Jump to content

skate bowl


Recommended Posts

  • Vectorworks, Inc Employee

Its possible to use just the one profile curve with the last mode of Loft, but since the two rail curves aren't identical, the profile warps undesirably in your example.

Seems like this should be a simple enough feature request for "Allow multiple profile curves in Birail Sweep Mode" which I'll submit now. I'll take a look at how Rhino and some other modeling packages handle it as well.

Link to comment

I came close. But it's not perfect.

I made each one of the arcs entirely flat, and adjusted the connector NURBS so that they are tangent to the flats in z direction. xy stays same.

The resulting surface has some overhangs and gentle ripples. It does shell.

New layers in the vwx file. 3d 3 is my setup, 3d4 is the Bi-Rail Loft. Strangely, the loft only worked with one of the profiles - on the right in Top view. Click order seems to make different lofts, too. I think best is bottom curve, then top, then profile.

Also tried simple Loft with lots of ribs (Layer 3d2). That might offer more control. But lots of work and additional problems with ripples..

-B

ubbthreads.php?ubb=download&Number=14709&filename=Skate%20Top:Plan.png

ubbthreads.php?ubb=download&Number=14710&filename=SkateBowl.png

Edited by Benson Shaw
Link to comment

Benson, i figured you would be the one to figure it out. here is the thing. we need to have a way to constrain connection points. like this

"this is always tangent"

"this is always at a right angle to the bottom line"

"this is always at a right angle to the top line"

"this radius never has an under cut"

"there should be no kinks in this path"

etc

Link to comment

these constraints would stay with the surface so if you adjusted the bottom rail the radius would adjust..

the bottom line is,

NOTHING SHOULD FOLD OVER, WRINKLE OR HAVE AN UNDERCUT this should be the default. then if you wanted an edge then you would need to say "put edge here"

granted there are over verts in some designs but that would need to be controlled by a horizontal plan that says "anything above this line can cure back on itsself"

Link to comment

DM, your list is right on. Another problem with my approach is that I altered the design to accommodate the software. A Loft fail can indicate non doable geometry, but it still seems kind of backwards.

This skate park should be as easy as a single command for a new surface that is a fillet between the two curves. Everything tangent. If the command fails, there should be feedback, such as highlight problem areas along the curves.

But in Vectorworks, this obvious solution is not an option. sigh. To really model this, I think the P. Cipes method is called for - Model it like you build it. Establish the upper and lower curves, screed (loft) between them with arc shaped tools. Isn't that how they build skate bowls? But that's LOTS of work in vwx. Rib after rib, then multiple lofts and adding solids, then it's too many objects to shell, blah blah blah.

A few of my complaints & wishes about this kind of modeling:

1. Difficult to make two connected NURBS curves tangent in all dimensions. The Combine tool has some modes that approach this, but they can change things about both curves to accomplish the join.

2. Bi-Rail Sweep needs options to control the profile plane - eg perpendicular to layer plane, or normal to surface of the rail, or ???. Seems like it leans radically as it sweeps, perhaps trying to compensate for different rail lengths.

3. Bi-Rail Sweep needs an option to sweep always tangent to plan projection of chosen rail, or working plane projection.

4. Splitting this model into 3 segments could be good, but even splitting it is difficult. The split by line is always an infinite line. We could use a split by drawn segment mode.

5. This is the poster child for needing snap to "screen" intersection of 2d and 3d edges - should work if elements are touching or only virtually coincident.

6. Tangent controls and notifications need enhancements.

7. If conversion to interpolated surface worked consistently, point controlled edits could eliminate the wrinkles and overhangs and bulges.

•mumble, mumble, gripe, gripe

-B

Link to comment

for the vw development team:

i don't need a "skate bowl" tool. but need sub-parametric tools that allow the creation of a complete surface. this master surface will be on its own layer and is generated from 2d lines (layout lines) on another layer. 99% of skate parks can be solved by just 2d lines. then this master surface can generate a 3d slab.

in the video below you will see:

https://youtu.be/7a3HYECNtvQ

1. skate parks require a set of plans that show it blending into environment.

2. skate parks have controversy (good 3d fly thru and landscaping needed to calm opinions)

3. skate parks are review by skaters (this is where vw needs a browser based approach. this way the model can be released to anyone with the ability to have notes added by others, in real time

4. skate parks are expensive so to have a cost of materials generated from 3d model would be good.

the sub-parametric tools developed for building skate parks can be also used for any non skate concrete plaza. showing small slopes for water drainage etc.

has anyone actually tried to model concrete to show drainage in the model with out creating hard lines? and make it a solid? very very hard. i don't want to hear about the "slab" tool. to much time is waisted on parametric tool development. spent your time on sub-parametric tool. then we can model anything with relative ease.

Link to comment
  • Vectorworks, Inc Employee
DM, your list is right on. Another problem with my approach is that I altered the design to accommodate the software. A Loft fail can indicate non doable geometry, but it still seems kind of backwards.

This skate park should be as easy as a single command for a new surface that is a fillet between the two curves. Everything tangent. If the command fails, there should be feedback, such as highlight problem areas along the curves.

But in Vectorworks, this obvious solution is not an option. sigh. To really model this, I think the P. Cipes method is called for - Model it like you build it. Establish the upper and lower curves, screed (loft) between them with arc shaped tools. Isn't that how they build skate bowls? But that's LOTS of work in vwx. Rib after rib, then multiple lofts and adding solids, then it's too many objects to shell, blah blah blah.

