Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ildifa]

I agree with Sniperton, the plots you posted raise a few questions... It doesn't seem to me that they are just mislabeled.

In very general terms, the lift should start with a negative value at -15 AoA, and reach zero at a slightly negative AoA (as the wings have a built-in angle of incidence); increasing more or less linearly until the max lift AoA (which is the stall angle of attack, let's say 20°), after which lift goes down abruptly.

The drag plot should start slightly positive, reach a minimum around the AoA where lift is zero, then increase in a less steep curve than lift; at the stall AoA drag starts to rise sharply.

Here is attached an example plot:



I took it from a videogame, so it's not super accurate, but I think it explains well what I was trying to describe!

Sorry if I've been too wordy, or explained stuff you already know   

Cheers

[Stainless]

Yes, I think something is very wrong.

I will have to look at it tonight.

What I think is happening is that the code is rotating the resultant forces into the world frame when I want them to be in the local frame.

So the graphs would be upside down to start with , and also transformed by the pitch of the aircraft

Then it starts to make a bit more sense. You can see the wing stall in the lift graph , and the drag would be the right general shape, maybe with an offset I am missing.

Anyone know roughly what aoa the spitfires wing stalls at?



 

[sniperton]

The Spitfire wing is somewhat special, but for your preliminary test you can take a number between 15 and 18 degrees.

[Stainless]

Well this looks a bit better.

The forces were in the wrong coordinate set.

[sniperton]

Lift is a matter of opinion and judgement (and I leave it to the savants), but I personally doubt that stalling the wings makes just a small bump down.
On the other hand, drag is definitely not what I would expect. It's hard to believe that drag at -10 could be way off the value at +10 degrees of AoA. The values should be largely the same with a small margin (the differences being due to wing incidence and airfoil shape).
Just my uneducated guess 

[Stainless]

Experiments with my new deferred renderer and FFT water

[hello]

Perfect storm brewing...   

[slibenli]

In case you haven't seen it:



I will probably integrate this into il2ge at one point.

Demo with code: http://www-evasion.imag.fr/Membres/Eric.Bruneton/OceanLightingFFT.zip
Paper: http://www-ljk.imag.fr/Publications/Basilic/com.lmc.publi.PUBLI_Article@125fe8a8322_1ac3379/article.pdf

Another one - looks a bit more natural I think

[Pursuivant]

Is there any way to easily get white combers and wind spray on the top of the waves? Curling waves that "break" at the top?

From altitude, it's far easier to to determine sea state by the white wave crests than from viewing wave motion. If you're trying to land on or near water, the white crests also give you a sense of wind direction and speed, as well as dangerously shallow areas and wind-sheltered areas.

[Stainless]

I am working on that, at the moment I am using the steepness of the surface normal, which is just wrong.

I need to work out the derivative of the surface normal, so where the normal rapidly changes direction is where the foam will form.

Of course this need to be at the top of the wave, so I need to use the sign of the derivative as well.

[Pursuivant]

I am working on that, at the moment I am using the steepness of the surface normal, which is just wrong.

Any possibility of just using the elevation of the wave? It's very rare that you have low waves with combers on them, unless it's something unusual like downdraft from a helicopter's blades or very high winds across a narrow body of water.

I need to work out the derivative of the surface normal, so where the normal rapidly changes direction is where the foam will form.

Couldn't you just use wind direction? Unless the waves are getting pushed up by a beach or reef, the combers will be on the downwind side of the wave.

Arguably, that would be better physics modeling, since the foam/comb at the top of the wave is formed by more or less straight-line air currents pushing the wave top.

Of course this need to be at the top of the wave, so I need to use the sign of the derivative as well.

Not necessarily. If you look at videos of really agitated seas, like in a Beaufort Force 10+ Storm/Hurricane, the entire sea surface appears to be white. The main thing is the degree to which the water's surface is whipped by wind, creating froth.

Of course, the real trick is getting a decent looking effect while still being thrifty with graphics processing power required.

[Stainless]

So I get the derivative of the normal to find the potential points to add foam.

Then get the dot product of the normal and the wind direction to filter those potential points to points that are down wind.

And vary the decision variable of the last stage based on the strength of the wind.

so in pseudo code

Code: [Select]

  float foam = 0;
  if (delta_normal > constant1)
 {
      float dw = dot(worldspacenormal,winddirection);
      if ((dw > 0)&&(dw > constant2))
      {
           foam = dw;
      }
 }