KSP2 EA: About the warp (time zoom)

[Vl3d]

I've been thinking about it and I think this subject needs to have it's own thread.

Nate said that time warp will be using a "non-eliptical solver" - I think he's referring to the on-rails warp. But remember that in KSP1 we can physics warp under acceleration.

So what about the new physics warp? Will it still be a separate thing? Is it still going to be buggy?

Will on-rails warp and time warp be merged? Will non-eliptical solver warp be allowed in atmosphere? Will we still have speed limits depending on altitude? Any theories or evidence?

Edited January 13, 2023 by Vl3d

[Bej Kerman]

KSP 1 persistent thrust constantly changed your orbit as you accelerated. I think all they mean by non-elliptical is that if your ship accelerates, the game will predict how your trajectory curves under thrust, and make the rail your ship follows under tinewarp conform to this.

[GoldForest]

There's little to no information about it, so all we can do is speculate and base our opinions on things like Persistent Thrust for KSP 1.

Physics warp will probably be in game. It will probably allow for faster than 4x speed as well, though not too much higher. Probably 8x speed. A new physics warp system will be needed if they add plants such as Super Earths or Mini Neptunes. Seriously, imagine roving around on a Mini Neptune, that would be slow and boring without Physics Warp. 

It will of course be separate from Time Warp. Time Warp can't be used in atmo, and I doubt it will be buggy because it's being made by professionals and not backyard programmers.

Time Warp and "on-rails warp" is the same thing. I doubt Time Warp will be allowed in Atmo since that would cause so many problems.  

Speed limits? I think you mean warp limits, in which case, yes, because the closer to the planet, the more the game has to render and that can cause issues. 

[Vl3d]

7 minutes ago, GoldForest said:

Seriously, imagine roving around on a Mini Neptune, that would be slow and boring without Physics Warp.

Neptune is composed primarily of gases and liquids, it has no well-defined solid surface.

[The Aziz]

I assume physwarp to still be there. Timewarp under acceleration may still lock the controls, you point the the right direction - now that maneuver nodes are not instantaneous I believe the marker will point you in average direction for the entirety of the burn - you start the burn, you warp, you go out of warp, you stop the burn. No pitching, yawing, rolling in the meantime. And it won't be available in atmo, (unless you're stationary) because there's more than just the engines affecting your ship, the ground, the air, all these constantly changing your vector. So you're still limited here to physwarp, and while I don't think it'll cause joint breaking or other weird behavior, you're still asking for trouble if you overuse it, as slight touch on the controls may end up with flipping rolling crashing - you're going 4x normal speed, so you have 4x less time to react, and your half a second long button push is actually 2 seconds long push for the craft. And I hope physwarp to still be available in space because I ain't wasting several minutes on turning my interstellar ship around for deceleration. Preferably using SAS controls. But, again, any input may cause unwanted results even without breaking the ship in half.

Edited January 13, 2023 by The Aziz

[GoldForest]

8 minutes ago, Vl3d said:

Neptune is composed primarily of gases and liquids, it has no well-defined solid surface.

Neptune, yes, but there are giant ice+rock planets which are called "Mini-Neptune" due to their size. Too big to be "Super Earth" and too small to be "Gas Giants." Those are the planets I'm referring to. 

Edit: Some Mini-Neptunes have water, so 'roving' is probably the wrong word. 'Subbing' is probably a better term... though, would the probe still be called a rover if it's a submarine? Submarine rover? So, 'roving' might be appropriate... hmmm.

5 minutes ago, The Aziz said:

And I hope physwarp to still be available in space because I ain't wasting several minutes on turning my interstellar ship around for deceleration. 

If Persistent rotation is in game, you could use time warp to turn around. 

Edited January 13, 2023 by GoldForest

[Bej Kerman]

7 minutes ago, The Aziz said:

I ain't wasting several minutes on turning my interstellar ship around for deceleration.

This will probably work like in SimpleRockets2/Juno. 

[The Aziz]

[referring to persistent rotation]

I don't think it will be. The artificial gravity rings may be completely unaffected by timewarp, and I still think that controls will be off during warp. There was no word on "you can control your ship during timewarp" other than acceleration, so I assume there's no such thing.

