We have observed different results when running the FilterLaplace function between the 32-bit and the 64-bit version of the IPP libraries.
Specifically we are calling: ippiFilterLaplace_8u_AC4R with both 3X3 and 5X5 kernels.
The binary results differ between the 32 and 64 bit libraries. Is this expected? We are currently running IPP 7.0.
an explanation from an IPP engineer is the following
The functions with prefix AC4 processes only three of the four channels (except special cases).
The fourth channel (alpha channel) remains unchanged.
So the results of function execution may be different (if before alpha channels of destination images were different).
It would be much better if you can share your code so we can inverstigate more in detail.