I am the developer that was working on this, so hopefully i can help. The config file contains all of the information for the tile assembler works off of. An example one is here with some extra info.
TileCount: the total number of tiles for each image. Required
ImageWidth: The width of the final image
ImageHeight: The height of the final image
if dimensions are not given and the tiles are not cropped then the size of the first image will be used to determine output image size
TilesCropped: are the tiles precropped
DistanceAsPixels: are the distances given in pixels or in 0-1.0 range. Defaults to True(distance is in pixels)
BackgroundSource: an image to be used as a background source which can be written over (useful to modify just a part of an image) if not given then defaults to black image, also will use this images dimensions if an image is given
Tile#X:The X coordinate of the left side of the image
Tile#Y:The Y coordinate of the Bottom of the image
Tile#Height: the Height of the image only necessary if images are not cropped
Tile#Width: the width of the image only necessary if images are not cropped
if single file:
ImageFileName: output filename if you are doing a single file
Tile#Filename: the name of the tile file # going from 0 to TileCount-1
if a series
ImageFolder: If the Assembler is given a folder it will assemble all files with a given prefix
Tile#Prefix=The prefix of the file to be used as the tile
ImagePadding=the total amount of padding after the prefix before the extension
ImageExtension= the extension of the image files for a series
file names in series will be everything after the prefix of the file name so _tile_1x1_2x2_Test.png will have a final image name of test.png
Hopefully this answers your questions. If anything is unclear/ you still have questions feel free to ask,
Grant Bartel
TileCount=4
#number of tiles
ImageFileName=PATH\default\images\CompletedImage.png
#output Image filename if not rendering using folder
#ImageFolder=PATH\default\images
#if a folder is given the assembler will go through the folder and find all files with the prefixes specified in tile prefixes with ImagePadding worth of characters after
#and Image extension it will then assemble all like named tiles eg tile_1x1_2x2_circle with tile_1x2_2x2_circle
ImageWidth=640
ImageHeight=480
#end image dimensions if not given will go based off first images size
TilesCropped=True
#if the tiles are cropped
Tile0FileName=PATH\_tile_1x2_2x2_Pyramid.png
Tile0X=0
Tile0Y=0
Tile0Height=240
Tile0Width=320
Tile1FileName=PATH\_tile_2x2_2x2_Pyramid.png
Tile1X=320
Tile1Y=0
Tile1Height=240
Tile1Width=320
Tile2FileName=PATH\_tile_1x1_2x2_Pyramid.png
Tile2X=0
Tile2Y=240
Tile2Height=240
Tile2Width=320
Tile3FileName=PATH\_tile_2x1_2x2_Pyramid.png
Tile3X=320
Tile3Y=240
Tile3Height=240
Tile3Width=320
Tile0Prefix=_tile_1x2_2x2_
Tile1Prefix=_tile_2x2_2x2_
Tile2Prefix=_tile_1x1_2x2_
Tile3Prefix=_tile_2x1_2x2_
ImagePadding=8
ImageExtension=.png
BackgroundSource=C:\Users\Grant\Documents\maya\projects\default\CompletedImage2.png
#The background image that will be overwritten if one is specified
This is working in both releases, also if you want to take a look at SubmitMayaToDeadline.py their is an example of how we do our submission around line 2205.
I have another question about file formats supported with tile rendering, specifically with Modo, but I presume this is restrictive for the format and not the render engine.
My list for supported tiling file formats:-
[code]DO SUPPORT TILING
Targa
Windows
JPEG
PNG
PNG 16 bit
SGI RGB
TIF 16 bit
TIF 16 bit Uncompressed
OpenEXR 16 bit
OpenEXR 16 bit tiled
OpenEXR 32 bit tiled
Layered OpenEXR 32bit
DO NOT SUPPORT TILING
Flexible Precision FLX
Compuserve GIF
High Dynamic Range HDR
JPEG 2000
JPEG 2000 16 bit
JPEG 2000 16 bit Lossless
PSD
TIF
LayeredPSD
Multi-file PNG Layers
Multi-file PNG16 bit Layers
Layered OpenEXR
OpenEXR 32 bit
[/code]
Is there any reason only some formats are supported? Is it a restriction of the format or Deadline and the assembly methods?
I really need Multi-file PNG Layers to be rendered with tiles and currently have to render a whole frame per slave for this format.
There seems to be only .mel version in the submission Maya Main folder. I’m checking the SubmitMayaVRayDBRToDeadline.py version instead as I can’t understand mel script or how to apply it as Python.
And sorry you are correct about the .mel i misread that, another example that is actually a .py file is SoftimageToDeadlineFunctions around line 860 which constructs the file for softimage it should be similar again if you have any questions feel free to ask.
Does Draft support Multi layered png files or just png files? Is your list of formats for 8 bit, 16 bit & 32 bit versions?
I see that it supports .hdr which the tile assembler does not, so I’m going to have use both in a custom setup I think, at the job submission stage based on file format required.
Any particular reason Draft is restricted to certain file types? Is Draft a Thinkbox creation?
Draft is a Thinkbox product a lot of information on what can be done with it can be found at http://www.thinkboxsoftware.com/draft/. For your other questions I will have to get someone who worked on draft to answer since I am not sure myself.
I’m not familiar with multi-layer PNG files. In general, Draft does not support layers, so I would assume that they are not supported. If you’d like to send us an example file, we could test it.
Draft can normally read any bit depth.
Its output bit depth currently depends on the image file format, for example, it always writes 32-bit floating point EXR files.
Sorry, I’m not quite sure what you mean. Grant listed the common file formats supported by Draft, but Draft supports other less common file formats too. I’d encourage you to try it with whatever image format you use, and please let us know if it doesn’t work.