Skip to content
This repository was archived by the owner on Jul 30, 2023. It is now read-only.

3. Developer's guide

Nik Yotis edited this page Jul 10, 2016 · 2 revisions

Arbitrary Output Values (AOVs) are organised in FLOAT/RGBA layers. A driver is a pluggable output-file writer that uses Arnold's built-in AOV pipeline to write output image files and send pixels to a display device. The driver gets AOVs one bucket at a time, which is ideal for tiled image formats. This way one can iterate through all the AOV layers contained in the bucket per single layer - bucket size 64x64 pixels.

lightsPreview allows the artist to run multiple Arnold sessions in parallel and render images in the same RV session. Each instance of the display driver running an Arnold render is identified with a unique name in order to avoid conflict between naming ports. Support for multiple instances of the driver -running in different Arnold processes- that connect to RV simultaneously is provided by passing one string (PID or anything uniquely identifying an instance of the renderer running). Note also that different Arnold sessions should show up with different names.

3D artists may pick one render's Region of Interest (ROI) to be redrawn by holding the middle mouse button and dragging the mouse - no support for Wacom tablets. Two RV events are triggered: one keyboard button and one mouse event respectively.

Note that the ROI functionality takes into account any Zoom or Pan applied to the Maya scene. RV's eventToImageSpace() is particularly useful, since it creates a correspondence map between the Image space and the Event space. In RV’s terminology, Image space is normalized to the Source Image geometry - the drawing canvas - while Event Space is the entire RV window. The artist draws "on the view" and not "on the image", which is subject to Zoom or Pan. Eventually, the ROI selected by the artist for re-rendering in RV is drawn in Eventual coordinates and will be in Event Space. These coordinates are then converted into Image Space so that any potential Zoom or Pan is taken into account. Note that the Image space coordinates are normalized: by default, they go from 0-1 in Y and from 0- in X.

Clone this wiki locally