When I use the PDFEncoder I can specify a compression but how do I do this when I am using the PdfTranslator (for creating PDF with OCR'd text) documents. I have a 24bpp image that when saved to a PDF with Jpeg compression results in a 1mb document - the same image that is output via the PdfTranslator (image on text) results in a document that is nearly 6mb, almost the same as saving it to a tiff file using the default compression.

As usual for most developers I have answered my own post by reading the documentation! I guess thats what its for. I am a little puzzled though, because I didn't use the CompressionSelector delegate, the documentation states the translator would use the most appropriate compression but in my case it wasn't - when I explicitly set the compression to jpeg for a 24bpp image the resulting document was 1mb instead of 5mb - I guess it was using the Flate compression type. What I would like to now is the rationale behind this and whether I should be using jpeg based on an image being 24bpp - are there instances where this will not compress that well?

We use flate by default because it is lossless compression. With Jpeg, even on the highest quality setting, some image data is lost.