Little Q 'bout the dropship thing

[chriso20]

heywo hl2'ers,

when gabe said th dropship physically simulated the telephone wires to move in the coastline video, i wonder if its just that the dropship moves and there's a particle simulator ( or what ever) that simulates the downforce which effects the wires OR that the downforce is the thing that keeps it up using an equation, eg, grav 800, downforce 1000 the thing will rise? get me?

And if the latter is true, i wonder if the distance from ground is used, so the closer you are the greater the upward force, if so that'd make for some cool stuff, like it flies across a canyon and it dips down into the canyon a bit


Also is this pic old? i never seen it before;

 

[NeLi]

That pic is official, isn't it?

 

[chriso20]

That pic is official, isn't it?

i got it off a yahoo review so i'd guess it'd have to be.

 

[Maxi]

That pic is official, isn't it?

it's a pic from 2003. Yes official.

 

[westie]

I have never seen that pic but i dont really look at many screenshots i just watch the movies.

 

[Roland Deschain]

From a physical stand point, the distance from the ground for an airplane is almost negligable if you are considering change in gravitation.

Even the astronauts in mir or the internation space station are experiencing almost the same gravitational force as we on the ground, and in an airplace the difference is even smaller.

 

[chriso20]

no i mean downforce from th fan blades. Like if you hold a hairdryer close to a table it pushes off more than 1m away.

 

[Roland Deschain]

no i mean downforce from th fan blades. Like if you hold a hairdryer close to a table it pushes off more than 1m away.

Ahh, that makes sense, but who knows what kind of Einsteinien fan blades those combines use . Good idea though, gonna watch out for that once I get HL2. Sigh.

 

[jet jaguar]

I think they just surrond the dropship with a bubble of influence. Where ever the dropship's BOI goes it disturbs the surround environment according to the characteristics given to it(like when Freeman bumps into a object). It would be too much to think it actually simulates fluid dynamics of moving air. Too many computing cycles for that level of reality.

 

[noluck]

i remember reading there is no water or air dynamics.. my guess is its just an animation sorta thing that plays when the dropship is over it or something

 

[DeltaBlast]

It's just an animation triggered by coming close to it.

 

[jet jaguar]

i remember reading there is no water or air dynamics.. my guess is its just an animation sorta thing that plays when the dropship is over it or something

So that's why the water doesn't ripple when the gunship starts it's shooting run towards Gordon.

 

[Tyrsenus]

gabe said th dropship physically simulated the telephone wires to move

It's not a dropship in the video. What you're talking about in the coastline video is the alien gunship. The dropship is a transport that has only been seen so far in concept art.

 

[Animal]

no i mean downforce from th fan blades. Like if you hold a hairdryer close to a table it pushes off more than 1m away.

Not true. It's just that the particles are reflected off the surface and add to the force.

The heights the combine aircraft is flying at will mean that this reflective force will have a negligible effect.

 

[(TK)Deltashadow]

Ahh, that makes sense, but who knows what kind of Einsteinien fan blades those combines use . Good idea though, gonna watch out for that once I get HL2. Sigh.

Lets get bernoulli and Einstein in on this one, and maybe they can get the Game release physics on track as well. LOL.

 

[(TK)Deltashadow]

no i mean downforce from th fan blades. Like if you hold a hairdryer close to a table it pushes off more than 1m away.

By the way just for a physics 101 lesson. Fan blades, or non fixed wing aircraft use on of two effects. The most common is the bernoulli effect were as the air on the top of the blades has more distance to travel(than on the bottom) and has to move faster to meet the air on the bottom, thus creating less air pressure on top creating lift. this means that in this case the desity of air effects how much lift will be produced(i.e. high elevation=less lift)
The second effect is ground effect which happens at a lot less altituded. This uses the bernoulli effect as well as the concept of hydroplaning or ice skating. Which means it takes advantage of fluid dynamics, basicly using the effect that low as a cross between backdraft and ball berings. This effect is best utilized and controlled on the water.(a hovercraft uses total ground effect to cause lift)
In the end it would kill a computer to acuratly represent either of these two physics concepts and even articifally you would see degredation in game play.
So in short its like neither, they apply the artifical physics to a single model(the dropship) like you could to people. Its preinstructed coded, the effects to the ground just happen to the ground when it is at a certain z coordinate away.

