The Bitmap class provides storage of and low-level access to 3ds max bitmaps.
Bitmaps can be temporary objects that are resident only in memory, or they can be associated with a file on disk. Those bitmaps associated with files have a non-null .fileName property and are either load-only or save-only, but not both (this is with respect to file I/O; all bitmaps in memory can have their pixels modified by the setPixel() functions). You make load-only bitmaps with the openBitmap() and selectBitmap() functions. You make temporary or save-only bitmaps with the bitmap() constructor, or with the render(), copy(), or getChannelAsMask() functions. To make them savable, they must have a valid file name, supplied either on the constructor with a fileName: parameter or by setting the .fileName property.
Constructors
bitmap <width <height> [ filename:<filename_string> ] \
[ numframes:<integer> ] \
[ color:<color> ] \
[ gamma:<float> ] \
[ pixelAspect:<float> ]
Creates an empty bitmap. The bitmap value returned is a savable bitmap.
The filename: parameter allows you to set the name of the file the bitmap will be saved to using the save method. Specifying an existing bitmap file name does not load the contents of the bitmap file into the bitmap value.
The numframes: parameter allows you to create a multi-frame bitmap.
The color: parameter allows you to set the initial color of all pixels in the bitmap. For example:
b = bitmap 320 240 color:white
The gamma: parameter allows you to set the gamma for the newly created bitmap. For example:
b = bitmap 320 240 gamma:1.9
render camera:c to:b
The pixelAspect: parameter allows you to set the pixel aspect for the newly created bitmap.
Note that it is not possible to change the gamma or pixel aspect of a bitmap once created, though you can copy one bitmap into a newly created one that has a different gamma or pixel aspect to achieve this effect.
Returns a bitmap value containing the contents of the specified bitmap file. If the specified bitmap file does not exist, a runtime error is generated. The bitmap value returned is a load-only bitmap.
selectBitMap [ caption:<open_dialog_title_string> ]
This method displays a Image Input selection browser for the user to select a bitmap file. Returns a bitmap value containing the contents of the selected bitmap file, or the value undefined if the user presses Cancel in the browser. The bitmap value returned is a load-only bitmap.
Copies an existing bitmap, creating an unique instance of the bitmap. If the source bitmap is load-only, the bitmap value returned is a temporary bitmap value, with a null filename and just one frame. If the source is a multi-frame file, it snaps the current frame into the new copy, which becomes frame 0.
Properties
<bitmap>.filename : String
Lets you get or set the file name associated with the bitmap.
<bitmap>.palette : Array
Yields an array of 256 colors (a bug in the current 3ds max SDK prevents working with paletted files of other than 256 entries). Take care accessing this array with colors indexes retrieved by the getIndexedPixels() function described below, because these indexes are 0-based and MAXScript array indexes are 1-based. Yields undefined if the bitmap is not paletted. Currently you cannot create a paletted bitmap in MAXScript, however you can read and write to a pre-existing paletted bitmap.
<bitmap>.frame : Integer
Get and set the current frame number in a multi-frame file. Setting this property is permitted only on load-only bitmaps. The frame number is 0-based for inherently multi-frame files, such as .avi and .mov.
<bitmap>.numframes : Integer
The number of frames in a multi-frame file, yields 1 for a single image file.
<bitmap>.height : Integer, read-only
Returns bitmap's height in pixels.
<bitmap>.width : Integer, read-only
Returns bitmap's width in pixels.
<bitmap>.gamma : Float, read-only
Returns bitmap's gamma value.
<bitmap>.aspect : Float, read-only
Returns the bitmap's pixel aspect. Note that this is the aspect of the pixel, not the aspect of the bitmap image.
Methods
Opens a virtual frame buffer (VFB) displaying the image. Changes to the bitmap are not automatically displayed to the VFB. To update the VFB, you need to call this function again. Setting the frame property does cause the VFB to update. Each bitmap has its own VFB, i.e., if you display two different bitmaps, two VFBs will be displayed.
Close the VFB associated with the bitmap if open.
save <bitmap> [frame:<integer>]
Saves current state of the bitmap. Make sure the file name has been set first. When saving multi-frame image files, the effect of the frame: keyword parameter is different depending on the type of output file. Inherently multi-frame files, such as .avi and .mov, ignore the frame: parameter and always write to the next output file frame in sequence on each save(); you cannot randomly write frames or rewrite frames already saved. Saving image sequences (such as .bmp, jpg, .tga) requires supplying the frame: parameter and causes the frame number to be appended to the output file name for each frame saved. This effectively generates a sequence of image files suitable for building an IFL. The output file remains open until you call close on the bitmap.
