What will show up on an InterpCrv the original points clicked when drawing it?


having drawn my aerofoil sections with InterpCrv through original x (distance from datum) and y height above chord axis data from a table, I need to place lines between the profiles sitting in their positions on the tailplane through those points. Such points are to a standard half cosine spacing or similar established procedure, scaling the x distribution of them from one profile to another I see they all follow the same distribution, all except the last profile oddly enough so I have now redrawn that to the same number of x/y points. as per the rules of best surfacing and also aerofoil wing creation.

I see no way of getting those points to show though.

If I select edit point I get points but not where the x/y points are. It is tempting to draw lines through these instead but that would be incorrect. I need to draw lines between the profiles original draw points.

The only other thing I can think of is to place rhino points along the profile at the x/y dimension points on the curve, group the lot then reposition these in their tailplane locations and draw between the rhino points.


Hi Steve1

Try to choose Knots=Uniform in InterpCrv.
Once drawn curve use the InsertEditPoint command.

Assuming, which I cannot tell for sure, that the goal is to edit the curve by changing the original input point locations, the tricky way would be to place the points first (Points command) at the locations you like, then with the points selected and with History on, use CurveThroughPt > CurveType=Interpolated. Now, move the input points around to edit. But, I am not at all sure what you need the points for so that could be useless.


This may not be possible for you currently, but I always keep the points I use for drawing curves so they are available for future reference.

I had knots set to Chord (default option as I never have altered it)

I will experiment to see what the Knots options do.

Pascal, When drawing aerofoil sections , as illustrated, I am using InterpCrv Degree 3 Knots Chord PersistentClose=No
It was not for editing the curve.
It is required when skinning the wing from profile to profile that the correct method is to split each profiles curve at the points it was drawn through.
One then creates curves with InterpCrv or straight lines ( I am wishing to find out which is best or correct, having ‘lost’ the memory of that ) between those points, heading out to wingtip, then one sweeps2 the first profile segment from rib 1 to rib 2 then to 3 etc, repeat for next segment, and so on.
To sweep the entire profile unchopped up from rib1 to rib 2 or if its a straight line generator rib 1 to rib 12 is wrong. So I am told by my learned expert.

One reason why each profile should have the same number of CV (points) mine have 20.

It was to be able to split the profiles at these points having sited them on the wing and at their correct spacing along the wing spar that I needed to find again the points in the curve so as to cut curve at them and also draw between them rib 1 to 2 to 3 etc.

I find it so odd that the points one picks with mouse click cant be identified afterwards in the curve.

David…whilst awaiting for replies and needing to move on, I had to return to the drawn profiles , dims and datums, and use the point tool, lay down points for all the ribs at the tips of the dimensions in jpg above, group them with the curve, replace all profiles on tailplane with these, rotate them back into place and upright, then set about drawing the lines between ribs !

Knot = uniform hopefully will save that hassle.


Is this the crux of your question Steve? Honestly it’s hard to tell from your post. Are you asking why after using InterpCrv that the points are no longer on the curve itself?

My understanding is Steve inputs the points, probably on a construction plane, builds the curve using InterpCrv or CurveThroughPt > CurveType=Interpolated, and then moves and/or rotates the curve into the needed position. He would then like to have the points available on the relocated curve.

That sounds like one way to “skin a wing” but I would not call it the correct method.

Exactly so.


Are you asking why after using InterpCrv that the points are no longer on the curve itself?

When one draws a curve with InterpCrv, one goes click click click with mouse on places one wishes the curve to pass through. (in my case the dimension arrow tips) For me these are the points in a 2D world (drawn on a construction plane) obtained from paired dimensions in an aerofoil data table. The aerofoil section line (wing surface line) is then drawn this way.

I then take this curve and place it in the 3D world where it should go so as to create a wing.
I am saying that having moved my InterpCrv drawn curve I am in need of seeing these points again yet there seems no way of displaying them.

david That sounds like one way to “skin a wing” but I would not call it the correct method

it wasnt until I was told it was the CORRECT way that I did this. I have a VERY learned friend who instructed me so.

