Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
FPGA community forums and blogs have moved to the Altera Community. Existing Intel Community members can sign in with their current credentials.
7234 Discussions

Two new functions in V10 Beta.

rbarron
New Contributor I
‎09-24-2007 12:41 PM
581 Views

Release notes and example code for MKL 10.0.006 Beta mention new functions CGEMM3M and ZGEMM3M, but no offer explanation of what they do. Does anybody here know? A link would suffice.

0 Kudos

Link Copied

1 Reply
y_sigal
Beginner
‎09-24-2007 12:53 PM
581 Views
In the help it is written:
"These routines perform the complex multiplication by forming the real and imaginary parts of the input matrices. It allows to use three real matrix multiplications and five real matrix additions, instead of the conventional four real matrix multiplications and two real matrix additions. The use of three real matrix multiplications only gives a 25% reduction of time in matrix operations. This can result in significant savings in computing time for large matrices."
It would be desirable to note, that for IA32 dgemm in Intel MKL it is realized inefficiently (see: http://www.thesa-store.com/products/ ).
0 Kudos
Copy link
Reply

Community support is provided Monday to Friday. Other contact methods are available here.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

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