Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Intel Community
- Software Development SDKs and Libraries
- Intel® Integrated Performance Primitives
- Is there an IPP function for this 3-rd interpolation function?

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

Highlighted
##

This is a standard 3rd order itnerpolation:

missing__zlw

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

02-11-2011
09:05 AM

8 Views

Is there an IPP function for this 3-rd interpolation function?

for (int iy=-2; iy<=3; iy++) {

y_coeff = interp_coef_y[iy+2];

x_interp_value = coefx0*pf[index_signal_start + 0] +

coefx1*pf[index_signal_start+1] +

coefx2*pf[index_signal_start+2] +

coefx3*pf[index_signal_start+3] +

coefx4*pf[index_signal_start+4] +

coefx5*pf[index_signal_start+5];

signal_value3 += y_coeff*x_interp_value;

index_signal_start += nx;

}

Is there an IPP function for this?

Thanks.

1 Reply

Highlighted
##

Your "x_interp_value = ..." statement should be handled nicely by the DotProduct function:

http://software.intel.com/sites/products/documentation/hpc/composerxe/en-us/ippxe/ipp_manual_lnx/hh_...

since you are essentially just multiplying two vectors together. One vector is contained in pf[] and the other is your list of coefficients (coefx0-5). With such a small vector you may not see much benefit from using an IPP function, due to the call overhead; but if the vectors are longer or you are doing large numbers of them, there may be an advantage.

PaulF_IntelCorp

Employee

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

02-11-2011
04:27 PM

8 Views

http://software.intel.com/sites/products/documentation/hpc/composerxe/en-us/ippxe/ipp_manual_lnx/hh_...

since you are essentially just multiplying two vectors together. One vector is contained in pf[] and the other is your list of coefficients (coefx0-5). With such a small vector you may not see much benefit from using an IPP function, due to the call overhead; but if the vectors are longer or you are doing large numbers of them, there may be an advantage.

For more complete information about compiler optimizations, see our Optimization Notice.