Hillshading #923

Open
devemux86 opened this Issue Feb 12, 2017 · 4 comments

Comments

Projects
None yet
3 participants
@devemux86
Collaborator

devemux86 commented Feb 12, 2017

Hillshading from digital elevation model data like HGT format.

@devemux86

This comment has been minimized.

Show comment
Hide comment
@devemux86

devemux86 Feb 12, 2017

Collaborator

See PR #922 by @usrusr and discussion in forum.

Collaborator

devemux86 commented Feb 12, 2017

See PR #922 by @usrusr and discussion in forum.

@devemux86

This comment has been minimized.

Show comment
Hide comment
@devemux86

devemux86 Feb 26, 2017

Collaborator

On Desktop Samples the 1st argument can be used as dem folder.

Will leave the issue open for monitoring further hillshading improvements.

Collaborator

devemux86 commented Feb 26, 2017

On Desktop Samples the 1st argument can be used as dem folder.

Will leave the issue open for monitoring further hillshading improvements.

@demiantres

This comment has been minimized.

Show comment
Hide comment
@demiantres

demiantres Sep 15, 2017

The hillshading algorithm produces visible seams along HGT file borders:

mfbound

The hillshading algorithm produces visible seams along HGT file borders:

mfbound

@usrusr

This comment has been minimized.

Show comment
Hide comment
@usrusr

usrusr Sep 18, 2017

Contributor

Visible seams in rough areas are known (interpolation artifacts: the interpolator assumes that the neighboring pixel outside the texture is the same color as the closest available, the dark part at the northern hill is one of those), But the dark line in the lowlands must be something different.

Except if the screenshot is from a build based on the stale state of my morehills branch, then it is a known error in my attempts of fixing the interpolation issue (which is why the morehills has no PR).

Contributor

usrusr commented Sep 18, 2017

Visible seams in rough areas are known (interpolation artifacts: the interpolator assumes that the neighboring pixel outside the texture is the same color as the closest available, the dark part at the northern hill is one of those), But the dark line in the lowlands must be something different.

Except if the screenshot is from a build based on the stale state of my morehills branch, then it is a known error in my attempts of fixing the interpolation issue (which is why the morehills has no PR).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment