- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
I observed a factor of 2 or 3 performanceLOSS upon transitioning from usual dyanmic linking toinstead IPP static linking with threaded static libswith dispatch (*emerged.lib, *merged_t.lib).As suggested in the manual I addedippStaticInit() to the beginning of the main program but that did not make any difference.
I am using IPP 6.0. Any suggestions on what I may be missing ?
Thanks.
Manoj
I observed a factor of 2 or 3 performanceLOSS upon transitioning from usual dyanmic linking toinstead IPP static linking with threaded static libswith dispatch (*emerged.lib, *merged_t.lib).As suggested in the manual I addedippStaticInit() to the beginning of the main program but that did not make any difference.
I am using IPP 6.0. Any suggestions on what I may be missing ?
Thanks.
Manoj
Link Copied
1 Reply
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Manoj,
Do you experience the same problem with the latest IPP 6.1 version? What is the IPP function or functions affected? How do you measure performance, what is target OS and processor?
Regards,
Vladimir
Do you experience the same problem with the latest IPP 6.1 version? What is the IPP function or functions affected? How do you measure performance, what is target OS and processor?
Regards,
Vladimir
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page