Hello to all, and hope all is well. The UI improvements to v8 are impressive.
On importing curves from outside Rhino, exposing control points can present a real mess.
I’d like to know if there is a way to ‘Rebuild’ a curve with the absolute minimum of control points, user controlled. The result would be a curve of the suggested number of control points, with each points ‘gravity’ automatically adjusted as required.
As an example, I can import an open curve and on so doing, duplicate the curve by tracing it, adjusting the control points weight(s) as needed. Just by looking at it, I can guess the minimum points I’ll need. I’ve experienced many cases where the result so closely matches the original that Rhino cannot measure the delta between the rebuild and the original. Surfaces built from these curves are superior.
Does this type of functionality exist, or might it exist in an add-on I can find on Food For Rhino?
Thanks David. I have not used CrvDeviation. I’ll give that a go.
To an extent though, measuring is not the main point. Reducing the number of coordinate points to the absolute minimum is.
FitCrv does not help. In fact, it worsens the problem by adding points.
As it is, as a user since 1.1, I can import a coordinate file of a hydro/aero foil in .dat format, which consists of straight lines between coordinate points, and just by looking at it, understand immediately I can trace the plot in as little as 3 or 4 control points from an original consisting between 60-150 plot points with accuracy so fine no aero/hydro analysis can tell me any different.
The win is in the surfacing David. Aero/Hydro surfaces built from absolute minimalism translates to machine code that drives higher end NC equipment as if it were alive. Male/Female parts coming off the machine need only final polishing. Cut time is reduced as well.
I imagine such functionality as an extension of _FitCRV. An advanced dialog when/if activated might ask how many points do you want on the refit. Given the response, the function traces accordingly, adjusting point weight in the process. Final results though not explicitly, exactly matching the plot are so perfect the flying result exceeds the original prediction, this due to the excellent surfacing that results, which again results in super accuracy in machining.
I’d be happy to post an example if you are interested. I work without obviously, but I think this would be a spiff feature.
Your attention is appreciated. Puffins are way cool.
As far as I know, there’s no straightforward way to reconstruct curves effectively in Rhino. I’ve encountered this issue before and found a relevant discussion thread:
Actually, to a point I very respectfully disagree, though I see your point clearly. I think _FitCRV is straightforward and does effective work, but not being a developer, I imagine it going farther and doing higher quality work, under the control of experienced Rhino users. That said, I believe we are on the same page.
I am tracing primitive coordinate data generated from programmatic output from software that dates from the DOS days. That alone is ultimately an intuitive process. _FitCRV is on the path, but the destination includes taking the absolute minimalist approach, adjusting point weight along the way to absolute minimum. I personally think _FitCRV should ask a question; “I can name that tune in ‘x’ notes. Do you approve this preview?”
This topic is not new. I’ve mentioned it in times past, but now my surfacing work is being analyzed in physical and virtual wind tunnels driven by exotic computational environments on an international level. It turns out one of my advisors has interest as well, plus supposedly a connection within Rhino.
Ex-Microsoft; if our terms even remotely match, I would respectfully ask that this be classed as a DCR (Design Change Request) classed level C, but begging for level B. I get that this may well be considered a stretch goal, but I would be willing to bet there are sufficient surfacers out there willing to hold it up.
Please consider? Thank you for your support as always.