Edited January 13, 2023 by The Aziz

[GoldForest]

Just now, The Aziz said:

I don't think it will be. The artificial gravity rings may be completely unaffected by timewarp, and I still think that controls will be off during warp. There was no word on "you can control your ship during timewarp" other than acceleration, so I assume there's no such thing.

Well, like I said, persistent rotation may be in game, not persistent controls. You could initiate the rotation, time warp, then un time warp to halt the rotation.

[Bej Kerman]

9 minutes ago, The Aziz said:

[referring to persistent rotation]

I don't think it will be. The artificial gravity rings may be completely unaffected by timewarp, and I still think that controls will be off during warp. There was no word on "you can control your ship during timewarp" other than acceleration, so I assume there's no such thing.

Or it's just another basic thing alongside the ability to jump when on EVA, another thing the dev team hasn't explicitly confirmed. SR2 devs could do it, Intercept probably can as well.

Edited January 13, 2023 by Bej Kerman

[Vl3d]

So if we want to merge physwarp and timewarp one of the main issues is with control inputs.

Because the game could theoretically decide on its own if it only calculated on-rails trajectories or all the local physics at an accelerated pace depending on the view (ship or map view) and position (if in atmosphere) and warp factor.

Maybe merge them like this: up to 4x always use physwarp (which allows control inputs) and for 8x+ use the non-eliptical solver?

Persistent rotation should 100% be in the game. Also higher warp factors. And I would love to also have slowed warp (0.1x .. 0.25x .. 0.5x).

Edited January 13, 2023 by Vl3d

[GoldForest]

4 minutes ago, Vl3d said:

So if we want to merge physwarp and timewarp one of the main issues is with control inputs.

Because the game could theoretically decide on its own if it only calculated on-rails trajectories or all the local physics at an accelerated pace depending on the view (ship or map view) and position (if in atmosphere) and warp factor.

Maybe merge them like this: up to 4x always use physwarp (which allows control inputs) and for 8x+ use the non-eliptical solver?

Persistent rotation should 100% be in the game. Also higher warp factors. And I would love to also have slowed warp (0.1x .. 0.25x .. 0.5x).

You can't really combine the two. 2x physics warp is not the same as 2x time warp. Different calculations completely. 

[Bej Kerman]

I hope timewarp is "smooth" for a lack of better word, like in Universe Sandbox. Stepped warp gets messy when you're trying to ff to a maneuver and have to predict how fast the game's going to decelerate. 

[Vl3d]

Just now, GoldForest said:

2x physics warp is not the same as 2x time warp. Different calculations completely. 

Physwarp does what timewarp does and also extra calculations. You can replace timewarp with physwarp for the 1x up to 4x or 8x speeds.

When it comes to timewarp 99% of the time I use the automatic "warp to position / maneuver node". I really hope this is used more to streamline the experience. I find manually controlling warp kind of fiddly and annoying.

[GoldForest]

Just now, Bej Kerman said:

I hope timewarp is "smooth" for a lack of better word, like in Universe Sandbox. Stepped warp gets messy when you're trying to ff to a maneuver and have to predict how fast the game's going to decelerate. 

To be smooth, it might have to be implemented I'm every step. 2x, 3x, 4x... 15x... 30x, 31x... etc

[Bej Kerman]

Just now, Vl3d said:

Physwarp does what timewarp does and also extra calculations. You can replace timewarp with physwarp for the 1x up to 4x or 8x speeds.

Er, no. Normal physics for physwarp, your vessel's orbit is a result of experiencing a gravitational force. For rails warp, the game calculates a fixed orbit from your current speed and position and places you on a rail according to that orbit.

Just now, GoldForest said:

To be smooth, it might have to be implemented I'm every step. 2x, 3x, 4x... 15x... 30x, 31x... etc

Why? Did the US devs have to implement every step from 1.001x, 1.002x, etc.? Nope, they just didn't use steps in the first place. KSP 1 already had smooth warp to smoothly transition between steps, it just didn't let you manually choose how fast your game goes.

