Terrain from contours - improve on native 'from contours'?

[fredo6]

@utiler said:

Fredo, would you like contours in DWG or SU?

I prefer SU as I am not familiar with CAD and Archi software (I don't know DWG format).

Thanks in advance

Fredo

[pilou]

@unknownuser said:

I prefer SU as I am not familiar with CAD and Archi software

Except one! SU Itself

[utiler]

Here's a few contour Skippy's, Fredo.

Survey01 - contours or a regular residential site of about 800sqm with a 15m fall. Contours at 250mm....

survey01.skp

Survey02 - contours over a commercial development site of about 2800sqm. Contours at 200mm increments with minimum grad.

survey02.skp

[fredo6]

Andrew

First of all, thanks very much for posting these 2 samples. They are extremely useful for fine tuning the algorithms.

I have problem about the calculation of the concave hull for the first one, but here is the second one, in 2 flavors:

• the original altitude (which are very flattish)
• the exaggerated altitudes which help me to better check the calculations

utiler_survey02.skp

Fredo

[Krisidious]

subscribed...

[utiler]

No problem, Fredo, I'm not surprised at issues with the first one as it was a survey of a site with an existing residence on it which prevented a uniform data to be taken...

What you've done with the other one looks good! Can't wait to see how this plugin ends up. I'll see if I can dig out some other survey plans off file and convert them for you.

[fredo6]

@utiler said:

No problem, Fredo, I'm not surprised at issues with the first one as it was a survey of a site with an existing residence on it which prevented a uniform data to be taken...

I evolved the algorithm for calculating the concave hull on your complex model and could obtain a result on your model, without any change, here attached.

A lot of the problem come from the small terminations hooks in several contours, like illustrated on the picture below

The issue with terrain construction is that you need to have a correct hull in order to determine the spatial distribution of the influence of the contours.

utiler_survey01.skp

Fredo

PS: I hope to finish a first version in 2-3 weeks time, as I have actually very few moment available currently

[EarthMover]

Can't wait! Thanks Fredo!

[utiler]

that is looking very sweet, Fredo!!!! Can't wait as well!!!

[Krisidious]

here's a topo I just got from a survey company... if you need tests.

---

HANNA-TOPO.dwg

[EarthMover]

Here's another set of contour lines from my current project. All welded up.

---

Coleman Topo.skp

[pbacot]

I've gone ahead and started using one of the tests you made of my contour set, Fredo. I don't have a lot to do as yet but it was really clean leveling some areas and "filling" others with Artisan. Maybe we can illustrate the cut and fill on the project with before and AFTER terrains. Thanks!

[fredo6]

@earthmover said:

Here's another set of contour lines from my current project. All welded up.

Adam,

Here is what I obtain on the original file. I think however that the contour should be simplified a little bit because there are a lot of edges and it slows down the calculation.

Earthmover Coleman Topo.skp

Fredo

[EarthMover]

A quick play with the contours. Simplified the curves using Thom's Edge Tools. Then did a From Contours terrain. Then a Drape Cpoints / Triangulate method to retopo. Then opened the high res contour lines in 3ds Max and used the free Populate Terrain plugin to generate a similar terrain. Then brought it back into sketchup and compared the two.

---

Earthmover Coleman Topo (1).skp

[dtrarch]

Terragen2 download
http://planetside.co.uk/index.php?option=com_content&view=article&id=30&Itemid=170
Someone here was looking for this and here tiz

dtr

[fredo6]

By the way, did anyone try SimuTerra? (see this post)

Fredo

[Chris Fullmer]

Does anyone know if Erez is still maintaining his Simuterra program? We haven't seen him around in ages!

[solo]

Nice clean results.

I have a set of contours that I'd like to compare if you have the time, I used sandbox and got a mess, wondering if this will give a better result. Here is the skp, however I'm wondering if the dwg would not be better.

---

test contours.skp

[fredo6]

@solo said:

Nice clean results.

I have a set of contours that I'd like to compare if you have the time, I used sandbox and got a mess, wondering if this will give a better result. Here is the skp, however I'm wondering if the dwg would not be better.

Here is the result on your model without any change, with the current version of TopoShaper. Grid definition is 60 x 27. I did some tests with a higher definition but cannot post it back because the size of the skp file is bigger than 4 Mb.

Note that your contours add up to more than 44,000 edges, with a quite complex concave hull. So it takes a minute to calculate it and there are still some problems on the borders.

Solo test contours.skp

Fredo

[solo]

That's very clean, wow, this is awesome.