NEW ADDON : Earth's Craters

[Cham]

And I don't see any problem with volcanoes on islands either (on the contrary, the placement appears to be VERY precise here) :

volcanoes.jpg

Of course, I also checked the borders. They are fine too.

Some other islands example, with very precise placements (on Jestr's 64K texture, at least) :

Hawai.jpg

islands1.jpg

[Cham]

And three other examples (sorry, I'm yet unable to find any mismatch) :

islands2.jpg

islands3.jpg

islands5.jpg

[ElChristou]

So that means the BMNG is wrong and also the data for borders from Grant. We need to solve this. Any idea how to proceed?

[Cham]

So that means the BMNG is wrong and also the data for borders from Grant. We need to solve this. Any idea how to proceed?

BMNG ? What is this ?

The borders from Grant are perfectly fine with Jestr's texture. I don't see any problem on my side yet.

[Cham]

Here's Chypre. How can this be more precise ?

chypre.jpg

I suspect you have a tiling problem with your VT.

[Cham]

I found a small offset on some river, in South America :

offset.jpg

In some other cases, there are no offset at all.

So I think that the 64K textures are reaching the spheroid limit (Earth isn't a perfect oblate sphere). This is a good lesson. We can't trust Celestia at this scale.

[ElChristou]

Blue Marble Next Generation. It's the Nasa repository of raw data set for color, night, elevation, topo map etc... You need those huge raw data to process them throught Fridger's tools to build a Highres set.

Now, concerning the borders and Jestr's map, there is something odd. From my side (and I've post some images) the borders match perfectly the 64k VT but not the volcanoes or some craters. Now you say all is fine with Jestr's map so perhaps that's mean I do have a problem with LABELS just like this elevation with contries...

Someone using a BMNG map could confirm all is OK with Cham's volcanoes for example?

[Cham]

ElChristou,

read my previous post (cross posting...).

We can't trust Celestia at this 64K scale. We're reaching the oblate spheroid limit.

[Cham]

I just noticed that most small borders offsets are located in the south hemisphere. In the north hemisphere, I didn't found any offset yet.

This is consistent with the fact that Earth is shaped more like a pear than an orange, say. The spheroid limit is stronger on one hemisphere than on the other.

Most of the time, I'm seeing the borders offset (in the south hemisphere) when I'm reaching the texture's and borders accuracy limits.

And I suspect that the tiling on a sphere is also deforming a bit the textures.

[ElChristou]

Can be but I'm sure there is aproblem with labels from my side. For example I just saw that the "Chilie" was in the ocean! Concerning rivers, don't forget they do change their shape with time.

[BobHegwood]

Christophe?

Methinks there is something wrong with your installation of the BMNG textures. I just checked a few of the
locations you mention above, and they all look fine on my machine. The Deep Bay area, for example, is perfectly
centered using Dr. Schrempp's finished BMNG 64k VT textures.

Just FYI...

Thanks, Bob

[Cham]

Bob,

please, show us some picts.

[ElChristou]

Bob,

please, show us some picts.

No need, I'm sure it's a problem on my side. I will do a fresh build of Celestia and test on it.

[Cham]

I'm reaching Jestr's 64K texture's limit here. Here are three views in the northern hemisphere (again, no bumps, no spec...).

A river in Laos :

laos.jpg

The orange line at the right is to indicate some texture tiling deformations on a sphere...

A border in the north of Germany (see the precision of the two ends of the border) :

germany.jpg

A mountains chain, somewhere in Eurasia :

mountains.jpg

The small offset you see in the case of the Elbrus volcano is caused by the elevation.

[BobHegwood]

Click the images for larger views.

[Cham]

Nice Bob ! Thanks !

Here's another example of the precision of the volcanoes locations, with URL (show us your view !) :

volcanoes.jpg

URL view :
cel://Follow/Sol:Earth/2008-08-05T04:20 ... 6276&ver=2

[Cham]

And another river in Laos, with URL (reaching the texture resolution limit) :

river.jpg

URL view :
cel://Follow/Sol:Earth/2008-08-05T04:20 ... 6276&ver=2

Please, show us you view.

[Cham]

ElChristou,

in the case of the large Vredefort crater in Africa (the one you were sceptical at), here are two web sites describing it. The radius used in my data is perfectly consistent with the info given on these web pages (about 250-300 km in diameter, possibly even larger, so the radius of 150 km used in my addon) :

http://en.wikipedia.org/wiki/Vredefort_crater
http://www.hartrao.ac.za/other/vredefort/vredefort.html

The smaller circular structure you see on the texture is simply the rebound structure at the center of the crater !

[BobHegwood]

(show us your view !)

Sorry, couldn't get the CEL:URL to work no matter what I did...

At any rate, this is close.

[ElChristou]

Incredible!