Camera locations

December 22, 2013

I can reduce the complexity of the model by defining the camera locations for the still and video shots.  Areas that are far or out of view of the camera will not have to be modeled or textured in detail.  Here are the locations that I have in mind:

  • Inside the compound.  Ground level and on the ramps looking inside and outside the walls.  On the gun tower over the church. From the window and back stairs of the convento.
  • A few 10s of meters outside the compound walls looking towards the compound.  In particular, from the South, East and North.  The Northwest view of the pecan tree.
  • At the Portrero street footbridge panning from the town to the compound.
  • At the Powder House looking toward the town and compound
  • Main Plaza toward San Fernando Church
  • Veramindi house.  The house details and panning from the church to the compound.
  • Military Plaza. Plaza buildings and the back of the church. Move up to the level of the church bell tower and view the compound.
  • Birds eye view approximating the area overview scene in Nelson’s The Alamo, An Illustrated History.
  • Bird’s eye view overlooking the compound.

Terrain work

December 16, 2013

My first project is to redo the terrain for the area around Bexar (San Antonio).  This is probably the third or fourth iteration.  The first terrain used in the TrueSpace model was limited to a small area around the Alamo compound.  A matte plane was used to simulate the view of the buildings in the town and hills.  The terrain was completely flat.  The second version had a much wider area and used a height map that I generated from USGS topo height data.  There were several problems with this.  First, the mesh consisted of an array of vertices separated by several meters.  This resulted in a high poly count but the detail of the area around the river was not satisfactory.  The second problem was the unevenness of the terrain made it hard to place structures (houses or walls) without showing a gap between one edge and the ground.  On the other hand, at the scale and viewpoint of my cameras, the detail that was causing the problem added no perceptible realism to the scene.

To solve these issues, I’ve decided to break the ground texture into regions that have different levels of detail.  The hills beyond the eastern most acequia will use the height map data.  The area around the river will use a contour topology that is subdivided to get the desired detail.  The other regions will be made flat for easy structure placement.  Hopefully, this will put the detail where it is needed and reduce the poly count where it is not.

My first task is to realign the topology according to a better map that I created for the 13-Days project.  This map was created to align the river, acequias, structures, fields and roads according to a modern topo map with adjustments from other historical sources.  For example, the river has a vastly different course today.

An are map based on topo and historical maps.

An are map based on topo and historical maps.

I’m back!

December 8, 2013

Over 2 years since the last post here.  Wow.

In 2010, I got sidetracked on another Alamo-related project.  I was asked to create an animation (2D) of the action that led to the battle of the Alamo.  I decided to use a Microsoft technology called Silverlight because it was supported for multiple platforms.  If you have a modern PC with Silverlight installed for your browser, you can check it out here:

13 Days: The Siege and Battle of the Alamo

I was fortunate to be able to demonstrate this program at the 175th anniversary of the battle in San Antonio courtesy of Valero Oil Company.

This taste of programming got me hooked on Windows 8 and app development.  While participating heavily in the user forums, I was contracted to be the technical editor for a couple of books on app development.  In 2012, I released my own app for displaying worldwide earthquake data and recently updated it for Windows 8.1.–QuakeView.

In the meantime, Blender has undergone several iterations of improvements.  New tools have also come on the scene that should be interesting to try.

My near term goal is to clean up the terrain mesh and add textures to make a convincing baseline scene for rendering 1836 Bexar.