Jump to content

Mainnormal file Where is this generated?


Recommended Posts

The main normal is a gray scale file that shadows the ground for differences in elevation by light.

My question: is this generated within the builder? Or is it a third party tracer that makes this.

Another question: blocked cells indicated by red on the pathfinding tga. Can this be regenerated? It seems the old cells remain and block movement if you change a map and remove the obstacle. Can you simply colour out the red to white in photo shop. Hunting for these on the map is difficult with the little cell PFS cursor.

thanks

Link to comment
Share on other sites

How to create the mainnormal.dds file:

You need to have the Nvidia DDS plug-in installed in Photoshop (http://developer.nvidia.com/object/photoshop_dds_plugins.html).

open NearHF.raw in photoshop using the following options:

channels->count: 1

depth: 16 bits

Bytes order: IBM PC

Convert image to RGB

Run the image through the Nvidia DDS plug-in with the following options:

Height generation:

filter type: 4 samples

MinZ: 0

Scale: 600

3D view options:

Animate light. turn it on

Alpha Field: select Unchanged

click OK to valid your settings and create the normal map.

You may need to blur the image a little bit if you find the result is a bit too pixellated.

Link to comment
Share on other sites

About the red pathfinding cells.

If everything is OK with your map, there should be no need to repaint the red pathfinding cells when you delete a building: simply delete the building, then save the map. While saving, the editor will recalculate the pathfinding cells automatically. You'll not see the result immediately, you'll have to reload the map to check that the red pathfinding cells are no longer here.

If anything goes wrong and the pathfinding cells are no longer working, you may need to force the editor to recalculate them. First delete the content of the pfsmap.bin and pfswatermap.bin (files created automatically to allow a faster loading time of the map). Simply open them in notepad, then delete everything in the file, and save again with the same name to create an "empty file". Perhaps simply deleting the files will work (they might be created again when you save the map - not sure about that). Better to make a backup of theses files first before trying to modify anything.

To recalculate the pathfinding cells, use the menu View -> recalculate pfscells in the map editor.

Link to comment
Share on other sites

Oh So close.....The map appears very dark.

I converted to rgb 16, used the nvidia filter to the settings you specified, and saved as a DDS file from photoshop. This saved it as a blue colored file 16 bit. The contours are correct but the map is too dark. It has nothing to do with brightness. I also tried converting this rgb bluish mainnormal file to grayscale and bump that up to 32 and 24 bits but the map still comes out very dark.

Link to comment
Share on other sites

Hi Knokke, using an existing game map as an experiment tourmauville I took the Near HF raw file and processed it through the Nvidia photo shop filter using the instructions above. What I got was this. Compared to what should have been as in the bottom pic.

ddsrgb16.jpg

What I was hoping to get was what I pulled from the game below. So somehow the image is not being save correctly. Am I missing a setting? I am using plug-in 8.23.110.xxx

mainnormal.jpg

Link to comment
Share on other sites

What is in the bottom pic is not really a bump map. it is a greyscale picture showing the relief of your map with highlights and shadows (light seems to come from the top of the picture). Your picture shows a highlight on the side of each "canyon" that receive light, and a shadow on the other side.

Processing this texture in the Nvidia DDS plug-in will give you a normal map, but it will not reflect the real elevation you're expecting. Darker and lighter areas on the textures are not supposed to be a depiction of highlight and shadows, but are supposed to depict micro-details of the ground elevation: the lighter the color, the higher the elevation is. White is for the highest point on the map, while black would be the lowest elevation, with all the gradations in between depicted in various shades of grey. So the lighter shades of grey should be found on the flatter higher areas, while the bottom of the canyons should be the darkest where the ravines have more depth.

With which 3D utility did you generate the greyscale picture? I'm using microdem to create all the elevation files I need, and I remember there is a mode where you can extract a greyscale texture like the one you show above from your SRTM elevation data, but its purpose is only to help visualise the relief of the DEM, not to create an elevation map.

It's probably the reason why you're not able to obtain a correct normal map.

Besides, I don't understand what you mean by "What I got was this. Compared to what should have been as in the bottom pic"

Once you process your greyscale picture in the Nvidia plug-in, you should obtain a "blue" image (or rather a Red-Green-Blue) picture where the amount of each of theses three colors correspond to a x,y,z orientation of each pixel of the texture.

Link to comment
Share on other sites

×
×
  • Create New...