The Star Citizen Thread V10

[Cobra1984]

Sure, right after you post a video of a starfarer that's zipping and buzzing and changing direction like a cardboard cutout regardless of speed or momentum

Is it modelling inertia and momentum or isn't it?

edit - here's a video of a massless Starfarer landing
https://youtu.be/S7B-WY65gIo?t=1375

So the weight of the Starfarer is suppposed to be around 4,000 tonnes. That would require something like 3,500 kilo newtons of thrust to do all of those little sideways motions you see in the video as the player lines it up on the pad. In fact as the video shows it's taking pretty much zero time to achieve several metres of lateral movement.

So the brain looks at this and perceives the ship as having no weight. This is why traditional model animators slow model footage down by 4-6 times, to give the model some weight.

What CIG have done with the 3.5 flight model is to increase those goal times to lend more weight to the models, in an attempt to mitigate this massless movement. That doesn't mean they are modelling inertia, which if they were, would have saved them YEARS of work and given them something resembling an actual flight model.


edit2 - in fact they've set this ship up with a HUGE goal time to reach full forward speed, yet it can blip sideways back and forth in an instant, which makes even less sense than I originally thought. What an absolute mess, they really don't know what they're doing.

 

[Agony_Aunt]

I rarely pop into this tread anymore

Be careful, Preston. You're treading on my dreams.

 

[sovapid]

Is it modelling inertia and momentum or isn't it?

edit - here's a video of a massless Starfarer landing
https://youtu.be/S7B-WY65gIo?t=1375

So the weight of the Starfarer is suppposed to be around 4,000 tonnes. That would require something like 3,500 kilo newtons of thrust to do all of those little sideways motions you see in the video as the player lines it up on the pad. In fact as the video shows it's taking pretty much zero time to achieve several metres of lateral movement.

I dunno, that video is pretty short and the starfarer is only moving a few meters. I guess it could be better.

I was thinking something more along the lines of this:

https://www.youtube.com/watch?v=ffH1Ixfh2ng

but where the starfarer is zipping and buzzing and changing direction like a cardboard cutout regardless of speed or momentum. And is actually able to dog fight the m50 and gladius at the end instead of floating around like a giant potato.

Its hard to tell in the video, but most of the time I was at full deflection on the stick/pedals trying to strafe as fast as possible. Especially at the end, I am strafing up as fast as the starfarer will go for most of it trying to get away.

 

[Zyll Goliat]

I dunno, that video is pretty short and the starfarer is only moving a few meters. I guess it could be better.

I was thinking something more along the lines of this:

https://www.youtube.com/watch?v=ffH1Ixfh2ng

but where the starfarer is zipping and buzzing and changing direction like a cardboard cutout regardless of speed or momentum. And is actually able to dog fight the m50 and gladius at the end instead of floating around like a giant potato.

Its hard to tell in the video, but most of the time I was at full deflection on the stick/pedals trying to strafe as fast as possible. Especially at the end, I am strafing up as fast as the starfarer will go for most of it trying to get away.

So this is how new FM looks like with the huge ship as Starfarer?

 

[sovapid]

So this is how new FM looks like with the huge ship as Starfarer?

Nah, its the current flight model. The video is from January, so some version of 3.4.

 

[Cobra1984]

I dunno, that video is pretty short and the starfarer is only moving a few meters. I guess it could be better.

I was thinking something more along the lines of this:

https://www.youtube.com/watch?v=ffH1Ixfh2ng

but where the starfarer is zipping and buzzing and changing direction like a cardboard cutout regardless of speed or momentum. And is actually able to dog fight the m50 and gladius at the end instead of floating around like a giant potato.

Its hard to tell in the video, but most of the time I was at full deflection on the stick/pedals trying to strafe as fast as possible. Especially at the end, I am strafing up as fast as the starfarer will go for most of it trying to get away.

Is the flight model modelling inertia and momentum or is it a time based goal system?

It's one or the other.

