Armoured glass refraction must see video

[VIMANAMAN]

Hrm, I am not so sure. A resolution 0.5 degrees would make an enemy aircraft probably skip up and down quite a bit if you move your head while aiming through the gun sight ...

 

But, yeah, what do we know. Maybe something like this is really "cheap" as performance is concerned and the resolution could be made so fine that it really would not matter too much anymore.

 

edit:

skipping? err ... a more general question: if the amount of "view shifting" is depending on the view angle, why is that not a problem in the real aircraft when using the gun sight? The target would move optically, but the guns sight reticle would not as it is not refracted and instead projected to infinity?

 

edit2:

Perhaps because the FoV when looking through the gunsight glass is so narrow that the change of view angle would be in fact negligible?

 

Yeah - maybe it is workable - I really don't know :) - it would be great if it was... even a compromise solution would be great if it eliminated that last, probably less than 1 degree of visibility over the nose. But I honestly don't know if its possible.

 

One advantage performance wise is that the Dora doesn't have mirrors (or a TGP :megalol:), so there's no double / triple whammy... :)

 

Anyway cheers for the interest in the idea. And I love the Dora as it is but any improvements along these lines would be great.

[Bushmanni]

If front and back faces are parallel in glass (like it is in this case) effects of refraction are a apparent shift of the viewpoint or "camera" and compression of the insides of the glass. The apparent shift of the camera needs to be calculated only once per frame. You can achieve this effect with same tricks as portals in Portal and using separate compressed 3d model for the inside edge of the glass. How doable this is in current DCS engine or EDGE is another matter.

[sobek]

Correct, could use tables whit pre calculated values though. increments of 0.5 degrees might do?

In the end you need the result, there's no need to calculate it on the fly.

 

It's not that easy. The angle is not uniform over the whole area so the image is distorted. You won't get away doing it properly by just shifting the image.

[NineLine]

Yup, I think the only way to truly do it, is with a real refraction effect, and I am not sure it would be efficient performance wise right now and I doubt the engine supports refraction anyways.

[VIMANAMAN]

It's not that easy. The angle is not uniform over the whole area so the image is distorted. You won't get away doing it properly by just shifting the image.

 

Probably cost of development / end user performance would make this not worth it... But it could be 'fudged' I guess to just work in the vertical axis - solving the problem...

 

I know DCS don't do 'fudging'... :)

[Random]

guessing it would be at least as resource hungry as the mirrors? perhaps in a few years?

 

Also maybe its not such a problem as people think.... as it effects all aircraft! We should be able to see a little further over the nose of the P51 too. Although i agree its not as big a deal as with the dora.

[VIMANAMAN]

guessing it would be at least as resource hungry as the mirrors? perhaps in a few years?

 

yeah I would guess you're right but we're only guessing.

 

Also maybe its not such a problem as people think.... as it effects all aircraft! We should be able to see a little further over the nose of the P51 too. Although i agree its not as big a deal as with the dora.

 

It affects the real world view from the front of the FW190 D9 - that's all I need to know.

Edited August 28, 2014 by VIMANAMAN

[aaron886]

That's a classic German proof! Engineering meticulousness. How can you even argue against it?

[genbrien]

a bit OT, but what would be happenning to the refraction if the glass get shot and cracked?

[sobek]

How can you even argue against it?

 

Why don't we use real time Navier-Stokes by now for flight dynamics? :)

[Bushmanni]

It's not that easy. The angle is not uniform over the whole area so the image is distorted. You won't get away doing it properly by just shifting the image.

 

The ray angle inside the glass is different but it always exits the glass at the same angle as it entered if the faces are parallel hence no distortion but apparent shift of the viewpoint but no shift of the image. What the angle difference does is make the glass look thinner as it really is (things only inside the glass seem compressed or "distorted"). This effect is the reason why the front edge looks thinner through the glass.