[magnemoe]

27 minutes ago, GoldForest said:

Well, like I said, persistent rotation may be in game, not persistent controls. You could initiate the rotation, time warp, then un time warp to halt the rotation.

As I understand rotation is preserved through time warp.  Now I do not know how this will work, if you spin up an ship so it rotates every 20 seconds having an .02 second rotation speed at 1000x warp would be weird  
So I assume ship stops rotating and continues ones you leave warp. 

[Vl3d]

12 minutes ago, Bej Kerman said:

I hope timewarp is "smooth" for a lack of better word, like in Universe Sandbox. Stepped warp gets messy when you're trying to ff to a maneuver and have to predict how fast the game's going to decelerate. 

Exactly.. that's why I prefer autowarp. In the interface I've seen the steps of warp factors, it's not a slider.

9 minutes ago, GoldForest said:

To be smooth, it might have to be implemented I'm every step. 2x, 3x, 4x... 15x... 30x, 31x... etc

It also depends if the type of numeric calculation (if the warp factor is integer of float).

9 minutes ago, Bej Kerman said:

Normal physics for physwarp, your vessel's orbit is a result of experiencing a gravitational force.

Yeah but that's what the new non-eliptical solver is supposed to do - account for accelerations. So it more closely resembles physwarp now, but only applies to the craft as a whole I assume.. To calculate forces on individual parts you still need physwarp - but that could easily replace the 1x .. 8x steps of timewarp.

Edited January 13, 2023 by Vl3d

[GoldForest]

7 minutes ago, Bej Kerman said:

Er, no. Normal physics for physwarp, your vessel's orbit is a result of experiencing a gravitational force. For rails warp, the game calculates a fixed orbit from your current speed and position and places you on a rail according to that orbit.

Why? Did the US devs have to implement every step from 1.001x, 1.002x, etc.? Nope, they just didn't use steps in the first place. KSP 1 already had smooth warp to smoothly transition between steps, it just didn't let you manually choose how fast your game goes.

Sorry, what I meant was the system would probably need to go through all the steps to be smooth. Not necessarily stopping, but instead of jumping from 100 to 1000, it could go through all the steps.

 

3 minutes ago, Vl3d said:

 

Yeah but that's what the new non-eliptical solver is supposed to do - account for accelerations. So it's more closely resembles physwarp now, but only applies to the craft as a whole I assume..

Accleration on rails and acceleration off rails are two separate monsters though. One calculates real time data and the other calculates predetermined routes. 

[Bej Kerman]

9 minutes ago, GoldForest said:

Sorry, what I meant was the system would probably need to go through all the steps to be smooth. Not necessarily stopping, but instead of jumping from 100 to 1000, it could go through all the steps

Why do steps need to be involved?

[Vl3d]

9 minutes ago, GoldForest said:

Acceleration on rails and acceleration off rails are two separate monsters though. One calculates real time data and the other calculates predetermined routes. 

Good point.. but physwarp also calculates trajectory changes (as displayed in map view when flying in the atmosphere).

[GoldForest]

1 minute ago, Bej Kerman said:

Why do steps need to be involved?

I mean, I don't see the point in letting someone choose to go 5.52x speed. Why allow decimals? 

Just now, Vl3d said:

Good point.. but physwarp also calculates trajectory changes (as displayed in map view when flying in the atmosphere).

Yeah, real time data and not on-rail predetermined data. 

[Bej Kerman]

4 minutes ago, GoldForest said:

8 minutes ago, Bej Kerman said:

Why do steps need to be involved?

I mean, I don't see the point in letting someone choose to go 5.52x speed. Why allow decimals? 

I don't get your point. If you could explain why this is at all a bad thing, please do

[GoldForest]

Just now, Bej Kerman said:

I don't get your point. If you could explain why this is at all a bad thing, please do

I'm not saying it's bad, I'm saying to be smooth the time warp system might have to go through all the numbers instead of instantly jumping around. 

[Bej Kerman]

Just now, GoldForest said:

I'm not saying it's bad, I'm saying to be smooth the time warp system might have to go through all the numbers instead of instantly jumping around. 

So just round it up or down. Never did Squad have to manually add 76,825x warp (et al) for the transition between 100,000x and 50,000x warp.