The save() function can only be called on a savable bitmap. You will get a runtime error if you attempt to save a bitmap opened with openBitmap() or selectBitmap().
Closes the bitmap file associated with the bitmap if it is open for output. A bitmap file is opened for output when you call save on the bitmap value the first time. If a VFB associated with the bitmap is open, it will be closed. This function also frees all memory allocated by the bitmap, including the internal pixel frame buffer. If you are generating many bitmaps, say in a rendering loop, you can use the close() function to free these often large amounts of memory, that won't otherwise be released until the next garbage collection.
copy <source_bitmap> <dest_bitmap>
Copies the source bitmap to the destination bitmap. If the size of the two bitmaps is different, the image will be scaled to fit the destination bitmap.
Position multi-frame file (avi, flc, etc.) to frame frame. This method is permitted only on load-only bitmaps. The frame number is 0-based for inherently multi-frame files, such as .avi and .mov. Updates VFB if open.
getPixels <bitmap> <coord_point2> <num_pixels>
Retrieve a sequence of pixels from a row in the bitmap as an array of color values. The coord_point2 argument specifies the starting pixel coordinate with [0,0] at the top-left-hand pixel. The .x component of the Point2 coordinate is the column, and the .y component is the row. The sequence must come from one row only; you cannot retrieve a sequence of pixels that spans multiple rows. Using any out of-bounds coordinates or retrieving over row boundaries will fail and yield an empty array.
setPixels <bitmap> <coord_point2> <color_value_array>
Sets the sequence of pixels in a row in the bitmap starting at the Point2 coordinate to the values in the given array. The .x component of the Point2 coordinate is the column, and the .y component is the row. The number of pixels to set is taken from the size of the array. As with getPixels(), you cannot set pixel sequences across a row boundary.
getIndexedPixels <bitmap> <coord_point2> <num_pixels>
Retrieve a row of pixels as an array of palette index numbers. The coord_point2 argument specifies the starting pixel coordinate with 0,0 at the top-left-hand pixel. The index numbers are origin 0. Note that this is in keeping with indexed color conventions but not MAXScript's. Take care accessing any retrieved palette array using these numbers because the retrieved palette is returned as a MAXScript array and these arrays use 1-based indexes. You cannot get pixels sequences that cross row boundaries.
setIndexedPixels <bitmap> <coord_point2> <number_array>
Sets the row of pixels starting at the Point2 coordinate to the color indexes in the given array. The number of pixels to set is taken from the size of the array. The color indexes are zero-based. You cannot set pixels sequences that cross row boundaries.
getChannel <bitmap> <coord_point2> <chan_name>
Retrieves the g-buffer channel data for the named channel for the pixel specified by 2D coordinate. The chan_name argument specifies the channel identifier, chosen from the following:
#zDepth
#matID
#objectID
#UVCoords
#normal
#unClamped
#coverage
#node
#shaderColor
#shaderTransparency
#velocity
#weight
If the bitmap contains the requested channel, the returned value is always an array of values, one for each layer present in the channel at that pixel, ordered from front to back. If the bitmap does not contain the requested channel, the function returns the value undefined. Typically, you get bitmaps containing channels by opening .rla files or using the MAXScript render() function with the channels: parameter supplied. You will get multiple values in the array if the front objects have any transparency. For example:
getChannel bm [x,y] #zDepth
might return:
#(-354.413, -354.467, -355.271, -1e+030)
indicating there is a front surface with non-zero transparency at depth -354.413, another at -354.467 and a final one at -355.271. The û1e+30 value represents the background image plane. You could retrieve the #node channel (if it was generated) to get the actual scene objects at those depths, eg:
getChannel bm [x,y] #node
might return:
#($Sphere:Sphere02 @ [-4.7,24.3,0.0], $Sphere:Sphere02 @ [-4.7,24.3,0.0],$Sphere:Sphere01 @ [-8.3,56.2,5.7])
The type of values returned in the array depend of the channel, as follows:
#zDepth : <float>
#matID : <integer>
#objectID : <integer>
#UVCoords : <point2>
#normal : <point3>
#unClamped : <color>
#coverage : <float>
#node : <node>
#shaderColor : <color>
#shaderTransparency : <color>
#velocity : <point2>
#weight : <color>
See the 3ds max Software Development Kit Help file for information on the data stored in the g-buffer channels.
