<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Hi Jaak,<br>
      <br>
      thanks for your detailed response!<br>
      I was in the hope that nutiteq somehow provides/sells the height
      information (beyond OSM).<br>
      <br>
      Regards,<br>
      Peter.<br>
      <br>
    </div>
    <blockquote
      cite="mid:56CB9DDC-6F9B-452D-BBE3-5AA5A6259E74@nutiteq.com"
      type="cite">
      <div>Hi,</div>
      <div><br>
      </div>
      <div>Regarding 3D buildings you can easily have "shoebox" 3D
        models, which are actually 2D polygons with height. See code
        sample from  <a moz-do-not-send="true"
href="https://github.com/nutiteq/hellomap3d/wiki/Online-OSM-3D-buildings">https://github.com/nutiteq/hellomap3d/wiki/Online-OSM-3D-buildings</a> .
        Height value is taken from OSM tags (height and floors), and if
        it is missing (well, this means about 99.9% cases) just a fixed
        average value.</div>
      <div><br>
      </div>
      <div> Having real buildings with textures is more complex task.
        There is another wiki page with code instructions: <a
          moz-do-not-send="true"
href="https://github.com/nutiteq/hellomap3d/wiki/Nml-3d-models-map-layer">https://github.com/nutiteq/hellomap3d/wiki/Nml-3d-models-map-layer</a> .
        I added now also a sample of it as Offline3DActivity to the
        HelloMap/AdvancedMap samples, and a link to wiki to download set
        of Barcelona buildings which are taken from the </div>
      <div><br>
      </div>
      <div>Anyway, there are the steps needed:</div>
      <div><br>
      </div>
      <div>1) get the data. We don't have our own 3D city datasets
        (yet), you should be able to license it from TomTom or Navteq,
        or from other providers. If you can suggest one for us, please
        do. Or you can use free 3D models from <strike>Google</strike> Trimble
        3D Warehouse. My understanding is that you are actually free to
        use these models in your apps, but I don't have official
        confirmation from Google or Trimble about it. You can always
        create your own dataset.</div>
      <div><br>
      </div>
      <div>2) convert it to a suitable format for Nutiteq SDK. We have a
        tool which reads KMZ/collada files and creates nmldb file which
        is heavily mobile-optimized special 3D format. This data can be
        used as offline package (nmldb file), or put to a server and be
        used as online service. This converter is licensed separately as
        add-on. We do test conversions with your data for free. </div>
      <div><br>
      </div>
      <div>3) Add a 3d model layer to your map, see the wiki page above</div>
      <div><br>
      </div>
      Jaak
      <div><br>
        <div>
          <div>On 07.04.2013, at 16:16, Peter K wrote:</div>
          <br class="Apple-interchange-newline">
          <blockquote type="cite">
            <meta content="text/html; charset=ISO-8859-1"
              http-equiv="Content-Type">
            <div bgcolor="#FFFFFF" text="#000000">
              <div class="moz-cite-prefix">Hi Jaak,<br>
                <br>
                I tried the example and it worked out of the box - very
                nice :) ! (to enable 3D view I had to change
                setTilt(45))<br>
                <br>
                How can I enable seeing buildings like in your
                screenshot:<br>
                <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://dl.dropboxusercontent.com/u/3573333/mapxt_3d_tallinn_device-2012-07-25-124845.png">https://dl.dropboxusercontent.com/u/3573333/mapxt_3d_tallinn_device-2012-07-25-124845.png</a><br>
                <br>
                or is this only possible with your custom file storage?<br>
                <br>
                Regards,<br>
                Peter.<br>
                <br>
              </div>
              <blockquote
                cite="mid:23F2B929-8A08-4D18-8194-660AF823FB54@nutiteq.com"
                type="cite">Hi,
                <div> regarding 1) . Maybe I used somehow older version
                  of lib. Now I tried with Maven dependencies and it
                  seems to get correct results, like in <a
                    moz-do-not-send="true"
href="https://dl.dropbox.com/u/3573333/public_web/graphhopper_sample1.png">https://dl.dropbox.com/u/3573333/public_web/graphhopper_sample1.png</a></div>
                <div><br>
                </div>
                <div> ps. I uploaded the graphhopper-1.0-SNAPSHOT to my
                  Maven repo (<a moz-do-not-send="true"
                    href="https://repository-nutiteq.forge.cloudbees.com/release/">https://repository-nutiteq.forge.cloudbees.com/release/</a>),

                  is there better place to fetch it from?</div>
                <div><br>
                </div>
                <div>Jaak</div>
                <div><br>
                  <div>
                    <div>On 04.04.2013, at 19:06, Peter K wrote:</div>
                    <br class="Apple-interchange-newline">
                    <blockquote type="cite">
                      <meta content="text/html; charset=ISO-8859-1"
                        http-equiv="Content-Type">
                      <div bgcolor="#FFFFFF" text="#000000">
                        <div class="moz-cite-prefix">Hi Jaak,<br>
                          <br>
                          cool, this is really great news! As I already
                          wanted to look into this, but hadn't the time
                          so far.<br>
                          <br>
                          Regarding 1) The returned value is as expected
                          (normal) meters. Hopefully I can digg into the
                          code next week.<br>
                          <br>
                          Regarding 2) ghz is a zipped folder and can be
                          opened via the Helper.unzip<br>
                          <br>
                          > mapsforge library<br>
                          <br>
                          Oh, so the mapsforge data can be opened? Nice!<br>
                          <br>
                          Regarding 3) This will be solved in issue-17.
