Jump to content

Site model simplifying or 'bridging'??


nca777

Recommended Posts

I have topo data (contours) from a cad survey with accurate z elevations. Problem is, when I convert to site model they simplify or bridge and lose a lot of detail and accuracy to the original survey topo. I've searched the forum and founf similar issues, but not sure this was ever resolved ??

 

Is there a setting I'm missing?

bcb 1.JPG

bcb 2.JPG

Link to comment

I think the resolution was to:

 

1.  add spot elevations and/or small contour segments halfway between other contours in valleys, etc. where the truncations tend to occur, and

2. Pray for a service pack or upgrade in 2018 that helps with this problem.:$

Edited by Rossford
  • Like 1
Link to comment

I don't think you are losing ever detail in a DTM as long as you don't use Ssimplify 3D Poly Tool.

 

I think it is just that you set your DTM display option (contour lines here ?)

to have a larger height distance as your survey lines are.

You could also set 2D contour appearance to every 5 mm. So it would show even more lines.

(Not more real information though)

Edited by zoomer
Link to comment
1 minute ago, zoomer said:

I don't think you are loosing ever detail as long as you don't use simplify the 3D Poly Tool.

 

I think it is just that you set your DTM display option (contour lines here ?)

to have a larger height distance as your survey lines are.

You could also set 2D contour appearance to every 5 mm. So it would show even more lines.

(Not more real information though)

 

I'm not sure what else I would do. These are 1' contours, set at 1' intervals at real-world elevations per the survey. Final construction documents need to be 1' contours as well. Unfortunately, this has left me using vectorworks for limited site modeling pplications but not quite meeting needs for documentation.

Link to comment

Sorry my english ist too bad.

 

Have you already bin in OIP > Site Model Settings, and applied the same contour line "resolution" as your source data ?

And it still leaves out lines ?

 

Another thing is get resulting contour lines at the same hight as the original ones.

If they contour lines created lay in between the original it will interpolate too much from the resulting 3D mesh and

they may look quite different in shape.

Link to comment
17 minutes ago, zoomer said:

Sorry my english ist too bad.

 

Have you already bin in OIP > Site Model Settings, and applied the same contour line "resolution" as your source data ?

And it still leaves out lines ?

 

Another thing is get resulting contour lines at the same hight as the original ones.

If they contour lines created lay in between the original it will interpolate too much from the resulting 3D mesh and

they may look quite different in shape.


Hmmmm...I'm not seeing anything about 'resolution.' The site model uses the original cad topo from the survey, not moving anything , so not sure there either...thanks for the help!

oip site model scrn.JPG

Link to comment
  • Vectorworks, Inc Employee

I am not highly versed in the site model, but I suspect that the "resolution" may refer to the Minor Contour Interval, which generally should be set as low as the smallest distance between two of the imported contours, if it's higher, it will round off the other contours. Sometimes setting this LOWER than the distance between the contours you are importing is needed as well. (6" or even 3" perhaps in this case to test)

I am fairly certain the Major Multiplier multiplies the minor interval, but it has been awhile since I worked with it.

Link to comment
1 minute ago, JimW said:

I am not highly versed in the site model, but I suspect that the "resolution" may refer to the Minor Contour Interval, which generally should be set as low as the smallest distance between two of the imported contours, if it's higher, it will round off the other contours.

I am fairly certain the Major Multiplier multiplies the minor interval, but it has been awhile since I worked with it.

 

Correct. The (minor) contour interval in this case is 1'. The major contour multiplier is set to '5'. FOr every 5 contours the site model generates a graphically heavier contour line. Changing the contour interval arbitraily would not be an option. thanks!

Link to comment

Yes, you set your "resolution" minor contour intervall correct,

I hoped it would be set larger and easy to fix.

Even your max/min elevation looks plausible.

 

So it seems the created contour lines are a bit interpolated.

Didn't expect this.

Maybe that means using a 2D copy of your surveyor 3D lines for plans

and the DTM just for 3D and Sections ?

Edited by zoomer
Link to comment
7 minutes ago, nca777 said:

Changing the contour interval arbitraily would not be an option.

True - but to troubleshoot more effectively, temporarily changing the minor contour interval as JimW suggested might bring back the missing / or bridged sections of the contour lines. You can always change them back to your current setting of 1'.

Link to comment

Can someone give their experiences with the setting called "smoothing interval" or "3D Grid spacing" on the right side of the site model settings box on the 2D view/Top Plan view of generated contours? 

 

Can't say I recall ever setting those other than what was standard, but maybe they have an effect?

 

Also, I haven't tried it yet, but someone recently mentioned converting the 2D polys to nurbs rather than 3D polys, which creates fewer vertices. Does that method affect final results.

And, the help manual actually suggests trying Draped Surfaces to smooth site models, but I haven't tried that either....stick with the tried and somewhat true 2D to 3D then simplify 3D poly.

Edited by Rossford
Link to comment

Okay, spent the last hour comparing the differences of 2D poly-Convert to 3D poly-Simplify to 2D poly-Convert to Nurbs - Convert to 3D poly. In the end, same number of steps and using VW suggested values for simplify 3D polys, similar, but not exact number of vertices. Don't think it would have any benefit in reducing vertices, but keeping enough to not have VW skip over some valleys when converting to site model.

 

Will try smoothing angle after lunch and report back if on one else has any answers....good use of a Friday afternoon, no?