getChannelAsMask <bitmap> <chan_name> [to:<bitmap>] \
[fileName:"filename_string"] [pixelAspect:<float>] \
[gamma:<float>] [layer:<integer>] [invert:<boolean>] \
[node:<node> | objectID:<integer> | matID:<integer>] \
[velocity:#angle | #magnitude] | [angle:<float>]
Builds and returns a separate 8-bit gray-level bitmap, suitable for use as a mask in materials, maps, effects, and so on. This bitmap corresponds roughly to the display you see in the visual frame buffer if you select a g-buffer channel to view. In addition, you can cause this mask to be masked itself to a selected node, object ID or material ID. The parameters are as follows:
<bitmap>
The source bitmap containing the source g-buffer channel data.
<chan_name>
The channel from which to generate a mask. The chan_name values are as described for the getChannel() method. Note that the mask is a single 8 bit value while the channel may be a complex value of some kind. In each case, a mapping to 8 bits is performed. Unless noted differently below, this mapping is the same mapping as in used by 3ds max to generate monochrome images in the Virtual Frame Buffer.
to:<bitmap>
Specifies an existing bitmap to write the mask into. Its width and height must match the source bitmap.
fileName:"filename_string"
Specifies the file name associated with the generated bitmap.
gamma:<float>
Specifies the generated bitmap's gamma value.
pixelAspect:<float>
Specifies the generated bitmap's pixel aspect. Note that this is the aspect of the pixel, not the aspect of the bitmap image.
layer: <integer>
Specifies the channel layer of the source bitmap to extract data from. Defaults to 1.
invert:<boolean>
If true, inverts the generated bitmap. Defaults to false.
node:<node>
objectID:<integer>
matID:<integer>
Specifies a sub-mask. If one of these parameters is supplied, the generated mask is clipped to contain data only in pixels where the given node, objectID or materialID is visible. To use this sub-masking, you must ensure the #node, #objectID or #matID channel is present in the source bitmap. Further, if the #coverage channel is present in the source bitmap, the sub-mask is anti-aliased by the coverage data.
velocity:#angle | #magnitude
angle:<float>
These parameters can only be specified if the selected channel is #velocity. The default velocity parameter value is #magnitude. These parameters control the generated bitmap as follows:
velocity:#angle
Generated bitmap pixel values are proportional to angle of pixel movement direction, with an angle of direction of 0 degrees (horizontal to the right) mapping to a luminance value of 0, 180 degrees mapping to a luminance value of 127, and 360 degrees mapping to a luminance value of 255.
velocity:#magnitude
Generated bitmap pixel values are proportional to velocity magnitude. High velocities are darker than low velocities.
angle:<float>
Generated bitmap pixel values are proportional to movement direction deviation from the given angle, with a luminance value of 255 at the exact angle and falling off to a luminance value of 0 at 10 degrees from the specified angle. The specified angle is relative to 0 being horizontal to the right.
Related Methods
Sets Silent mode on or off. If Silent mode is off, any errors occurring during bitmap input or output will result in an error message dialog being displayed by the 3ds max bitmap I/O routines. If Silent mode is on, error message dialogs are not displayed. Returns a boolean signifying the prior Silent mode state.
The Silent mode state is an internal 3ds max state, and other 3ds max plug-ins can turn this state on or off. It is recommended that if you use this method that you turn the state on or off saving the return value, perform the bitmap input or output, and then restore the state to its original value.
Returns a boolean signifying the Silent mode state.
selectSaveBitMap [ caption:<open_dialog_title_string> ]
Displays a 3ds max bitmap save dialog and returns the fully-specified file path name as a string. Returns undefined if the user cancels out of the bitmap save dialog.
Frees up all the memory used by the image file bitmap caches. This is useful if memory is fragmented with a lot of different bitmaps and you want to have just the ones currently active reloaded.
getBitmapOpenFileName caption:<title> filename:<seed_filename_string>
getBitmapSaveFileName caption:<title> filename:<seed_filename_string>
Displays the bitmap file selection dialog. Both functions return a fully-specified file path name or undefined
if the user cancels out. If an existing file name is selected using the getBitmapSaveFileName() function, a File
Exists/Confirm Overwrite dialog will be displayed. Note: The getBitmapOpenFileName() function does not require
that an existing file name be selected - i.e., a file name is returned if the user types a file name that does not
exist. If necessary, you should check the file name that is returned to ensure the file exists.