Btw - the video I linked shows it changing direction without any inertia or momentum, like a cardboard cutout.

 

[sovapid]

Is the flight model modelling inertia and momentum or is it a time based goal system?

It's one or the other.

Btw - the video I linked shows it changing direction without any inertia or momentum, like a cardboard cutout.

Against my better judgement, i'll bite.

So why does it have to be one or the other?

The system models inertia, mass and acceleration. The time based goals system calculates the amount of force that the thrusters need to produce in order to achieve the desired acceleration for a given mass.

Why can't this be a thing?

Granted it can result in the thrusters producing a ludicrous amount thrust, but it is still "modeled".

 

[Cobra1984]

Against my better judgement, i'll bite.

So why does it have to be one or the other?

The system models inertia, mass and acceleration. The time based goals system calculates the amount of force that the thrusters need to produce in order to achieve the desired acceleration for a given mass.

Why can't this be a thing?

Granted it can result in the thrusters producing a ludicrous amount thrust, but it is still "modeled".

Because if you're modelling inertia, mass, momentum then you don't NEED a time based goal system.

 

[sovapid]

Because if you're modelling inertia, mass, momentum then you don't NEED a time based goal system.

Now that I can agree with.

Kinda like having no loading screens, but putting in long boring game mechanics that might as well be a loading screen.

 

[Gregstrom]

Because if you're modelling inertia, mass, momentum then you don't NEED a time based goal system.

Sorry, but bearing in mind the project history, why on Earth do you think that means they haven’t done both?

 

[Surefoot]

Sure, right after you post a video of a starfarer that's zipping and buzzing and changing direction like a cardboard cutout regardless of speed or momentum

Yes it does, regarding its supposedly huge mass, it's moving like a ballerina. It should take quite a few kilometers just to brake from cruising speed...

Sorry, but bearing in mind the project history, why on Earth do you think that means they haven’t done both?

From what i know they did do both but the time-based goal linear movement completely overrides any physics. The thruster values are adjusted in function of these goals and become beyond ludicrous.

 

[Asp Explorer]

Naah - the thrusters have been overclocked and punch above their weight

 

[sovapid]

Yes it does, regarding its supposedly huge mass, it's moving like a ballerina. It should take quite a few kilometers just to brake from cruising speed...

Yeah, if you watch the video I posted, I start slowing down from top speed 16km out.......

And never get anyway near cruise speed again through out the whole video.

 

[Tippis]

The system models inertia, mass and acceleration.

No. The system creates parameters that makes the underlying physics engine yield a result that is consistent with a predetermined outcomes. Basic underlying factors such as inertia, mass, and acceleration are outcomes of those parameters, so that something can be fed into CE, rather than being the actual fundamental mechanical constraints they should be — they're not actually modelled because the system exists expressly to make them irrelevant.

A ship with a mass of 1kg can have the same behaviour as a ship with a mass of 1 gram or 1 ton, because the system is there to ensure that mass is not a factor in the outcome — only the goaltime matters.

If the system modelled mass and inertia, acceleration would be a function of force and mass, and manoeuvring time would be a function of acceleration. That is not how SC works, per CI!G(G)'s own descriptions. Rather, it asserts a manoeuvring time, acceleration is a function of that, and force is horribly mangled into a function of acceleration. It is entirely backwards and wholly divorced from any kind of proper physical modelling.

Why can't this be a thing?

Because the design goal of the entire flight system is to remove such troublesome things as physical parameters and to boil them down to the single “designer friendly” parameter of a goal time.

 

[McGlashan]

I was wondering how it's supposed to be "designer friendly", instead of simply and intuitively saying "this needs to feel heavier" and cranking up the mass, or "this needs to yaw faster" and turning up the respective thruster... thrust.
But then I remembered it's CR making these calls, and of course he probably thinks it makes him sound clever to say dumb things like "I want it to yaw 60 degrees in half a second."