I would rather just take the entire profile and sweep2 it to the recipient profile, far less work.

To do it the ‘correct’ way, or so I am led to believe, and one doesnt argue with gurus, I need to locate the aerofoil table positions on the curve and then draw lines from them on rib 1 to same rib 2 and so on, split curve at those lines then skin wing in many strips ! then join it all together again. A recipe for disaster maybe but its the correct way.

if only I could find those points on the curve.


In that case, placing point objects and using CrvThroghPt, and then moving both curve and points will do what I think you need.


Guess my workaround is the actual solution then, I however used InterCrv to draw my way through them.

same result I can see …correct me if I am wrong.

I often start off tangent at leading edge of profile so InterpCrv allows me that luxury.

What is the significance of Knot uniform versus knot chord, any advantage additional to this method Pascal ?

Coming from Macromedia Freehand land (now in the distant past) where the points clicked were able to be shown again afterwards, as one needed to sometimes move them, Rhinos curve tools not displaying them afterwards always struck me as odd, if user felt they were important enough to draw through initially, user might want access to them again :smile:



From Help:


Determines how an interpolated curve is parameterized. When the spacing between the picked points is equal for the whole curve, all three parameterizations generate the same curve.

When you draw an interpolated curve, the points you pick are converted into knot values on the curve. The parameterization means how the intervals between knots are chosen.


The knot spacing is always 1 and is not based on the physical spacing of the points. Uniform parameterization can be used if the point spacing is roughly equal and it is desirable to draw several curves that have the same parameterization. This is possible only with uniform curves (unless you rebuild the curves). Uniform guarantees that every control point affects the surface exactly the same way, no matter how much the surface is edited.


The spacing between the picked points is used for the knot spacing. This makes curves that have widely varying point spacing behave better than uniform curves.


The square root of the spacing between picked points is used for the knot spacing.


My picked points for aerofoil have spacing which progressivly increases to start with (half cosine) so it would seem I need to stay with Chord as I have been using.
No extra advantage to your solution.

When you draw an interpolated curve, the points you pick are converted into knot values on the curve

Knots though are where one initially had picked the points when drawing curve, so here is the solution then …osnap Knot , that will find them ?


Just drawn InterpCrv through rhino placed points then osnap only knots finds every click I made on that curve when drawing it, so KNOTS = original click points when drawing InterpCrv.
if curve is away from the original drawing I can find those points on that curve with osnap Knots.

result :slight_smile:


Just keep in mind this will not be true for degree 3 curves - probably what you are using, but just in case.


ah… mine are deg 3, well it worked first time, fluke I guess.

back to plan B then.


I knew I should have kept my mouth shut - just carry on - use degree 3, that is perfectly fine.


1 Like

you mean it is ok after all.

oh bliss :smile:

I just go knot detecting along a deg3 curve like mine detecting. :slight_smile:


Right, so this cuts to a fundamental misunderstanding of how NURBS math work. Rhino is a NURBS modeler, and so it makes curves and surfaces using the underlying NURBS math. NURBS curves and surfaces are made up of control points. For a NURBS curve, the only time the curve is going to pass through the control point (provided your curve is not straight) is at the beginning and end. If your curve has any…curvature to it, it will NOT pass through the control points. So, when you are using InterpCrv with Rhino, what you are asking is “Hey Rhino, I want to make a NURBS curve that passes through some set points, and I don’t want to spend all day point editing it, if I show you the points where I want the curve to pass through, will you do your best to figure out where the control points need to be placed?” And to this, Rhino says “Sure! Here’s your curve, it passes through the points you showed me.” And, by definition of the math behind NURBS, those points will not be the same as the points you picked.


Well, I think the problem is that the EditPoints are not where the InterpCrv input points are placed, and that is counter intuitive…