Earth Curvature

[Schmeisser1942]

Hi

i reported this during the early beta but it seems that it hasnt get any attentions since then.
The earth curvature seems off quite a bit, it seems that it is too high.
Ships with in real close vicinity are already almost hidden halfway up from the earth curvature.

https://steamcommunity.com/sharedfiles/filedetails/?id=2588407116

[DECAFBAD]

You seem to be correct.

The UZO is about 11m 5m above the waterline. Using the Earth Curvature Calculator[www.omnicalculator.com], at that eye level the horizon is about 12km 8km away. Anything closer than that should be fully visible.

That C3 looks like it's 8km away. If you set the eyesight level to 0m, it tells us that the obscured part is 5m above the waterline, which seems to be what we are seeing.

Obviously the water in the game isn't curved, that would be a ♥♥♥♥♥ for the CPU to calculate, instead the ships seem to be sunk artificially, which gives you the same impression. I think the devs forgot to add the eyesight height to the equation.

[Dewgle]

How far away were the ships in that screenshot for reference, OP?

[wolf310ii]

Originally posted by DECAFBAD:

The UZO is about 11m above the waterline

That would be more than the high from keel to the top of the conning tower.
The conning tower had a high from the deck of 3.42m, so the UZO had a high above waterline of around 5-6m

[Dewgle]

Originally posted by wolf310ii:

The conning tower had a high from the deck of 3.42m, so the UZO had a high above waterline of around 5-6m

Based on that information then, the hulls of ships would start to dip below the horizon from the u-boat crews perspective starting at a distance of around 8km.

You can also use the earth curvature calculator in reverse to determine at what distance you should no longer be able to see the target vessel. Just take the mast height and add 5 meters of observation height from the u-boat and plug that number into the ECC. The resulting answer should be the distance at which the mast tips would no longer be visible.

[DECAFBAD]

Originally posted by wolf310ii:

Originally posted by DECAFBAD:

The UZO is about 11m above the waterline

That would be more than the high from keel to the top of the conning tower.
The conning tower had a high from the deck of 3.42m, so the UZO had a high above waterline of around 5-6m

Yeah the numbers on that schematic aren't in meters (wtf is that?). The UZO is barely above the top of the conning tower, though. 5m sounds more realistic.

[Schmeisser1942]

I was around 3nm away.
I got another screenshots, 2nm away:

https://steamcommunity.com/sharedfiles/filedetails/?id=2589172432

And the ship is already sunk in with half of the hull:
https://steamcommunity.com/sharedfiles/filedetails/?id=2589172339

It is not that extreme in Real Life IMHO
https://i.imgur.com/FUoafyb.png

[DECAFBAD]

So turning off the curvature option is the more realistic setting at the moment

This post was made by #flaterathgang

[Dewgle]

Originally posted by Schmeisser1942:

And the ship is already sunk in with half of the hull:
https://steamcommunity.com/sharedfiles/filedetails/?id=2589172339

It is not that extreme in Real Life IMHO
https://i.imgur.com/FUoafyb.png

The second screenshot seems correct for that distance. Empire class ships already sit pretty low in the water.

[nihilcat]

Nice to see you back Schmeisser1942

The elevation change is computed from this formula. I double checked it now, but it seems to be legit:

vertical_elevation_change = (cos(distance_from_camera / earth_radius * PI/2) - 1) * earth_radius

Where earth radius equals: 6371000 meters.

I may be missing something here, but everything seems fine overall.

Originally posted by DECAFBAD:

Obviously the water in the game isn't curved, that would be a ♥♥♥♥♥ for the CPU to calculate, instead the ships seem to be sunk artificially, which gives you the same impression. I think the devs forgot to add the eyesight height to the equation.

This is disinformation - the water is curved by the shader, as well as lands and everything else. That's the whole point of this setting. I mean, it wouldn't do anything without it.

[nihilcat]

There you can check my old sheet with some pre-computed distances:

https://1drv.ms/x/s!Atpx7H7m5qiwm4FzQ7a0rybSfIbJ8Q?e=De8deZ

The first column is a horizontal distance, while the second column is the resultant vertical elevation difference using the formula above.

[Urfisch]

Yes, for 6m view height (standing watch) it's almost 9km (4,5nm) view distance. I also calculated that out of interest when it comes to spawning distance for shore areas.

Maybe its just some kind of engine specific effect? The formula is correct, but the actual rendering effect makes it look different for some ships? Ruby also tested this curvature effect when we introduced the tdc mod and found it to be correct, afair.

[DECAFBAD]

The elevation changes are correct, it's just that the sea elevation has to change along with the object seen by the viewer.

Originally posted by Schmeisser1942:

I was around 3nm away.
I got another screenshots, 2nm away:

https://steamcommunity.com/sharedfiles/filedetails/?id=2589172432

And the ship is already sunk in with half of the hull:
https://steamcommunity.com/sharedfiles/filedetails/?id=2589172339

It is not that extreme in Real Life IMHO
https://i.imgur.com/FUoafyb.png

Here's a closer picture of the Global Success I[www1.picturepush.com]

The aft part of the tug, which is a few meters out of the water, is unobscured and visible 4.9nm away. Do we know where the photographer was, when the distant picture was taken?

[DECAFBAD]

So I did a bit of testing.

C3 cargo ships every 2km, furthest is 10km
https://steamcommunity.com/sharedfiles/filedetails/?id=2590915154

View from UZO, the furthest one (10km away) is already hull down
https://steamcommunity.com/sharedfiles/filedetails/?id=2590915303

View from the attack periscope fully extended, the hull of the furthest freighter is still barely visible at 10km
https://steamcommunity.com/sharedfiles/filedetails/?id=2590915458

The Earth in UBOAT appears to be a bit small, doesn't it?

[Dewgle]

Based on the earth curvature calculator you provided and the 5 meter uzo height that Wolf provided, revealed that 8km is when the curvature of the earth would begin to obscure targets.

The furthest target in your experiment at 10km away is 2km beyond the horizon and roughly 6-8 meters of her hull is being obscured by the horizon.

This seems correct to me.

[DECAFBAD]

Originally posted by McDewgle:

Based on the earth curvature calculator you provided and the 5 meter uzo height that Wolf provided, revealed that 8km is when the curvature of the earth would begin to obscure targets.

The furthest target in your experiment at 10km away is 2km beyond the horizon and roughly 6-8 meters of her hull is being obscured by the horizon.

This seems correct to me.

Should be only 32cm down
https://dizzib.github.io/earth/curve-calc/?d0=10&h0=5&unit=metric

8km away, it's already down to the anchor