No movies, nor animation: This is an static image compositing
tool (although its internal code has been developed thinking in a future
support for movie compositing).
Text layers are single-line only. Multiple lines and paragraph formatting
is not supported yet. If you need multiline text in your compositions,
you must render it with an external application, and import it as an image
file in Pegamento.
The exported images from the final rendering are always RGB. So, for the
moment it's not possible to generate an alpha channel in exported images.
Also, grayscale rendering is not supported (even if your layers are grayscale,
the final rendering will always be RGB).
If you want to use a grayscale image as a matte, the image must be either
a 2-channel grayscale-alpha image or a 4-channel RGBA image. Otherwise,
if it's a 1-channel grayscale, the image will be inserted as a fully opaque
quadrilateral containing the grayscale image (the luminance channel will
get the image in the file, but the alpha channel would be fully opaque),
and so you won't be able to get the matte you wish.
MIPmaps are generated in software, rather than in hardware.
Lack of "document concept". You work with "Load", "Merge", and "Write"
concepts instead of "Open", "Save" and "Close" concepts. So, there's no
"current document", and the current data in the application is not associated
with any file. In short: you don't open nor close a document.
DXFs newer than release 12 are not supported. However, most CAD applications
allow you to save in release 12 format.
The DXF 'Text' entity is not supported, so, if you want DXF text, it must
be defined with other entity types.
DXF line polylines are always understood as being made of straight line
segments, even if they have arcs or splines. All polyline vertices are
processed as fully sharp vertices, without any kind of curvature not smoothing.
DXF geometry enclosed inside block definitions is not optimized: it's drawn
as is, without grouping entities into line strips, loops nor triangle strips.
DXF blocks definitions are stored in memory as many times as they're inserted,
so the current version doesn't benefit from the "smart memory usage" that
a block-based drawing encourages. The memory used by a block-based drawing
will be the same as if you decompose the blocks.