Performance loss between IPP3.0 and IPP6.1

[rohwedder]

Hi,

we had the need to switch to WindowsXP-64Bit because our software needs too much memory for a WindowsXP-32Bit OS (huge images for computer vision). On Windows-32bit we were using IPP3.0. Therefore we had to upgrade because IPP3.0 is not available for 64Bit, so now we use IPP6.1. Additionally we were hoping for a small performance boost as side effect of the new optimizations (e.g. multithreading, since our system is a dual core system) in the new version, but in the contrary! We encountered cases in which the old IPP3.0 was faster than the new one and this effect got even worse when we used ippSetNumThreads(1) to limit computation to only one thread just as IPP3.0 already did. IPP6.1 was more than two times slower than the old IPP3.0.
Example:
ippiFilterMedian_16s_C1R(...) with the following parameters
srcStep=4090
dstStep=4050
dstRoiSize={ width=2025 height=6337 }
maskSize={ width=21 height=1 }
anchor={ x=0 y=0 }
Size of source image={ width=2045 height=6337 }
Image Content=Random noise created by a randomizer

Our test computer:
Intel Core 2 Duo CPU (E6750 @ 2,66GHz) with 8GB of RAM on Windows XP 64Bit

Does anyone have an idea why the older version is slower and is there a solution for our problem?

Thank you,

Rohwedder AG

[Vladimir_Dudnik]

Hello,

the reason of performance difference for ippiFilterMedian function between IPP 3.0 and IPP 6.1 is different algorithm we start to use from IPP 5.0 time frame. The new algorithm provides better performance for 2D filter kernels. We did not took into account that someone will use 1D kernels for 2D functions. Basically we recommend to use ippsFilterMedian for 1D kernels. If there is no Ipp16u data type support let's fill feature request for future IPP versions.

Regards,
Vladimir

[Chao_Y_Intel]

A feature request tracked in the engineering databasefor ippsFilterMedian_16u function.

Thanks,
Chao