ippsDecodeAdaptiveVector

[Elyasaf_B]

Hi,

I am using IPP G729 functions in my application The application is developed on Linux 64 bit platform. 

We have observed different results when running the  ippsDecodeAdaptiveVector function between the 32-bit and the 64-bit version of the IPP libraries.
Specifically we are calling:  ippsDecodeAdaptiveVector_G729_32f_I.     

We are currently running IPP 8.2 Update 1, for Linux.

The binary results differ between the 32 and 64 bit libraries.  Is this expected? 

Thanks

[Chao_Y_Intel]

Hello, 

What is the difference of the result?    Here is some comments from the expert on this: 
For FP, we don’t wave bit 2 bit same as the requirement. We used PESQ and SNR on the decoder output to verify the codec.

Some test code shows the difference may also help to check it. 

Thanks,
Chao