-
Notifications
You must be signed in to change notification settings - Fork 168
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
Remove unreliable signals very close to zero #4420
base: master
Are you sure you want to change the base?
Conversation
…umerical approximations
Once approved, this should be cherry-picked to maint/v4.1.0 as well. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I would very much prefer if we could update the comparison program instead.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not a specialist in MultiBody but for me it looks reasonable.
@HansOlsson the more I think of it the more I see you point. These signals are supposed to be near zero, and if we don't check they may (in principle) stray away. One could argue that they very likely can't do that without some of the non-zero components changing significantly, but at the end of the day this issue can be really fixed by just adding some sensible absolute tolerance to the CVS-compare tool. I'll keep this on hold for the time being. |
Pull request was converted to draft
I converted it to a draft PR to reflect the previous comment and avoid accidental merging. |
These signals only depend on numerical approximations and CSV compare cannot evaluate them reliably because it lacks information about their magnitude (nominal attribute).