Celestia Forums

Hi All,
I know that there has been some agreement that the oceans are too dark, coastal waters include. But everyone seems to have left the coastal waters a little darker then the deep oceans after brightening them. Should the shallow coastal waters be lighter than the deep oceans? Darker? Or generally the same?

The deep oceans have no data. Just 020514. The coastal waters have variation but is it accurate and worth it to preserve this data?

These textures are a real task to deal with given their size and all. So I was thinking of writing a program that will perform the corrections needed without an image editor. I'm trying to come up with the formulas now.

cartrite

I understand that the BMNG philosophy was to present Earth in colors
that we would perceive IN THE ABSENCE of an atmosphere. In such case, the dark oceans make sense.

My attitude would correspondingly be to leave the base texture untouched and contemplate any modifications via an appropriate colored layer in the clouds texture.

Bye Fridger

In the past I did some expeirimenting with haze density and tinting the clouds.
I may have no choice but to leave it as is because it would seem that 2 Gb of memory is not enough to recalculate pixels with the gimp on the original size tiles 21600x21600. What concerned me was the following paragraph taken from the Limitations section. http://earthobservatory.nasa.gov/Newsroom/BlueMarble/

Deep oceans are not included in the source data; the creator of the Blue Marble uses a uniform blue color for deep ocean regions, and this value has not been completely blended with observations of shallow water in coastal areas. The lack of blending may, in some cases, make the transition between shallow coastal water and deep ocean appear unnatural.

My last attempt at BMNG I covered over the coastal waters but this time I would like to keep this data and I thought by lightening it to about the same level as the deep oceans that in would blend in better. But I thought that shallow water would look lighter than deep water due to reflections from the sandy bottom. Not darker.

cartrite

Fridger is right. This is exactly what I am doing to each of the 12 BMNG textures. I have a ocean layer, that lightens the costal areas and corects some of the details that were lost in the new textures. Of course this still isn't totaly acurate either. My layers don't take into account the seasonal variables of alea blooms and the like in the ocean waters. But I have enough data to worry about now applying these layers and coreting glitches in ht e main textures in the snow regions.

Don. Edwards

Don Edwards wrote:

Fridger is right. This is exactly what I am doing to each of the 12 BMNG textures. I have a ocean layer, that lightens the costal areas and corects some of the details that were lost in the new textures.

That is what I'm trying to create. An ocean layer for a spec map.

So you are leaving the oceans the original color and just lightening the coastal areas?

Anyhow I just had a close look at the watermasks available from the BMNG site and the inland waters are way off. I think I'm going to try and create a hand made map of inland waters for the spec map.

The program I was thinking of writing would flag any pixel value less than 020514 and add 000307 to it. This would use the BMNG 84k raw flat world file as input. I did basicly the same thing with the gimp on a subtile from A1 and it looked pretty good. It came out slightly lighter then the deep ocean but some areas were still slightly darker. And the deep water would stay 020514. Then I could lighten all or leave as is.

I may be wrong but isn't the whole coastal area just representing the height value of the sea floor and has nothing to do with the reflective properties of the water? This is my basic question. The reason I started this thread.

cartrite

Well in a sense yes. The BlueMarble team seems to have over compensated for the lighter coloring of the coastlins. Thats why they apear so much darker than the water in the midle of the oceans. Of course it should be the other way around.

Yes my ocean layers are changing the overall color of the oceans. It is taking the slightly purple hue out and and shifting it more to a blue/green color. Of course the oceans are so dark that you can't really see the difference until the texture is loaded and rendered in Celestia with all the proper effects.

I will probably make 2 versions of all 12 textures. 12 with the standard dark oceans and 12 with much lighter oceans for those with video cards with rendering limitations. Many users with older or limited video cards oftencomplain that the oceans are to dark.

I will try and do a couple comparison shots in Celestia for you so you can what my layers do.

Don. Edwards

Don. Edwards wrote:I will probably make 2 versions of all 12 textures. 12 with the standard dark oceans and 12 with much lighter oceans for those with video cards with rendering limitations. Many users with older or limited video cards oftencomplain that the oceans are to dark.

It is a very big addon !
Is it a 16K or 64K-128K ?

All will be at 16k. I just don't have the memory to work on anything bigger.

Don't know if this link has already been given, but I found the information on how colors and reflections appear to astronauts from space

http://eol.jsc.nasa.gov/newsletter/RedSeaReefs.htm

Hope can help.
Bye

Andrea

Thank You Andrea,

How did you find this?

This kinda shows me that some of the color correction tests I was discarding may have been in error. I was discarding colors that showed a redish hue.

That small island shows no darkening around it's oastline. Eliminating all the darker water along the coasts seems to be a correct representation of a visual view from space. I wrote a program that does this for about 95% of the coastlines.

It would be interesting to put all that whitcap action into the oceans. I'll have to try this somehow.

cartrite

cartrite wrote:Thank You Andrea, How did you find this? cartrite

Cartrite, I'm subscribed to NASA's Gateway to Astronaut Photography

http://eol.jsc.nasa.gov/distribution/

that sends me regularly by email links to new astronaut's images from ISS.
One of this is the link I submitted here.
Hope it helps, I find it very interesting. If you dig over there (but there are thousands of images!) you'll find perhaps some replies to your doubts.
Bye

Andrea

Thanks Again,

Just subscribed.

cartrite

cartrite wrote:Thanks Again, Just subscribed.
cartrite

You are welcome.
Bye

Andrea