screen every frame everything has to transformed from what ever space it exists in to a camera relative space. Move stuff between different coordinate spaces is what a good chunk of game programming is about.
Thank you for such a detailed answer. So, if the planet's rotation doesn't have any effect on the camera, it sounds like the view of the planet will be pretty much the same as for the live stream of the 3rd. Will the sun, planets, moons asteroids, etc be visible in the background if the planet's rotation around the sun and your rotation about its surface swings them into and out of the field of view if you are zoomed out enough not to have the entire planet blocking your view of the system?
Whatever they do I hope they consider it carefully. For me, this is a major selling point of the game and I hope they keep it dynamic like everything else. If they implement the physics on galactic bodies dynamically that will add so many more dimensions to the strategy this game will be getting played and modded for decades (unless there is a sequel)
By the way, my question was answered by the May 17th Livestream and it was awesome! The moon rotated into view, the sun and other planet could be seen in the background briefly as the camera was rotated... was very cool.
