A implementation for CAD #31
Replies: 1 comment
-
Thanks!
If you have any thoughts on how this experience could be improved, I would love to hear that. Were there any specific obstacles? Or would you imagine the API structured differently?
Just to be sure, success/failure is deterministic right? So for a specific problem with specific initial values it always either succeeds or fails. Likely not the problem, but I suggest to zero the In order to debug the problematic cases, can you install a log-compatible logger (e.g., env_logger) and set it to output debug level logs? It will print a lot of things, but it should give a better idea about what is going on under the hood. If it is possible, you could also paste the logs here so I can try to make sense out of them.
Yes, this migration is planned for the upcoming weeks in my job, so unless something unexpected happens, I believe that it will be released soon-ish. |
Beta Was this translation helpful? Give feedback.
-
Hello,
Following our discussion here, I've done a small crate from my project where all the rendering and ui are removed to concentrate on the constraint problem, you can find it here. Was a lot of work to bind my data structure to yours.
In the example, I've four points on which I can apply constraint like segment parallel, horizontal, vertical, point fixed, point fixed on x axis, point fixed on y axis...
If I've exactly 8 equations then it's a determined system and the solver find every time the solution.
But if I remove a constraint (undetermined system), sometimes it converges, sometime not.
I can arrange to have always determined system but it will be not easy I think.
What do you think about this?
And do you plan to use Fear soon?
Olivier
Beta Was this translation helpful? Give feedback.
All reactions