That’s what I thought. Once I figure out how to combine untrimmed with other internal curves, this should become possible.
6 Likes
Great! 
People who do modeling of architecture and jewelry will be especially happy with this.
2 Likes
oh if we add this kind of wishes to the wish-list…
i would love to see inner tangent (g1) constraint to a surface edge. so the white surface edge should not only offer G0 but also G1…
green - given edge-surfaces (G0,G1,G2)
white - surface with inner G1 constraint
blueish - _fillSrf result
@menno please reserve some energy and time to make a great interface to this powerful command/feature.
looking forward to see the next evoltion - kind regards - tom
2 Likes
We are working on a GUI for FillSrf, but it is not yet ready for prime time.
3 Likes
This is also a good use case scenario. Reminds me of Rhino’s poor “Patch” command, except that it is unable to match with G1 to an inner target surface edge.
Meanwhile, xNURBS 7 has some new features reminiscent to what is discussed in this topic:
4 Likes
take your time - and hopefully surprise us.
I would love to see a very similar UI for
fillSrf
matchSrf
patch
(global) edgeContinuity
my 2 cents:
i like the idea of having a interactive list where elements can be added / removed and their parameters (G0,G1,G2…Tolerance etc) can be controlled. this list should be interactive - if an item is selected in the list, the viewport should highlight as well…
kind regards - looking forward to see the next steps -tom
@Cadworx I miss you in this topic
5 Likes
I love this too!
And it can be extended to Srf Creation tools as well so you can have a preview of the result before the “final surprise”!
Computational power today isn’t an issue anymore for this kind of operations so PLS more interactive tools!
6 Likes
That xnurbs tool looks like automagic i need to try it someday
They slightly cheat in their videos by showing the enhancements in a misreading way. For example, you will notice that in their last video the “bad input geometry” consists two fillets split by a straight plane. This is how xNURBS is able to achieve building a single-span untrimmed 4-sided surface with a minimum amount of control points, despite that they claim that the input edge curves consist hundreds of points. However, if the fillets were split by a curved surface, the resulting xNURBS surface would have multiple control points (or will be no longer untrimmed). But they purposely didn’t show that, in order to present the xNURBS tool as superior to the native tools in Rhino.
They used similar carefully setup tricks to present their older enhancements in the best possible way, too. This is a bit misleading, because the user gets the false impression that xNURBS is some magical tool that always delivers perfect results, no matter how bad the input geometry is. When the users fail to achieve clean results with xNURBS, the developer blames them that “they don’t use the tool properly” or that “they don’t know what are they doing”, further adding insults to their capability to think as human beings. Happened multiple times in several topics here, especially the one devoted for xNURBS. It’s a nice tool with handy features, but publicly putting the blame to the user when the tool fails is a bad habit.
2 Likes
Lol probably laughing all the way to the bank to thanks for letting me know
xNurbs isn’t that bad. Haven’t tried the QuadSrf feature.
What do you think about this setting? Good enough by your standards?
Needs to be checked with a dynamic zebra from various angles to get a better idea of the quality. Note that achieving tangency at the border does not guarantee a quality surface if the overall flow is wavy. Some amount of unwanted waviness is detected by the blue shading in your first image.
What are your thoughts on the Rhino CurvatureGraph that scan through a surface. Rhino now only gives edge CurvatureGraphs unless users generate separate IsoCurves on a surface and run CurvatureGraphs.
Ideally, do you harness CurvatureGraph or CurvatureAnalysis, such as Gaussian and mean? - You don’t need to answer.
Looks OK for average users like myself. 
The main advantage of xNurbs is that it facilitates Zebra flow, but users must still manually rebuild and adjust the surface.
1 Like
xNURBS SQUARE: TEST
You may or may not know this (probably you just were looking for a quick way to demonstrate xNurbs), but to me you got the hierarchy of surfaces wrong.
It makes no sense to construct the surface with g2 continuity from the edge of the narrow blend surface.
You should first construct the surface (as cleanly and lightly as possible) and then derive the narrow blend in g2 continuity from it.
Again, probably you did this just for demonstration purposes, but I think the importance of good patch layout and hierarchy is so often overlooked.
6 Likes
yes, it was for demonstration purposes, I was just trying to show the options and how the new xnurbs command works, and it would be nice to have it in rhino
1 Like
Yes of course, that’s what I thought.
A thing I forgot to show in the video is the ability to influence the quadsurface of xnurbs with internal points.
2 Likes
I totally missed that!
Thanks for pointing this out.