Friday update and discussions 2011-02-25

[B25Mitch]

Hi everyone. I see there's been some discussion about the gap in shadows, close to the base of the object casting them.



This is an inherent limitation of texture-projection type shadowing. Here is a quick example I did in Blender, also using a low-resolution texture projection shadow:



Notice how the light creeps underneath the wall (yes, the wall is attached to the ground). This can be reduced using a 'bias' factor, however this of course will drain more resources from the system. The simple fact is that this sort of effect will always be present to some extent when using texture-projection shadowing.

Now take another look at the first screenshot. The planes that are further away have a worse gap in the shadow and lower shadow map resolution than the plane in the foreground. This indicates that the team are well aware of the issue and have done everything they can to minimize it.

[Meusli]

Originally Posted by B25Mitch

Now take another look at the first screenshot. The planes that are further away have a worse gap in the shadow and lower shadow map resolution than the plane in the foreground. This indicates that the team are well aware of the issue and have done everything they can to minimize it.

Hurray, somebody who knows something. Thanks for the explanation as even I understand that.

[McHilt]

Originally Posted by B25Mitch

Hi everyone. I see there's been some discussion about the gap in shadows, close to the base of the object casting them.



.............

Now take another look at the first screenshot. The planes that are further away have a worse gap in the shadow and lower shadow map resolution than the plane in the foreground. This indicates that the team are well aware of the issue and have done everything they can to minimize it.

That explains it very clearly! thx a lot for your effort!
@BadAim: you're right about the stabiliser... hence the position indicator: - 0 + I could've known....

[Skarphol]

Originally Posted by B25Mitch

Hi everyone. I see there's been some discussion about the gap in shadows, close to the base of the object casting them.



This is an inherent limitation of texture-projection type shadowing. Here is a quick example I did in Blender, also using a low-resolution texture projection shadow:



Notice how the light creeps underneath the wall (yes, the wall is attached to the ground). This can be reduced using a 'bias' factor, however this of course will drain more resources from the system. The simple fact is that this sort of effect will always be present to some extent when using texture-projection shadowing.

Now take another look at the first screenshot. The planes that are further away have a worse gap in the shadow and lower shadow map resolution than the plane in the foreground. This indicates that the team are well aware of the issue and have done everything they can to minimize it.

Thanks for your explenation Mitch! I found that gap in the shadow peculiar. As this gap has not been seen on other pictures, I guess the problem occures when the light hit the joining of those to surfaces at very special angles.

Skarphol

[major_setback]

Originally Posted by B25Mitch

Hi everyone. I see there's been some discussion about the gap in shadows, close to the base of the object casting them.



This is an inherent limitation of texture-projection type shadowing. Here is a quick example I did in Blender, also using a low-resolution texture projection shadow:



Notice how the light creeps underneath the wall (yes, the wall is attached to the ground). This can be reduced using a 'bias' factor, however this of course will drain more resources from the system. The simple fact is that this sort of effect will always be present to some extent when using texture-projection shadowing.

Now take another look at the first screenshot. The planes that are further away have a worse gap in the shadow and lower shadow map resolution than the plane in the foreground. This indicates that the team are well aware of the issue and have done everything they can to minimize it.

Originally Posted by Skarphol

Thanks for your explenation Mitch! I found that gap in the shadow peculiar. As this gap has not been seen on other pictures, I guess the problem occures when the light hit the joining of those to surfaces at very special angles.

Skarphol

It has shown up in quite a few of the earlier pictures. I've noticed it quite a lot. It shows where the aerial mast joins the fuselage (look at the big/close 109 screenshot in this weeks update), and on exhaust covers for example. You can see it on an opened spitfire door, and under the Hurricane tail too.
It looks like they (understandably) try to avoid taking screenshots from certain angles because of it.



Aerial and 'floating' engine intakes:



Exhaust cover and tail:

[major_setback]

I'd say that the pilot here is the right size. I can't imagine he could be much bigger:



It looks an improvement on this:

[Sutts]

It does look better but could just be the angle of the shot. He sure looks like he's just had a broom shoved up his a$$ in the second shot.

I'd love to see some oxy masks too. Goggles down without a mask was not a common sight from all the evidence I've seen.

[philip.ed]

Originally Posted by Sutts

It does look better but could just be the angle of the shot. He sure looks like he's just had a broom shoved up his a$$ in the second shot.

I'd love to see some oxy masks too. Goggles down without a mask was not a common sight from all the evidence I've seen.

+1
also, few wore these mark IV goggles

[Biggs]

many pilots flew without goggles because they weren't the greatest quality back then and would distort the pilots vision. this effect was compounded when adding to the regular amount of distortion that the cockpit glass already created.

they figured it was best to have the least amount of material between the pilots eye and the sky.

I think Bob Doe wrote/said something to that effect at one point.

[Blakduk]

Regarding the pictures of the stabiliser shadows on the 109's- they were not fixed completely to the tail section. They could be tilted to trim the aircraft- as i understand it they were not adjustable in-flight, and could only be set by groundcrew. I think the shadow gap is overdone but it is not apparent on the models of the other planes (such as the Spits and Hurricanes).
The problem of floating elements on smaller parts such as the exhaust covers and aerial masts is definitely there, as illustrated by the excellent examples posted by Major_setback, but the 109 stabilisers shadow gaps is partly due to the design of the real aircraft and therefore the true-to-life computer model.