Celestia Forums

I created a specular map from SWBD files that can be used instead of the Watermask files at the BMNG server. It still needs data north of 60 degrees latitude and south of 60 degrees. I'm not sure how this will be done yet. The details on how it was created are at the Celestial Matters Forum in the thread http://forum.celestialmatters.org/viewtopic.php?t=114. Here are some screenshots I took.








As soon as I can I'll upload the binary file to my site. It is currently DOWN. or somewhere else?
cartrite

Okay, It's uploaded. This file is meant to be used with the F-tex tool tx2rgba.
I got over 300kbs download. The site goes down a lot so if there is no connection try again later.
cartrite

cartrite wrote:It still needs data north of 60 degrees latitude and south of 60 degrees.

By this you mean that north and south of those lines, there will be no specularity at all? Is this different from the watermask on the BMNG server?

By this you mean that north and south of those lines, there will be no specularity at all? Is this different from the watermask on the BMNG server?

Within the landmasses north of 60 and south of 60 there is no specular data except for coastlines. There are oceans and coastlines from 60n to 90n and 60s to 90s. There are previews at my site which show what the binary file looks like.
cartrite

cartrite wrote:Within the landmasses north of 60 and south of 60 there is no specular data except for coastlines. There are oceans and coastlines from 60n to 90n and 60s to 90s.

Ah. I understand now. I'll have to take a look and try them out when I get a chance.

Pretty interesting!
But I don??t have the SLIGHTEST idea what SWBD stands for!Or even how to use it...

Another stupid question from the Brain-Dead...

Are specular maps used with dry land masses? I know that they are
used to manipulate reflections on water and possibly ice, but do dry
land masses require some control of specularity also?

Thanks, Bob

BobHegwood wrote:Another stupid question from the Brain-Dead...

Are specular maps used with dry land masses? I know that they are
used to manipulate reflections on water and possibly ice, but do dry
land masses require some control of specularity also?

Thanks, Bob

Yes, one uses a black mask for the land masses to make sure that there the specular reflection is NIL ! White means total reflection, in contrast.

Actually my specular masks typically use just a little rest light (0.12), which enhances the details from the normalmap rendering (without making the surface look like plastic)...

I just made such a specmap using cartrite's new, much improved 84k swbd-watermap, added 0.12 of rest light, OR-ed it with the ocean boundaries determined from elevation=0 in the srtm-elevation map, added it as alpha channel to the BMNG basemap, cut the compound into VT tiles ....and --after 20 minutes-- I am now playing happily with the result!

All quickly done with my F-TexTools.

Bye Fridger

PS: for people wanting to try this as well: cartrite's 84k swbd-watermap is INVERTED relative to the official BMNG watermap! So you got to recompile my 'specmap' tool in the F-TexTools package, with one trivial modification



Skipping that '~' has the effect of not inverting the input watermap (cartrite did it already!)

PS: for people wanting to try this as well: cartrite's 84k swbd-watermap is INVERTED relative to the official BMNG watermap! So you got to recompile my 'specmap' tool in the F-TexTools package, with one trivial modification

Hi Fridger,
I'm going to have to try running one the way you did. With specmap tool. I started by running tx2rgba. I'm not sure what you mean by INVERTED though. I didn't have to do anything special to run tx2rgba. ????
I used the program png2bin for the final conversion to binary. The png map "looked" fine before I ran png2bin ??
cartrite

danielj,
SWBD = SRTM Water Body Data
Before you ask, SRTM = Shuttle Radar Topography Mission
It is a specular map. Water is white. Land is Black.
The specular map is making the river systems and lakes or oceans "stand out" or "shine" in the images I posted above.
cartrite

danielj wrote:Pretty interesting!
But I don??t have the SLIGHTEST idea what SWBD stands for!Or even how to use it...

Daniel,

if you try looking swbd up in Google you might come up with

swbd = switchboard

For earth there are various sources for elevation data
SRTM3, GTOPO30/SRTM30 and ETOPO2. SRTM3 are the best. That's why you see always the acronym srtm when we speak of elevations...and normalmaps.

For Earth, water bodies (rivers, lakes, ocean boundaries,..) are supplied by SWBD, GSHHS and VMAP0 data sets. So,

WBD = WaterBodyData, aha!

Each of them comes in a different data format. Like all new SRTM data, SWBD come in 1?°x1?° tiles.

Therefore they have first to be assembled into one big file. However SWBD involves a special so-called vector format, i.e. rivers etc are encoded as polygons, admitting a VERY high resolution. In order to be useful for Celestia maps, that vector format has first to be converted into the appropriate raster format. This is kind of tedious and time consuming.

cartrite just achieved this for the highest rsolution of 84k.

GREAT!

