Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Dual Contours #26

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

[WIP] Dual Contours #26

wants to merge 5 commits into from

Conversation

sjkelly
Copy link
Member

@sjkelly sjkelly commented Jun 16, 2019

No description provided.

@sjkelly
Copy link
Member Author

sjkelly commented Jun 26, 2019

@rdeits This adds quite a few dependencies. AFAIK there are not many dependents for Meshing (only Drake Visualizer), so it shouldn"t be a big issue.

This lets us get nice sharp corners on meshes so we can 3D print robots :)

@sjkelly
Copy link
Member Author

sjkelly commented Jul 21, 2019

I still have no idea why 1.0 and 1.1 give different results.

@rdeits
Copy link
Contributor

rdeits commented Jul 23, 2019

Hey @sjkelly sorry I dropped the ball on this. Would a code review still be helpful?

Also, I wonder if you"re seeing a difference between the results on 1.0 and 1.1 due to some additional SIMD optimizations. We had a similar issue in EnhancedGJK.jl where the dot product of vectors from StaticArrays could give infinitesimally different results on different Julia versions because SIMD was used in some versions and not others. As long as the difference in the resulting mesh is small (presumably it should be), then it should be OK.

@sjkelly
Copy link
Member Author

sjkelly commented Jul 24, 2019

Also, I wonder if you"re seeing a difference between the results on 1.0 and 1.1 due to some additional SIMD optimizations. We had a similar issue in EnhancedGJK.jl where the dot product of vectors from StaticArrays could give infinitesimally different results on different Julia versions because SIMD was used in some versions and not others. As long as the difference in the resulting mesh is small (presumably it should be), then it should be OK.

I appreciate you pointing me in the right direction. There are several dot products here. I need to setup a 1.0 environment so I can get a full diff.

I am waiting to merge this until I have setup a system for differentiable geometry in my solid modelling kernel to get some additional testing. The algorithm is also very sub-optimal since it is a port from a python example. I will ping you after I do more cleanup.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

자동등록방지를 위해 보안절차를 거치고 있습니다.

Please prove that you are human.

2 participants