7 replies to this topic

[zapquoc]

what might be the usefulness of TIN?

[Hans van der Maarel]

That's a very general question, could you elaborate a bit more about the context?

Personally, I use TIN's a lot for my 3D work, not so much for mapping. I generate a TIN from a collection of spot elevations and then drape 2D features over it so they will be shown at the proper elevation.

Hope this helps

[mike]

i agree with Hans. it's very useful when you have both the TIN data and then an orthophoto or raster to drape over it. i've done some 3D models of canyons, volcanos and mountains using TINs and rasters. i've also created a 3D model of a lake, using bathymetry DEM data. so i guess it depends on what you have to represent.

[kass]

he might be wondering (I am anyway), what's the advantage over using an dem and/or hillshade with a raster draped over it? is it just better for analysis?

[Hans van der Maarel]

he might be wondering (I am anyway), what's the advantage over using an dem and/or hillshade with a raster draped over it?  is it just better for analysis?

<{POST_SNAPBACK}>

Since a TIN is irregular, it can be more detailed in places where it's needed and less detailed in places where it's possible. A DEM has a certain resolution and will always have that many points. I.e. you can describe a perfectly flat, square, plane in DEM with gazillions of points, or do it in TIN and only need 2 triangles.

The intended application might also favor TIN over DEM. I've done work for a 3D environment that was built in 3D Studio Max, which uses triangular faces to describe 3D objects. Feeding it a TIN made it possible to influence the amount of data in an earlier stage of the process.

I think it really boils down to what you want to use it for and what data you are putting into it.

[frax]

wouldn't a TIN perform much better for displays too, in 3d games you would have more tin-like shapes for instance, and it would be hardware accelerated.

[mike]

he might be wondering (I am anyway), what's the advantage over using an dem and/or hillshade with a raster draped over it?  is it just better for analysis?

<{POST_SNAPBACK}>

i guess it all depends on what type of data is available to you. sometimes only a certain data type may be available and you will have to work with whatever you have. of course, some software can cconvert DEMs or TINs from each other, but that option is only available if you have the software. depending on what you need to do, one data type may work better. if you're doing analysis, then a DEM will work better for you. if you are doing a 3d model of some type, either can work for you.

wouldn't a TIN perform much better for displays too, in 3d games you would have more tin-like shapes for instance, and it would be hardware accelerated.

i've worked with some very large TINs... and they were difficult to work with b/c of their size. and i have a pretty fast computer too. it's the software that it's handled in. some handle TIN better than others.

[Kartograph]

You need a tin to interpolate Isolines from measurement points. Short of real Geostatistics like Kryging, there aren´t many other options to do this.
If you have the data already, no real need for a TIN. But somewhere in the process, most area-like data was a TIN at least once. If it is derived from discrete measurement points, that is.

BTW:
Anybody here who also had to interpolate isohypses using a TIN by hand? Took me hours and hours, just for a small ditch in the landscape, but was a neat way of learning.