So now the new much better watermap can be used to make new super textures with the F-TexTools.

Much respective discussion you will find over in CM, here

http://forum.celestialmatters.org/viewt ... 14&start=0

and in a new thread by cartrite, here

http://forum.celestialmatters.org/viewt ... 48&start=0

We are currently discussing, how to make this conversion process professionally sound, such that henceforth the new watermap can replace the bad, old BMNG one at a scientific level. Reto Stoeckli from NASA is VERY interested in this issue as well...As he wrote to me recently, they did not attempt this tedious process of raster conversion of the SWBD data simple because of lack of manpower!

Another present issue is the fact that the hires SWBD data extend only from -60 to + 60 degrees in latitude. So a substitute is required for the more polar regions.
These questions have to be discussed and solved ...

Bye Fridger

cartrite wrote:

PS: for people wanting to try this as well: cartrite's 84k swbd-watermap is INVERTED relative to the official BMNG watermap! So you got to recompile my 'specmap' tool in the F-TexTools package, with one trivial modification

Hi Fridger,
I'm going to have to try running one the way you did. With specmap tool. I started by running tx2rgba. I'm not sure what you mean by INVERTED though. I didn't have to do anything special to run tx2rgba. ????
I used the program png2bin for the final conversion to binary. The png map "looked" fine before I ran png2bin ??
cartrite

Your map is fine if you use it just with tx2rgba! But I used your map in 'specmap' first to logically OR it with the ocean boundaries from srtm and inject some rest light (0.12) . In the distributed 'specmap' an inversion of the input watermap takes place ('~') which is needed if the BMNG watermap is used. It is NOT needed, if your new one is used.

So I first applied 'specmap' and THEN tx2rgba.

That's all.

Cheers,
Fridger

Most interesting... hopefully should get started on tex tools sometime this week!

Fridger, cartrite, perhaps it could be interesting to see the new spec-watermap applied to these regions,


lake Meade and western side of Gran Canyon, AZ


and northern part of Tunisia

both at local noon time, whose we spoke in the Celestial Matters Forum, here:
http://forum.celestialmatters.org/viewt ... c&start=75
In those images most of white spots should not to be there!
I would like to see same images both both by cartrite, with his 84k swbd-watermap map from SWBD using tx2rgba only, and from Fridger, using the same cartrite's 84k but with "added 0.12 of rest light, OR-ed it with the ocean boundaries determined from elevation=0 in the srtm-elevation map, added it as alpha channel to the BMNG basemap, cut the compound into VT tiles" using full F.Textools capabilities.
Can we see the new ones, please?
This will give a clear idea of the real improvement obtained this way, IMHO.
Thans a lot!
Bye

Andrea

Andrea,

Please have a look at CM, there I will display truely spectacular shots with the new VT's. In about 15 minutes the post will be completed in CM.

Here is what you get at level5 with the new specmap based on cartrite's work:

Tunesia:



Lake Meade:






Cheers,
Fridger

t00fri wrote:Andrea, Please have a look at CM, there I will display truely spectacular shots with the new VT's. In about 15 minutes the post will be completed in CM.
Cheers,
Fridger

Wow, wow and wow again!!!
Absolutely wonderful!
A big "WELL DONE" to cartrite and you, for all the time and experience you both used to obtain such a stuff for all of us.
VERY APPRECIATED INDEED!
Cannot wait to see the full post and images in Celestial Matters Forum, see you there!
Bye

Andrea

ANDREA,
Your Welcome.
Fridger beat me to it. I also took snapshots of the areas but couldn't logon to imageshack. His look better anyhow. I've yet to process the specmap with the specmap tool so my bmng doesn't have the rest light at (0.12) yet. Have to do that soon.
EDIT:The file is updated and is also available here. http://www.celestialmatters.org/users/c ... ermask.zip
Many Thanks to rthorvald.
SO......keep in mind. If you are intending to use the specmap tool from F-TexTools toolkit use the file Watermask. If your looking for a specmap for whatever reason, use the file called Landmask. The Landmask file is only available at my site.
cartrite

I updated my site to contain 2 files. The original file I uploaded is a ready to use specmap. I have renamed it to SWBD_Landmask.zip. An updated file called SWBD_Watermask.zip is inverted. That is the water is black and the land is white. Fridger's F-Tex tool, specmap, should work properly with the SWBD_Watermask file without modifying it and rebuilding. The SWBD_Watermask is the file you want to use if you intend to use that tool.
cartrite

We have truly entered the realm of cartography... super job!

I just published some pages at Celestial Matters. The Stephen page and the SWBD Watermask page contain more information about the methods used, previews, links, and information about how this file was created. http://www.celestiaproject.net/forum/viewtopic ... 5977#95977
cartrite