Fastest way to mulitply 64fc by 64f (complex times scalar)

Fastest way to mulitply 64fc by 64f (complex times scalar)

Assume I have the following:

Ipp64fVector my_all_real_vector
Ipp64fcVector my_complex_vector

and I want to multiply them.

Is it faster to split complex and do 2 real multiplies and rebuild complex, or make real into complex and do complex multiplies??

Strategy 1:

ippsCplxToReal_64fc(my_complex_vector, my_real_parts, my_imag_parts);
ippsMul_64f_I(my_all_real_vector, my_real_parts);
ippsMul_64f_I(my_all_real_vector, my_imag_parts);
ippsRealToCplx_64f(my_real_parts, my_imag_parts, my_complex_vector);

Strategy 2:

ippsRealToCplx_64f(my_all_real_vector, NULL, my_real_but_complex_vec);
ippsMul_64fc_I(my_real_but_complex_vec, my_complex_vector);

I did a quick test to compare and they came out close.  I expected the real multiplies to beat the complex, but maybe the overhead to repack is significant.  What is the recommended best practice in this scenario?  (my vector lengths ~1k to ~26k)

Thanks!

-Brian

 

4 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

It is a shame there is not a 64-bit implementation of ippsMul_32f32fc_I() as that appears to be what you need!

Maybe feature request it from Intel? Alternatively, is there a way you could represent your data with 32-bit precision instead?

I think the appropriate scenario for this case is this:

ippsMul_64f_I(my_all_real_vector, (Ipp64f *)my_complex_vector, length);

ippsMul_64f_I(my_all_real_vector, (Ipp64f *)my_complex_vector + length, length);

Thanks.

I'm sorry.

My scenario is not appropriate.

Perhaps right scenario is such:

ippsRealToCplx_64f(my_all_real_vector, my_all_real_vector,, tmp_complex_vector, length);

ippsMul_64f_I((Ipp64f *)tmp_complex_vector, (Ipp64f *)my_complex_vector, length+length);

I'm sorry once more.

Thanks.

 

Leave a Comment

Please sign in to add a comment. Not a member? Join today