 

[-JeZ-]

By the way just for a physics 101 lesson. Fan blades, or non fixed wing aircraft use on of two effects. The most common is the bernoulli effect were as the air on the top of the blades has more distance to travel(than on the bottom) and has to move faster to meet the air on the bottom, thus creating less air pressure on top creating lift. this means that in this case the desity of air effects how much lift will be produced(i.e. high elevation=less lift)
The second effect is ground effect which happens at a lot less altituded. This uses the bernoulli effect as well as the concept of hydroplaning or ice skating. Which means it takes advantage of fluid dynamics, basicly using the effect that low as a cross between backdraft and ball berings. This effect is best utilized and controlled on the water.(a hovercraft uses total ground effect to cause lift)
In the end it would kill a computer to acuratly represent either of these two physics concepts and even articifally you would see degredation in game play.
So in short its like neither, they apply the artifical physics to a single model(the dropship) like you could to people. Its preinstructed coded, the effects to the ground just happen to the ground when it is at a certain z coordinate away.

Now THAT'S what I call a concise reply :thumbs:

 

[Yellonet]

i remember reading there is no water or air dynamics.. my guess is its just an animation sorta thing that plays when the dropship is over it or something

So this means that if you create a big cube with water in it and then blow a hole in it's side - the water wont pour out?? If so... ;(


- Yellonet

 

[Brian Damage]

Only game I can think of that might just possibly have anything like that is Black and White 2. Or at least, it is if you mean physically simulated water.

If you're talking a simple "pouring water" sprite and a decreasing water volume, I'm sure you could program it in. They did it in HL for the dam, IIRC...

 

[Yellonet]

Only game I can think of that might just possibly have anything like that is Black and White 2. Or at least, it is if you mean physically simulated water.

Yes, I mean physically simulated water... but I guess it wont happen :flame:

If you're talking a simple "pouring water" sprite and a decreasing water volume, I'm sure you could program it in. They did it in Hl for the dam, IIRC...

Not very hard to do in Worldcraft/hammer.. but it's not dynamic


- Yellonet

 

[Biscuit]

So this means that if you create a big cube with water in it and then blow a hole in it's side - the water wont pour out?? If so... ;(


- Yellonet

Unfortunately, Valve's stated that fluid dynamics will not be present in Half-Life 2.

 

[Yellonet]

Sad. Very sad. But I guess it must be quite a challenge to make it work right.


- Yellonet

 

[BlackBeltBob]

If an aircraft has the Bernoulli effect and the air on the top side of a wing indeed travels a further distance in a shorter timespan than the air on the bottom side of the wing, how could you ever make a loop?

The plane would be upside down, and the force pushing the plane up, would then be pushing it downward. This force is increased due to gravity pulling the plane down. The plane would need double the distance from earth at the startingpoint of the loop, in order to just scrape the ground at the end of the loop.

This principle has thus been questioned. Another, more viable, solution is the fact that the air behind the wing is moving downward. Because its moving downward, it pushes the plain up...

I heared this in the National Knowledge Quiz on TV. It's not imagined, my science and physics teachers were also wrong about it (just like the contestants, most of whom had Masters AND PhD's AND doctorates (like Prof. Dr. Ir. Freeman)).

 

[Yellonet]

If an aircraft has the Bernoulli effect and the air on the top side of a wing indeed travels a further distance in a shorter timespan than the air on the bottom side of the wing, how could you ever make a loop?

The plane would be upside down, and the force pushing the plane up, would then be pushing it downward. This force is increased due to gravity pulling the plane down. The plane would need double the distance from earth at the startingpoint of the loop, in order to just scrape the ground at the end of the loop.

This principle has thus been questioned. Another, more viable, solution is the fact that the air behind the wing is moving downward. Because its moving downward, it pushes the plain up...

I heared this in the National Knowledge Quiz on TV. It's not imagined, my science and physics teachers were also wrong about it (just like the contestants, most of whom had Masters AND PhD's AND doctorates (like Prof. Dr. Ir. Freeman)).