                          It is the problem of mapping from "real world
                          gps coordinates" to graph node ids.<br>
                          You can partially work around this by
                          increasing the precision of
                          Location2IDQuadtree index<br>
                          or checkout the master (not really recommended
                          yet, as incompatible file layout and probably
                          other minor issues)<br>
                          <br>
                          Regards,<br>
                          Peter.<br>
                          <br>
                          <br>
                        </div>
                        <blockquote
                          cite="mid:FA90F308-2FD1-49FA-BC9E-25A6B22553E1@nutiteq.com"
                          type="cite">Hello,
                          <div><br>
                            <div> I just pushed a sample with
                              Graphhopper routing to Nutiteq 3D SDK
                              AdvancedMap3D samples: <a
                                moz-do-not-send="true"
                                href="https://github.com/nutiteq/hellomap3d/">https://github.com/nutiteq/hellomap3d/</a>,
                              see GraphhopperRouteActivity.java .
                              Nutiteq SDK is quite similar to Mapsforge,
                              just with 2.5D perspective and rotation
                              capabilities. And with a lot of different
                              data source Layer implementations, from
                              geotiff and shapefiles to mapsforge
                              library.</div>
                            <div><br>
                            </div>
                            <div> Issues with grasshopper:</div>
                            <div><br>
                            </div>
                            <div>1) major problem - all the routes in my
                              area (Estonia) are a bit more than 2x too
                              short. Instead of 100 km I have 45 or so.
                              Could it be that it is so much north and
                              calculations of distances become wrong? If
                              the "meter" is really "mercator meter"
                              then the value could be even right, but
                              calculation of mercator meters to real
                              meters, when the route is from south to
                              north, is not trivial (or not possible at
                              all). I created test dataset for Baltics 3
                              countries: <a moz-do-not-send="true"
                                href="https://dl.dropbox.com/u/3573333/baltics.ghz">https://dl.dropbox.com/u/3573333/baltics.ghz</a>
                              , where it is observable. </div>
                            <div><br>
                            </div>
                            <div>2) some minor issues. I did not figure
                              out how to open properly .ghz file with
                              the API. So the sample asks now for .map
                              file inside <map>_gh folder, so the
                              files must be uncompressed in SDCARD. But
                              it would be nicer to enable just to open
                              ghz file, and GH would uncompress it
                              automatically.</div>
                            <div><br>
                            </div>
                            <div>3) start and endpoints are sometimes
                              not optimal, quite far from clicked
                              location, and there seem to be much closer
                              nodes to given points. I added lines from
                              clicked location to first and from last
                              location to make it nicer. Maybe some
                              tolerance in graph generation, or in
                              routing?</div>
                            <div><br>
                            </div>
                            <div><br>
                            </div>
                            <div>Your pull requests with fixes to the
                              demo app are very welcome, quite possibly
                              I just did something wrong</div>
                            <div><br>
                            </div>
                            <div>Jaak</div>
                            <div><a moz-do-not-send="true"
                                href="http://Nutiteq.com/">Nutiteq.com</a></div>
                          </div>
                        </blockquote>
                      </div>
                    </blockquote>
                  </div>
                </div>
              </blockquote>
            </div>
          </blockquote>
        </div>
      </div>
    </blockquote>
    <br>
  </body>
</html>