Difference between revisions of "22 March 2006"
From Organic Design wiki
(3D Space Ready!) |
m (Caretaker: typo) |
||
| (3 intermediate revisions by one other user not shown) | |||
| Line 3: | Line 3: | ||
The peer interface uses two geometry models, 3DSpace (polar) and BoxModel (rectangular). This 3D Space not only handles the z-ordering and the usual ''x,y,z → x,y,scale'' coordinate reduction, but also allows for a hierarchy (tree) of 3D objects who's coordinates and dynamics are relative to their parent. This will allow the construction of a dynamic 3D model of the Nodal reduction algorithm. | The peer interface uses two geometry models, 3DSpace (polar) and BoxModel (rectangular). This 3D Space not only handles the z-ordering and the usual ''x,y,z → x,y,scale'' coordinate reduction, but also allows for a hierarchy (tree) of 3D objects who's coordinates and dynamics are relative to their parent. This will allow the construction of a dynamic 3D model of the Nodal reduction algorithm. | ||
*[[3D Space Example]] | *[[3D Space Example]] | ||
| + | *[[Geometry]] | ||
</table> | </table> | ||
| − | *[[User:Nad|Nad]]: A bug that was in the original | + | *[[User:Nad|Nad]]: A bug that was in the original implementation of the 3D space was finally solved :-) It turns out that ECMA parses the hash-keys as dot-separated paths (due to the mergin of object-model and hash-array-structure in the syntax). The GUID's that were being used for the 3D-object instance names were constructued from Math.random() and contained a decimal point that was being interpreted as a path-separator for the object model. ''(see actual edit of the horrible bug fix [http://www.organicdesign.co.nz/wiki/index.php?title=space.as&diff=14480&oldid=14479 here]!)'' |
Latest revision as of 17:57, 28 October 2006
The peer interface uses two geometry models, 3DSpace (polar) and BoxModel (rectangular). This 3D Space not only handles the z-ordering and the usual x,y,z → x,y,scale coordinate reduction, but also allows for a hierarchy (tree) of 3D objects who's coordinates and dynamics are relative to their parent. This will allow the construction of a dynamic 3D model of the Nodal reduction algorithm. |
- Nad: A bug that was in the original implementation of the 3D space was finally solved :-) It turns out that ECMA parses the hash-keys as dot-separated paths (due to the mergin of object-model and hash-array-structure in the syntax). The GUID's that were being used for the 3D-object instance names were constructued from Math.random() and contained a decimal point that was being interpreted as a path-separator for the object model. (see actual edit of the horrible bug fix here!)
