<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content=text/html;charset=ISO-8859-1>
<META content="MSHTML 6.00.2800.1561" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY text=#000000 bgColor=#ffffff>
<DIV><FONT size=2>Simon,</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> VTP <A href="http://vterrain.org/"><FONT
size=3>http://vterrain.org/</FONT></A> (to paraphrase Schuyler Erle
"the best kept secret open source geo app on the net") uses libmini (<FONT
size=3> </FONT><A href="http://www.stereofx.org/terrain.html"><FONT
size=3>http://www.stereofx.org/terrain.html</FONT></A>) an open source library
for tiling images (textures) and elevations models ("height
fields").</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> In the VTP implementation, the authoring
tool VTBuilder takes a user-specified set of images and creates a tileset.
the Tileset has an "INI" file to describe it's properties:</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT
size=2>[TilesetDescription]<BR>Columns=12<BR>Rows=16<BR>LOD0_Size=1024<BR>Extent_Left=370826.6795802761<BR>Extent_Right=391921.9795802761<BR>Extent_Bottom=2408667.927330024<BR>Extent_Top=2436626.127330024<BR>CRS=PROJCS["NAD83
/ UTM zone 4N",GEOGCS["NAD83",DATUM["North_American_Datum_1983",SPHEROID["GRS
1980",6378137,298.2572221010002]],PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-159],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["metre",1]]<BR></FONT></DIV>
<DIV><FONT size=2>From the libmini doc, the tiles "can be saved in the PNM
format (PGM/PPM for grayscale/RGB images), but the native DB format of the Mini
Library should be preferred."</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>The tile files are named (for the db
format):</FONT></DIV>
<DIV><FONT size=2> tile.0-0.db
- first row, first column, highest level of
detail</FONT></DIV>
<DIV><FONT size=2>
<DIV><FONT size=2>
<DIV><FONT size=2> tile.0-0.db1 </FONT></DIV>
<DIV><FONT size=2> tile.0-0.db2
</FONT></DIV>
tile.0-0.db3 - first row, first
column, lowest level of detail</FONT></DIV></FONT></DIV>
<DIV><FONT size=2>
<DIV><FONT size=2> tile.1-0.db
- second row, first column, highest level of
detail</FONT></DIV> :</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>Brent Fraser</DIV></FONT>
<DIV><FONT size=2></FONT> </DIV>
<BLOCKQUOTE dir=ltr
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">.
Could somebody outline the solution used by the Virtual Terrain
project?<BR><BR>Cheers,<BR><BR>Sy<BR><BR></BLOCKQUOTE></BODY></HTML>