Guarantee of affine status of Mat4::look_at_rh() and Mat4::look_at_lh() output
#310
Replies: 2 comments 1 reply
-
|
Yeah it looks like that assert needs an epsilon, it's doing an exact comparison with |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
I've pushed a fix for this to main. I'm not sure when I'll make a new release, want to get some other stuff in as well. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I've been getting some asserts when I turn on
debug-glam-assertfrom callingtransform_point3()on the output ofMat4::look_at_rh(), but upon inspection, row 3 typically falls in the range [[-0.0, -0.0, -0.0, 1.0],[0.0, 0.0, 0.0, 1.0000001]]. There's no means of supplying a perspective to theselook_at_*functions, so shouldn't the error introduced by writing[0.0, 0.0, 0.0, 1.0]to row 3 be negligible? I'm admittedly not the most knowledgeable on 4x4 matrices, so apologies if I'm making baseless assumptions.Beta Was this translation helpful? Give feedback.
All reactions