Link to comment

So keep in mind that the surveys many of are working from to create Site Models are already (sometimes heavily) interpolated; that is, the surveyor walked the site and took a series of readings, not necessarily in any particular or strict grid pattern. Those "points" are then used by the survey software to interpolate the contours. So if you are creating a Site Model from contours, it will end up being double interpolated, so to speak.Three comments: a survey is only as good ad the person doing it; and often I ask for the survey "point file" and use those to create a Site Model ( and to check on the accuracy or frequency of the points); also, in every case (so far) the surveys and Site Models are more than accurate enough for what we do, and I don't usually fret about small (less than 1 ft) oddities. A lot of that will be worked out on site by the builder.

Link to comment
48 minutes ago, CipesDesign said:

So keep in mind that the surveys many of are working from to create Site Models are already (sometimes heavily) interpolated; that is, the surveyor walked the site and took a series of readings, not necessarily in any particular or strict grid pattern. Those "points" are then used by the survey software to interpolate the contours. So if you are creating a Site Model from contours, it will end up being double interpolated, so to speak.Three comments: a survey is only as good ad the person doing it; and often I ask for the survey "point file" and use those to create a Site Model ( and to check on the accuracy or frequency of the points); also, in every case (so far) the surveys and Site Models are more than accurate enough for what we do, and I don't usually fret about small (less than 1 ft) oddities. A lot of that will be worked out on site by the builder.

Makes me a little leery thinking about recreating topo--it's pretty standard in our practice to maintain consistency in referencing the survey from one discipline to another. The surveyor carries their own liability tied to the documents/files they create. A lot of our projects deal with pretty tight code constraints from building height limits to slope, cut and fill limits, etc, etc. Interpolation is not really acceptable for documentation. 6-12" in grade change for a landscape architect can be pretty substantial, whereas a builder might fudge it all together, a good LA will add steps, walls, terraces, grading to create defined spaces. 

Link to comment

Gotta chime in on this. Clipped contours is a pet peeve and pet project of mine.  Several comments:

1.  Take a look at several solutions with extra points in this video. It is interpolation, as others mention. But the most intense solution is pretty quick and adjusts the Site Model 3d contours to match the Top Plan 2d graphic. My extra step of 2d contour> NURBS>3d contour doesn't fix the clipping. It is just a way to cut down the excessive number of points. In my opinion, this step keeps the resulting 3d contours closer to the original 2d data than the Simplify contours option.

https://vimeo.com/195211767

 

2. If the survey is distributed in form of trusted contours made from point data, another vertex saver process is to generate the site model from new 3d loci or Stake Objects. Use Duplicate Along Path with a 3d locus or Stake on each contour.   Set the duplication distance to same as or less than (half?) the intended minor contour interval.  Color code and class each level for differentiation when viewing the source data.  Resulting DTM is an evenly spaced mesh. Unfortunately, clipping may still occur.

 

3. In my experience, when a site model suffering clipped contours is "fixed" by adding interpolation points, subsequent changes to the Minor Contour Interval generally introduces new clipping.  So, the initial Site Model contour interval should be the intended interval for design/display/print.

 

4. If the survey point data is available, consider using the points instead of contours as site model source data.  Everything between those survey points is interpolation, probably just triangle straight line interpolation, but everyone using the survey can at least agree on those initial points.

 

5. I really hope this "clipped contours" problem gets fixed soon.  It's a bug from every vwx release with site model functionality.

 

-B

  • Like 3
Link to comment

Good idea, Alan but the 2d Smoothed Contour and/or adjustments to the Smoothing Interval options offer no help with the clipping/bridging.  The 2d Smoothed Contour does alter and smooth the 2d display, but not necessarily closer to the 2d source shape.

 

The clipping problem manifests both 2d and 3d. The clipped area of a valley or headland produces a flat platform or shelf where the mesh connects parts of a contour to itself instead of to next contour up or down.

 

Even a draped surface over the site model will display the flat area. Lower resolution of the drape may deemphasize the flat, but then the overall shape of the drape less coincident to the site model surface. Not very helpful.

 

Something that would be helpful is if vwx could generate a series of interpolated mid interval contours in any selected area (or entire area) of the site model. eg if the source contours are 1.0 meter interval, vwx generates new polyline geometry interpolated at .5 meter interval. Similar to changing the Minor Contour Interval. But as far as I know, this is display only, no selectable geometry is created.  Even if the interpolation produced only 2d Top/Plan polylines or loci, new 3d contours or points could be generated and placed with current tools. This new interpolated data could be introduced as site modifier contours, or a new category of source data - y'know, alternate source data.

 

-B

  • Like 1
Link to comment

nca777,

I went through this entire string and perhaps I missed something. Perhaps the issue is with your original approach in importing the "cad file". Do you have the option for the surveyed to send you the TIN instead of his output of contours? I have found if you can get this you will have much more accurate site-model. I would also recommend that you draw you site-model boundary before creating the site model.

Link to comment
On 5/1/2017 at 11:00 AM, rowbear97 said:

nca777,

I went through this entire string and perhaps I missed something. Perhaps the issue is with your original approach in importing the "cad file". Do you have the option for the surveyed to send you the TIN instead of his output of contours? I have found if you can get this you will have much more accurate site-model. I would also recommend that you draw you site-model boundary before creating the site model.

 

Not sure, maybe I could get the tin. Sometimes I find it on a frozen layer in CAD. I'll look into it. thanks

Link to comment
  • 4 weeks later...

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...