How can I get a water world?

[cyphyr]

I've beeen playing about with planets and water. all I do is set the water sphere very slightly bigger than the terrain sphere.
Usually a value of about radius = "6.3783e+006" (red numeral is the one I've added to the world radius) seems to work. If things like shore line texture are then in the wrong place I'll raise or lower them proportionately rather than trying to get the water to a zero altitude. The biggest issue I keep running into is the the numeric feedback for anywhere other than the top of the world is pretty useless. Whats the point of knowing a mountain at the equator for example, that is clearly way above the water line is actually -1996m, everywhere on its surface, dont get it... :|

Richard

[Mr_Lamppost]

I've beeen playing about with planets and water. all I do is set the water sphere very slightly bigger than the terrain sphere.
Usually a value of about radius = "6.3783e+006" (red numeral is the one I've added to the world radius) seems to work. If things like shore line texture are then in the wrong place I'll raise or lower them proportionately rather than trying to get the water to a zero altitude. The biggest issue I keep running into is the the numeric feedback for anywhere other than the top of the world is pretty useless. Whats the point of knowing a mountain at the equator for example, that is clearly way above the water line is actually -1996m, everywhere on its surface, dont get it... :|

Richard

Too late to test that tonight but isn't your 3  outside the range of significant figures saved?

Sorry, but I cannot confirm your observation ... might be because I am using an alpha version not the beta. Somebody else has to try.

I am running Beta 1.10.23.1  which build ar you using?

I downloaded Mr. L's tgd test file and replaced all his fancy function with a with a simple power fractal terrain.  The normal displacement of the power fractal will give the water enough depth since displacement is both positive and negative.

I used the "Fancy functions", to deliberately generate a terrain that would be very sensitive to the actual scale of the water sphere as compared to the planet.  The change I was demonstrating, 1 metre, is less than 0.0001% of the radius of the planet.  If you are using standard power fractals to create your terrain it is far easier to move the terrain relative to the water than if the terrain is function based.  I could have lowered the terrain by 1 meter to obtain the effect I was after but that would have been fiddly compared to just being able to raise the water. 

[MortalSphere]

I used the "Fancy functions", to deliberately generate a terrain that would be very sensitive to the actual scale of the water sphere as compared to the planet.  The change I was demonstrating, 1 metre, is less than 0.0001% of the radius of the planet.  If you are using standard power fractals to create your terrain it is far easier to move the terrain relative to the water than if the terrain is function based.  I could have lowered the terrain by 1 meter to obtain the effect I was after but that would have been fiddly compared to just being able to raise the water. 

Ah ok.  So unless you want specific manufacture terrain features at a particular altitude relative to the water then you would use functions?

[Oshyan]

I believe the numerical accuracy has been increased in internal builds since the last beta. These fixes will be present in the final release.

- Oshyan

[Mr_Lamppost]

I believe the numerical accuracy has been increased in internal builds since the last beta. These fixes will be present in the final release.

- Oshyan

That's good to know. Although there does not, for this example at least., appear to be a problem with numerical accuracy.  It is simply that not enough significant figures are being written to the .tgd file. 

I used the “Fancy functions”, to deliberately generate a terrain that would be very sensitive to the actual scale of the water sphere as compared to the planet.  The change I was demonstrating, 1 metre, is less than 0.0001% of the radius of the planet.  If you are using standard power fractals to create your terrain it is far easier to move the terrain relative to the water than if the terrain is function based.  I could have lowered the terrain by 1 meter to obtain the effect I was after but that would have been fiddly compared to just being able to raise the water. 

Ah ok.  So unless you want specific manufacture terrain features at a particular altitude relative to the water then you would use functions?

The choice of functions or Power Fractals will depend on the type and shape of the terrain you need.  I used functions for the example because I wanted a terrain where large areas were at or very close to zero altitude and no areas below zero.  I just thought it was quicker to string a few function nodes together than to set and test power fractal parameters.   

[Oshyan]

I believe the numerical accuracy has been increased in internal builds since the last beta. These fixes will be present in the final release.

- Oshyan

That's good to know. Although there does not, for this example at least., appear to be a problem with numerical accuracy.  It is simply that not enough significant figures are being written to the .tgd file. 

Aren't these essentially the same thing in practice though? In any case I think it is this exact problem that is remedied with the changes.

- Oshyan

[Mr_Lamppost]

I believe the numerical accuracy has been increased in internal builds since the last beta. These fixes will be present in the final release.

- Oshyan

That's good to know. Although there does not, for this example at least., appear to be a problem with numerical accuracy.  It is simply that not enough significant figures are being written to the .tgd file. 

Aren't these essentially the same thing in practice though? In any case I think it is this exact problem that is remedied with the changes.

- Oshyan

Numerical accuracy and significant figures; are they the same thing?  Yes.

The problem is not with the way numbers are stored and manipulated during editing and rendering but with the degree of accuracy with which they are stored in the .tgd file.  It should also be noted that there is a similar reduction in the accuracy to which very large or small ar displayed in the editor.

As you say this problem has been fixed any further discussion should be reserved for English tutorial 09:15 Tuesday. Well that was when it was when I was at school.