BlenderGIS elevation height differs between computers

[bobg3723]

I created a map with BlenderGIS with srtm30 height data on an i3 processor system with 16 GB system ram, integrated GPU
When I tried creating the same map, same data on an older system with an AMD 8150 with 8 GB ram, 8 GB pci-e graphics card, the map's height data was not a detailed in heights (mountain, and peak were smoothed out and not as bumpy) in the 3D view on either perspective or orthographic. It's as if the mesh nodes were not as dense. The map is 45 km wide by 85 km long.

When the BLEND file converts to a 200k DXF file. It's big. I'm more interested in fast rendering instead of pretty rendering. Is there a setting in Blender I can adjust on the AMD system to compensate for this loss in height resolution?

Or am I just lacking in computer ram on the older system?

[domlysz]

I suspect the extents of your 2 tests are not exactly the same. Larger area need more vertices : try to increase the subsurf level in the modifier stack or add another one.
By the way 45x85km is probably a bit large for Blender especially if you expect a good resolution.

[bobg3723]

I went back and tried to repeat the problem on the original i3 computer and the problem is also there, now. Here's a picture of the original peninsula I converted to DXF and the a repeat on same computer a few days later.

[bobg3723]

PROBLEM SOLVED!

Your suggestion to subdivide the mesh worked. My problem was that I forgot to subdivide the mesh prior to getting the SRTM30 elevation from the server. It works perfectly on both computers. The x and y dimensions are 96km by 166km.

I did discover one more problem I didn't mention, but I solved that by scaling. It seems that the x and y scale of the map is different from Google Earth and it's by a significant percentage. But that's a non-issue if I re-scaled the map.

[domlysz]

Take a look at the FAQ about scale distorsions :
https://github.com/domlysz/BlenderGIS/wiki/FAQ#why-my-measurements-on-basemaps-reference-seems-wrong-

[bobg3723]

Dear Sir, I learned something today. Thank you, domlysz.

…

On 05/11/2021 4:47 AM domlysz ***@***.***> wrote: Take a look at the FAQ about scale distorsions : https://github.com/domlysz/BlenderGIS/wiki/FAQ#why-my-measurements-on-basemaps-reference-seems-wrong- — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub #485 (reply in thread) , or unsubscribe https://github.com/notifications/unsubscribe-auth/AT7VXDP6HUJIBT6F26VG6LDTND4LPANCNFSM44LEFUHA .

[bobg3723]

I get it about distortion. If the latitude is a point in the formula for correct distance X = COS(Latitude), how do I arrive at the correct measure of a Tape Measure that spans over multiple latitudes or longitudes or diagonally across both?

…

On 05/11/2021 4:47 AM domlysz ***@***.***> wrote: Take a look at the FAQ about scale distorsions : https://github.com/domlysz/BlenderGIS/wiki/FAQ#why-my-measurements-on-basemaps-reference-seems-wrong- — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub #485 (reply in thread) , or unsubscribe https://github.com/notifications/unsubscribe-auth/AT7VXDP6HUJIBT6F26VG6LDTND4LPANCNFSM44LEFUHA .

[domlysz]

There are some complex formulas to compute the distance between 2 angular points on the ellipsoid, but it's really not what you want.
If you need good measurements then you should use an adaptated projection system that minimize the distorsions on your area of interest. That's what projections are for.

[bobg3723]

The projection coordinate system I'm going to re-project to is (EPSG:27700) for the U.K. I'll learn how to do that with QGIS and when I've understood that process I'll try and use the data for that projection to bring into BlenderGIS. I'll have to study up on that process so I can be in a position to ask you pertinent questions. I just don't know enough to know what to ask, so until then thank you for your suggestions. Please feel free to suggest anything I may not have considered.

…

On 05/12/2021 7:39 AM domlysz ***@***.***> wrote: There are some complex formulas to compute the distance between 2 angular points on the ellipsoid, but it's really not what you want. If you need good measurements then you should use an adaptated projection system that minimize the distorsions on your area of interest. That's what projections are for. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub #485 (comment) , or unsubscribe https://github.com/notifications/unsubscribe-auth/AT7VXDOUEFBUPUNCMPL4EI3TNJZG5ANCNFSM44LEFUHA .