Notes
When you open a bitmap file, an entire frame is loaded into memory. When you are finished processing a bitmap, it is a good idea to either call close() on the bitmap or assign the value undefined to the variable storing the bitmap value, and then manually run garbage collection (gc()), so that it's memory can be released. Another technique for conserving memory when using the render() or getChannelAsMask() functions repeatedly is to use the to:<bitmap> keyword argument which allows those functions to overwrite an existing bitmap's pixel image with the new rendering or mask.
There currently is no access to the properties associated with the bitmap output plug-ins. This means that you cannot set the codec used for .avi files, nor the setup parameters for .rla and .tga files. The properties used are the properties that were last used when saving that file type in 3ds max.
Examples
The following script shows how to properly create and save a multi-frame image file.
Script
theTeapot=teapot() -- something to render
animate on at time 10 \ -- set animate and time context
rotate theTeapot 180 z_axis -- rotate the teapot
cam=targetcamera pos:[200,0,100] \ -- camera pointed at teapot
target:theTeapot
renderFrames=#{1,3,5..12} -- specify frames to render
b=bitmap 160 120 filename:"c:\\t.avi" -- create a new bitmap
for i = 1 to renderFrames.count do -- loop though renderFrames
(
if renderFrames[i] then -- if supposed to render frame...
(
at time i -- set time context
render 160 120 camera:cam to:b -- render to bitmap frame
save b -- save each frame as you advance
) -- if you save AFTER the loop,
-- just the last frame is saved.
)
close b -- close the output file. This will
-- also get rid of the reference to
-- the bitmap value and free its
-- memory.
Output
$Teapot:Teapot01 @ [0,0,0] -- result of line 1
OK -- result of lines 2 and 3
$Target_Camera:Camera01 @ [200,0,100] -- result of lines 4 and 5
#{1, 3, 5..12} -- result of line 6
BitMap:c:\t.avi -- result of line 7
OK -- result of for loop, lines 8 to 15
OK -- result of line 16
The following script reads a bitmap file, and outputs a script to Listener that implements a function that will recreate the same bitmap. This function can then be copied into your script. This is useful for building button images in MAXScript, so that you don't need to distribute bitmap files along with your script.
Script
(
b=selectbitmap() -- open image file browser
bname="bitmap_"+(getfilenamefile b.filename) -- build name from filename
w=b.width -- get properties of bitmap
h=b.height
format "----------\nfn load_% = (\n" bname -- start defining function
format "local %=bitmap % %\n" bname w h -- create bitmap in function
-- write out a function that unpacks an integer into a pixel color
format "fn unpack val = for p in val collect (r=p/256^2; g=p/256-r*256; b=mod p 256; color r g b)\n"
for r=0 to h-1 do -- for each row in the bitmap
-- have function write the column of pixels to the bitmap
( format "setpixels % [0,%] (unpack #(" bname r
pixels=getpixels b [0,r] w -- read in the column of pixels
for c=1 to w do -- loop through each pixel
( p=pixels[c] -- get the pixel
-- pack the pixel into an integer and write it out
format "%" (((p.r as integer)*256+(p.g as integer))*256+(p.b as integer))
if c != w then -- if not at end of data
format ", " -- write a comma
else
format "))\n" -- else close out the line
)
)
format "return %\n" bname -- function returns the bitmap
format ")\n----------\n" -- finish off function definition
)
Output
----------
fn load_bitmap_convert = (
local bitmap_convert=bitmap 4 4
fn unpack val = for p in val collect (r=p/256^2; g=p/256-r*256; b=mod p 256; color r g b)
setpixels bitmap_convert [0,0] (unpack #(12632256, 12632256, 12632256, 12632256))
setpixels bitmap_convert [0,1] (unpack #(255, 255, 255, 255))
setpixels bitmap_convert [0,2] (unpack #(255, 255, 255, 255))
setpixels bitmap_convert [0,3] (unpack #(255, 12632256, 12632256, 12632256))
return bitmap_convert
)
----------
Also see the example in Point3 Values for another usage of class Bitmap.