A question that puzzles me for quite some time now: What is this 'low-flavor' that is mentioned in the pipeline filters? I cannot find anything on it...
This mean that function takes for example Ipp8u data on input but do calculations and output the result in Ipp16u data type, so input data is "low-flavour" in that case.
I see, although I'm not sure I understand why this is used only for the pipeline filter routines, and in 16s format only. Anyway, the question 'what is it' is answered. I'll think some more about 'why' and 'why only here'...
It was additional to the original design, so was not spreaded for the whole set of functions yet. If you feel it might be useful somewhere else please submit your feature request to Intel Premier Support