A few of my complaints & wishes about this kind of modeling:

1. Difficult to make two connected NURBS curves tangent in all dimensions. The Combine tool has some modes that approach this, but they can change things about both curves to accomplish the join.

2. Bi-Rail Sweep needs options to control the profile plane - eg perpendicular to layer plane, or normal to surface of the rail, or ???. Seems like it leans radically as it sweeps, perhaps trying to compensate for different rail lengths.

3. Bi-Rail Sweep needs an option to sweep always tangent to plan projection of chosen rail, or working plane projection.

4. Splitting this model into 3 segments could be good, but even splitting it is difficult. The split by line is always an infinite line. We could use a split by drawn segment mode.

5. This is the poster child for needing snap to "screen" intersection of 2d and 3d edges - should work if elements are touching or only virtually coincident.

6. Tangent controls and notifications need enhancements.

7. If conversion to interpolated surface worked consistently, point controlled edits could eliminate the wrinkles and overhangs and bulges.

•mumble, mumble, gripe, gripe

-B

for the vw development team:

i don't need a "skate bowl" tool. but need sub-parametric tools that allow the creation of a complete surface. this master surface will be on its own layer and is generated from 2d lines (layout lines) on another layer. 99% of skate parks can be solved by just 2d lines. then this master surface can generate a 3d slab.

in the video below you will see:

https://youtu.be/7a3HYECNtvQ

1. skate parks require a set of plans that show it blending into environment.

2. skate parks have controversy (good 3d fly thru and landscaping needed to calm opinions)

3. skate parks are review by skaters (this is where vw needs a browser based approach. this way the model can be released to anyone with the ability to have notes added by others, in real time

4. skate parks are expensive so to have a cost of materials generated from 3d model would be good.

the sub-parametric tools developed for building skate parks can be also used for any non skate concrete plaza. showing small slopes for water drainage etc.

has anyone actually tried to model concrete to show drainage in the model with out creating hard lines? and make it a solid? very very hard. i don't want to hear about the "slab" tool. to much time is waisted on parametric tool development. spent your time on sub-parametric tool. then we can model anything with relative ease.

I will be wishlisting these, but I'll split them up for paperwork reasons.

I have always wondered if some sort of extreme error detection when creating surfaces was possible, as I have never seen a situation where: "Oh, look at this dramatic random bend occurring even though the source curves had no such bend, thats exactly what I wanted!" was how things went.

Link to comment

I'm Impressed, too! Nice solution, Bas.

Surface still has some bulges when I use DM's original curves and arc positions.

Maybe make a few more splits? or relocate the arcs onto outer curve centers? or redraw the flats so co radial with the outer arcs?

Bottom surface from the original bottom curves failed via the Create Surface from Curves command. But extract edges from the new solid and the command creates a (warped) surface.

-B

ubbthreads.php?ubb=download&Number=14733&filename=SurfaceFromCurves.png

Edited by Benson Shaw
Link to comment
It lofts fine in Rhino as is...... two rails, 3 profiles. Rhino also has options to rebuild the geometry or refit the rails.

+1 to all of Benson's wishes. This type of modelling needs to be improved in VW.

KM

Makes me wonder why VW cant do this, what's the difference between both, its 'just' math right?

And what makes that the surface wrap as it does?

Not to bash VW, but just wondering.

Really hope there are coming improvements to the nurbs part of VW. Using it allot last year for furniture made from sheet-metal.

Link to comment

It seems that something is not working properly with lofts at times, recently I have been working on a ship hull and could only make it look proper by splitting up the geometry. Even then there were some wrinkles, fortunately at the bottom so invisible, but the geometry was not that difficult and all NURBS were having the same direction etc. etc.

So +1 to all of the above suggestions from me as well.

And I'm impressed too, even saved this topic as a bookmark for future reference.

Link to comment

there seems to be a need to control a complete surface from the top plan, using that as your guide line geometry then have the ability to fine tune in 3d without it going wacky. this would be true of all surfaces. skate parks being just one. another would be all the paving around a shopping mall. same idea one big surface that twists to the many drains and curbs.

Link to comment

Not having VW nearby, but what I would try:

- Draw from the center of the quadrant in a side view 1 line that follows the wanted slopes of the 'bowl' and the flat outer parts

- the line has to be longer then the quadrant and the 'square' its a part of.

- sweep the line 360 deg

- shell it downwards

- subtract the unwanted geometry (with this step you make the slab rectangular instead of round)

- substract all the unwanted parts

- create a symbol

- rotate duplicate 3 times

- add all the symbols in one solid addition

- changes needed to the model? added the symbol, 'open' and 'close' the solid addition and you are done

Link to comment

yes that is what i did with a sweep. then trimmed off square. but now i need to add a swale etc.. when doing drainage planes and the surface is concrete a person always needs to tweak points in xyz to get the water to flow to a drain. this requires special adjustments that cannot be done with sweeps etc.

again the skatepark surface concept. keep in mind that the flat concrete in a skatepark is not really FLAT but will have a slight slope to shed water to drain.

so this means that the flat spots on the bottom of a bowl are not really flat but slope to the drain. so in reality there are no flat surfaces in a skate park.

hence we need to be able to control all this slope(iness) from the surface geometry somehow. similar to the grade tool that can control the surface of the DTM BUT with one big adjustment in that an interlocking web of guidelines (curves also) is made that can have slope entered into each line.

this type of control should get top preference than spending time on a "Slab" tool.

concrete is rarely if ever flat

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...