 

[Tippis]

I was wondering how it's supposed to be "designer friendly", instead of simply and intuitively saying "this needs to feel heavier" and cranking up the mass, or "this needs to yaw faster" and turning up the respective thruster... thrust.
But then I remembered it's CR making these calls, and of course he probably thinks it makes him sound clever to say dumb things like "I want it to yaw 60 degrees in half a second."

The horror stories from behind the scenes, where the backers who were deconstructing the flight model and providing more sensible performance profiles (yes, this happened — their suggestions made it into a patch) and asked why things worked this way, was that “the ship makers didn't grasp what m/s² meant”, so something they did understand was required to give them something to tweak. Completion time for [whatever the base manoeuvre is] was arrived at as the simplification they could get their heads around.

Granted, this is something like third-hand information, so a slight salt mine might be required… but since this is CIG we're talking about, I'm sure it was actually worse than that.

 

[sovapid]

Got em!

Not enough simulated enirsha!

[video=youtube;oWpdOsVWidc]https://www.youtube.com/watch?v=oWpdOsVWidc&feature=youtu.be[/video]

 

[Borgkube]

Well it’s internal in the company so don’t think so, unless they change profession.
However, when I was very passionate about the project I was also very angry, but now I don’t care and can talk about it in an objective POW, still I always get the same follow up questions, “why do they keep making the same mistake?” and my answer is always, I don’t know, and I still can’t answer this question, why keep making the same mistake repeatedly!?

Because it either always was, or has become a SCAM.

 

[[VR] DDM_Reaper20]

The people who answered your question don't actually play the game.

In a small light ship like a razor there is no perceived inertia. It will buzz around like a hummingbird.

If you fly a larger ship like a starfarer, there is more inertia than I like. You really have to plan ahead for stopping. It is like captaining an ocean liner. Somewhat similar to flying a cutter in ED, but much worse.

Even in a small, light fighter, there SHOULD be inertia. Heck, with a Sidewinder I cannot pull manoeuvres such as you pulled. Not even with an SLF, and that's really quite a bit smaller than anything floating around (hah!) in the SC "verse" (half a solar system, to be honest...). Unless they somehow, magically, found a way to build inertia compensators that totally negate inertia (no idea how that should happen), mass remains mass.

Your example only shows that they simply adjust values by hand, which is even worse than I thought.

@ Tippis

Thanks for explaining that -- though I'm wondering why they chose to override those values in the first place. Doesn't make much sense to me, but then, what do I know about game development?

@ Cobra1984

Well explained, even a layman like I could understand that.

 

[Surefoot]

Yeah, if you watch the video I posted, I start slowing down from top speed 16km out.......

And never get anyway near cruise speed again through out the whole video.

I have flown one, i know pretty well how it moves, i also possess a Retaliator which is also quite a huge monster of a ship, these have a perfect linear movement, with near-infinite instant accelerations (when changing directions).. And yes, the Razor or the Hornet should have quite a bit of inertia, being between 30t and 70t.. Those tiny tiny manoeuver thrusters at the moment produce more power than multiple Saturn V's, each. Without any disturbance around, by the way...

The horror stories from behind the scenes, where the backers who were deconstructing the flight model and providing more sensible performance profiles (yes, this happened — their suggestions made it into a patch) and asked why things worked this way, was that “the ship makers didn't grasp what m/s² meant”, so something they did understand was required to give them something to tweak. Completion time for [whatever the base manoeuvre is] was arrived at as the simplification they could get their heads around.

Granted, this is something like third-hand information, so a slight salt mine might be required… but since this is CIG we're talking about, I'm sure it was actually worse than that.

Almost that much has been told by the lead designer in charge of the new FM, so it's first hand info..

Thanks for explaining that -- though I'm wondering why they chose to override those values in the first place. Doesn't make much sense to me, but then, what do I know about game development?

They chose to do that because they wanted an arcade space shooter, with FPS-like movement (like in Quake or Unreal Tournament) since the main developers in charge of the project used to be FPS developers in charge of an FPS engine...