There are more than a couple of forces in action here, not just gravity and the Bernoulli effect. The law of inertia for one.


- Yellonet

 

[sHm0zY]

THe drop ship looks coolieo

 

[Logic]

I think people are underestimating (or not understanding) the Havok physics engine. It's definately true that there's no calculation of movement of air particles (that would be far too cpu intensive), but it's extremely easy to calculate directional forces, and for those forces to be proportional to the distance between two objects. It would be quite easy to code the dropship to realistically exert a force on physical objects beneath it, such as boxes, and powerlines (which are physically calculated, I believe, there's something about it somewhere in the Valve info thread), just as it would be easy enough to calculate the dropship's vertical movement according to gravity and thrust.

So basically, no true to life calculations of air movement are made, but that kind of accuracy is completely unneccissary in a game. The Havok physics engine is more than capable of mimicking the effects of much more complex physical calculations, just by using much more generalised calculations (such as applying a single force, thrust, from the dropship, rather than calculating the movement of air particles).

 

[Darg]

The wires have been said to be made up of points which are then joined by a bezier like curve, i.e. the wire itself. The amount of physics points depends on the length of the wire and how realistic you want it to act.
I am assuming that the easiest way of making the wire interact with the dropship would be to use quite a simple equation calculating how much force the dropship would apply to the wire and in what direction. The direction would simply be opposite to the direction of the dropship (Or towards the dropship if the rotors pull the wire towards them).
The amount of force would be a number depending on both objects (dropship and the wire) over the distance squared between the rotor point of the dropship and the physic points of the wire. This would give a nice proportional fall-off to the effect.
To not have this equation running when the dropship is half the map away they could just have it come into effect when the distance between the two objects drops below x distance.

It would provide the necessary simulation of realism without taking up too much CPU power. It could even be turned off for non necessary objects like wires on lower end computers. Honestly seeing that wire 30 metres up waving in the breeze of the dropship is just eye candy. It's not going to affect the game in any way apart from diverting your attention from the two dozen combine soldiers that just unloaded from the ship.

 

[killahsin-[CE]]

I think people are underestimating (or not understanding) the Havok physics engine. It's definately true that there's no calculation of movement of air particles (that would be far too cpu intensive), but it's extremely easy to calculate directional forces, and for those forces to be proportional to the distance between two objects. It would be quite easy to code the dropship to realistically exert a force on physical objects beneath it, such as boxes, and powerlines (which are physically calculated, I believe, there's something about it somewhere in the Valve info thread), just as it would be easy enough to calculate the dropship's vertical movement according to gravity and thrust.

So basically, no true to life calculations of air movement are made, but that kind of accuracy is completely unneccissary in a game. The Havok physics engine is more than capable of mimicking the effects of much more complex physical calculations, just by using much more generalised calculations (such as applying a single force, thrust, from the dropship, rather than calculating the movement of air particles).

This has to be the best post on top of the true aspect of what your discussion is about. Although I would like to add that HL2 is basically Havok 2.5 which is a combined venture between The havok team and Valve. It's really 2 but I only say 2.5 because I'm sure there will be some things valve will keep exclusively for hl2.

 

[Brian Damage]

I was pretty certain that the Havok code in Source was actually more like Havok 1, plus massive modifications by VALVe.

 

[Rupertvdb]

i think he means they used havok 1 and they have added bits and modded it to the point where it could be seen as 2.5.

I personally have no idea what level the physics are at, i hope to see them in action soon....sigh.

 

[Logic]

I was pretty certain that the Havok code in Source was actually more like Havok 1, plus massive modifications by VALVe.

My understanding is that many of the major changes and improvements Valve made to Havok served as the framework for Havok 2, though Havok 2 has been developed separately from HL2 for long enough to have slightly different features. So it's basically true that HL2's physics engine is a version of Havok 2, minus one or two of Havok 2's newer features, and possibly plus some features Havok 2 doesn't have.

 

[Brian Damage]

So... Havok 1.9 (+ .5 x Havok 2) + [WKW*]... :E ?

(*"Who Knows What")