[Sceptre]

Who knows maybe EDGE will allow it

[Hans-Joachim Marseille]

Video at 08:00 mark ... so after all potential refraction modeling efforts we will be looking at the cowling as view limiter LOL!

[Flagrum]

An other thought ... refraction of glass - not every glass is the same and different types of glass refract light differently, doesn't it? So, the effect of the "optically shrunken" frame in the video could be in reallty even smaller ... or bigger, right?

[Hans-Joachim Marseille]

An other thought ... refraction of glass - not every glass is the same and different types of glass refract light differently, doesn't it? So, the effect of the "optically shrunken" frame in the video could be in reallty even smaller ... or bigger, right?

And since he also used 1.5 the thickness of the original in his experiment, the original refraction (distance) would have been much less.

[Sceptre]

And since he also used 1.5 the thickness of the original in his experiment, the original refraction would have been much less.

 

This, and is it just me, or is the D9's front glass not as steeply angled as the guy put it in the video? I may be wrong as I don't have the DCS FW-190, just going by youtube videos here...

[Cedaway]

To properly model this effect, you would use Ray-Tracing. But it's not a 'real-time' rendering solution since it consume huge amount of calculation...

[Fox One]

How much the refraction is shifting the image vertically can be approximated reasonably well from such pictures (A-5 and D-13):

 

[bongodriver]

The pictures also show clearly the 'bar' is likely to obscure the bottom of the revi.

[IJG27_Nemesis]

Hi mates,

Sorry to lift the matter of refraction again but this german guy has made a fantastic video to demonstrate this effect and it's simply a must see video.

 

 

 

For some reason I cannot post the video so if some one could please do.

 

 

Hi Pilots!

 

There is also a written thread with plenty of detailed pictures, which I think is worth reading.

 

I've called it "The Ultimate Fw190 photo evidence thread".

 

check it out here:

http://forum.il2sturmovik.com/topic/9152-ultimate-fw190-photo-evidence-thread/

 

 

Best regards,

Nemesis

Edited August 28, 2014 by I/JG27_Nemesis

[NineLine]

We dont need anymore evidence, we need to know how you make it look like this in a game engine ;)

[IJG27_Nemesis]

We dont need anymore evidence, we need to know how you make it look like this in a game engine ;)

 

 

Hi SiThSpAwN!

 

I don't know about the current capabilities of modern 3D-Engines, but what I know from 1999 is the Unreal Engine of the original game Unreal by Epic.

 

you see that "door" - It is a teleporter and when you look on its entry you see something different than what really is behind the teleport-door - this should work for the windscreen effect as well - at least it is worth taking a look at the technology back then (unreal engine 1 from the year 1999).

 

 

you saw a different room than the space which really was behind that door - if now the door would have the frame of the INNER windscreen frame, and behind that you see a separate "room" that is exactly so but only 37mm lower...? ;)

Edited September 14, 2014 by I/JG27_Nemesis

[NineLine]

I dont think that is the same effect you want here...... you want an actual live, dynamic refraction effect.

[zaelu]

I think the refraction should be way way way easier to simulate than flight dynamics of FW190 as depicted in DCS.

 

Just looking at this picture and makes you laugh remembering Oleg :D

 

[Bushmanni]

All the refraction does is show the world from a little shifted viewpoint so the technique in the unreal engine video is capable of reproducing essentially what some brute force simulation of photons would give as an image. The compression effect can be simulated with separate compressed 3d model of the window frame for the refracted viewpoint as there's not going to be anything but glass inside the glass. If you want cracks in the window for damage model you can make them also compressed. If you want to make the refraction simulation really hifi you can scale the window frame model in axis perpendicular to the window plane to simulate the slight difference of apparent compression when the viewing angle changes. This effect is very small though and is visible to the eye only at very shallow angles. You can actually see it in the demonstration video if you look closely. But IMHO simulating it is overkill as I doubt it's noticeable from the pilots seat.