I decided to test how HDRI gets wraped onto a sphere. It turned out that it
does it not so well :/ (it happens with any HDRI and any size of sphere)
Is there any solution?

http://img143.imageshack.us/img143/5183/hdritesttgaup0.jpg

Post a reply to this message

"motorsep" <nomail@nomail> wrote:
> I decided to test how HDRI gets wraped onto a sphere. It turned out that it
> does it not so well :/ (it happens with any HDRI and any size of sphere)
> Is there any solution?
>
> http://img143.imageshack.us/img143/5183/hdritesttgaup0.jpg

Sorry, forgot about material code:

pigment {
       image_map {
              hdr
             "c:BlenderTextureshdr02dh006lp.hdr"
             once
             interpolate 2
             map_type 1
         }
     }
     finish {
          ambient 1.0
   diffuse 0.0
     }

Post a reply to this message

never mind :) I got it. For HDRI MegaPov has special map_type that equals 7

Post a reply to this message

"motorsep" <nomail@nomail> wrote in message 
news:web.451c72fa9f6ff81ce79570f0@news.povray.org...
> never mind :) I got it. For HDRI MegaPov has special map_type that equals 
> 7
>
>

Make sure your map-type matches your image type as HDRI probes of various 
types are found on the web.  The special map_type 7 is for mapping angular 
maps to a sphere.  Angular maps are circular in shape and look similar to a 
direct picture of a mirror ball, but the two should not be confused, as they 
are quite different.  For an angular map, the distance from the center of 
the image corresponds linearly to the angle of the line of sight, whereas 
with a mirrorball image, the angles get crunched together as the angle 
increases.  With angular maps, less resolution loss occurs at the higher 
angles (depending on how it is assembled however).  Several probes from Paul 
Devbec's site (don't have the link onhand) are angular and can be used with 
the map-type 7.  Mirrorball images can't be used directly, they must be 
converted to either angular or longitude-latitude and mapped appropriately. 
Longitude-latitude is probably the most common type to be found, and the 
only one I like to use (all the light probes I have made my self are in this 
format).  This corresponds to the traditional spherical mapping of a 
rectangular image to a sphere (it doesn't require a special mapping type), 
the x coordinates correspond to the 'longitude' or east-west angle and the y 
coordinates correspond to the 'latitude' or north-south angle.  With 
lat-long maps, the rectangular image is basically stretched towards the 
poles, basically giving 'extra' information, the opposite of mirrorball.

Hope this is of help.  I know there are sites out there that can explain it 
better.  I'll try to recover some tomorrow if I can for you.

-tgq

Post a reply to this message

"Trevor G Quayle" <Tin### [at] hotmailcom> wrote:

> angles (depending on how it is assembled however).  Several probes from Paul
> Devbec's site (don't have the link onhand) are angular and can be used with

Sorry, Paul Debevec

>
> Hope this is of help.  I know there are sites out there that can explain it
> better.  I'll try to recover some tomorrow if I can for you.
>
> -tgq

This HDR Shop tutorial explains the different types a bit with
illustrations.
http://gl.ict.usc.edu/HDRShop/tutorial/tutorial4.html

-tgq

Post a reply to this message