[phpBB Debug] PHP Warning: in file [ROOT]/includes/functions.php on line 3781: ob_start(): output handler 'ob_gzhandler' conflicts with 'zlib output compression'
Oolite Bulletins • Leesti is Moon 8k - Page 2
Page 2 of 2

Re: Leesti is Moon 8k

Posted: Tue Jan 05, 2021 6:21 am
by another_commander
stranger wrote: Mon Jan 04, 2021 11:57 pm Regarding copyright. Almost all work, and all most critical part of work, is yours, and I feel I have no right to declare myself as author. Co-authorship maybe, another_commander & stranger?
Feel free to author it as you prefer. I have no particular preference and am happy either way. :-)

Re: Leesti is Moon 8k

Posted: Tue Jan 05, 2021 6:28 am
by stranger
another_commander wrote: Tue Jan 05, 2021 6:21 am
stranger wrote: Mon Jan 04, 2021 11:57 pm Regarding copyright. Almost all work, and all most critical part of work, is yours, and I feel I have no right to declare myself as author. Co-authorship maybe, another_commander & stranger?
Feel free to author it as you prefer. I have no particular preference and am happy either way. :-)
All right, sir! :D

Re: Leesti is Moon 8k

Posted: Tue Jan 05, 2021 7:41 am
by another_commander
Just a quick note that I forgot to mention previously: the texture combo I used is still not perfect. There seems to be a very slight offset at certain areas between the normal and the diffuse map, probably because I could not get a straight conversion from the 8192x4096 NASA bump map to a normal map of the same size (the normal map was coming out always a few pixels smaller, like 8144x4075 for some reason and I had to resize it to 8k to fit the diffuse dimensions). I think it's only a problem if you are trying to pay very close attention though and overall I believe the textures are still usable.

Re: Leesti is Moon 8k

Posted: Fri Jan 08, 2021 4:36 am
by stranger
@ another_commander
New diffuse map is fine (needs some light curve rebalance to adopt to Oolite lighting - already done).
New normal map is not so good. Too soft shadows, complete loss of shadows from small craters, ridges and gullies - it ruins impression of celestial body lacking atmosphere. So I decide to use original normal map. There are noticeably polar distortions in this normal map, but I hope most Moon surface will look fine with new diffuse map. Both maps are centered using 0°latitude as reference point, so seems this is no noticeable issue with displaced shadows. Maybe I'll try later to combine van Vliet's normal map (fine details!) with NASA CGI Moon Kit map to fix issue with polar regions.
Still need some time to assemble and test updated OXP in both versions.
Thank you once more for lesson!

Re: Leesti is Moon 8k

Posted: Fri Jan 08, 2021 6:31 am
by another_commander
The normal map was made by converting NASA's displacement map using this online tool: https://cpetry.github.io/NormalMap-Online/

If you want to give it a try, the parameter that defines the intensity of the shadows is Strength. For the moon displacement source file, the higher this is set, the softer the shadows in the resulting normal map (which is a bit of a reverse thing, but whatever). I used a strength of 10 (maximum), which is why shadows are so soft. You may want to try experimenting with that if you want more intense surface details.

Re: Leesti is Moon 8k

Posted: Fri Jan 08, 2021 6:47 am
by stranger
@another_commander
Thank you for info!
There are no 8192x4096 px resolution displacement map on NASA CGI Moon Kit page. Is it right to rescale 23040x11520 px file first and than convert it to normal map as second step?
And what original file will be more suitable - float or uint? Seems that floating format will be more accurate, right?

Re: Leesti is Moon 8k

Posted: Fri Jan 08, 2021 6:53 am
by another_commander
stranger wrote: Fri Jan 08, 2021 6:47 am @another_commander
There are no 8192x4096 px resolution displacement map on NASA CGI Moon Kit page. Is it right to rescale 23040x11520 px file first and than convert it to normal map as second step?
Yes, this is exactly what I did too.
And what original file will be more suitable - float or uint?
Not entirely sure on that, but I used uint.

Re: Leesti is Moon 8k

Posted: Fri Jan 15, 2021 1:45 am
by stranger
Leesti is Moon 0.3.0 and Leesti is Moon LSM 0.3.0 uploaded.
Original van Vliet's diffuse map in these updates was replaced to diffuse map from CGI Moon Kit by NASA's Scientific Visualization Studio. This diffuse map has only color information, not shadows.
Polar distortions on van Vliet's normal map were patched using normal map, generated from CGI Moon Kit displacement map. Comparing with van Vliet's normal map this normal map has too soft shadows with loss of small terrain details, so I used it only to correct polar distortions in zone approx 90°...80° latitude.
There are still some polar distortions remaining on normal map, but most of Moon surface now seems pretty cool.

Chang'e 5 landing site is added to Leesti is Moon LSM. Seems Chinese missing Mons Rümker declared as landing site. Nice result for first attempt in any case! Respect, China!