Intel® IPP 2018 Update 3 is now available. Intel(R) IPP 2018 Update 3 release improved LZ4 compression and decompression performance on data with high entropy.
Check Intel® IPP 2018 Update 3 release notes to learn more information.
What's New in Intel® IPP 2018 Update 3:
What's New in Intel® IPP 2018 Update 2:
Image processing: ippiSeparableFilter, ippiNorm, ipprFilterBorder_L
Signal processing: ippsAdd_32u
What's New in Intel® IPP 2018 Update 1:
Added new Platform-Aware APIs ipprFilterBorder for 3D data filtering, and ipprCopyBorder for 3D border data copying.
Extended optimization for Intel® AVX-512 instruction set in the image and signal processing domain.
Extended optimization for Intel® SSE4.2 instruction set in image processing and signal processing domain
What's New in Intel® IPP 2018:
Added new functions to support the LZ4 data compression and decompression.This release also introduces the patch files for LZ4 source to provide drop-in optimization with the Intel® IPP functions.
Introduced the standalone cryptography packages. The cryptography functions no longer depend on the main Intel® IPP packages, and can be used without the main Intel® IPP packages.
Introduced the optimization code for the GraphicsMagick source.The code can provide drop-in optimization on GraphicsMagick with the Intel® IPP functions:
The code supports GraphicsMagick version 1.3.25, and provides optimization for the following GraphicsMagick APIs: ResizeImage, ScaleImage, GaussianBlurImage, FlipImage, and FlopImage.
The optimization code can improve the APIs performance by up to 4x, depending on the functionality, input parameters, and processors.
Made the Integration Wrappers APIs part of the Intel® IPP packages.
Added the 64-bit data length support to Canny edge detection functions (ippiCanny_32f8u_C1R_L).
Added the ippiDemosaicVNG functions that support the demosaicing algorithm with VNG interpolation.
Added the Elliptic Curves key generation and Elliptic Curves based Diffie-Hellman shared secret functionality.
Added the Elliptic Curves sign generation and verification functionalities for the DSA, NR, and SM2 algorithms.
Extended optimization for the Intel® Advanced Vector Extensions 512 (Intel® AVX-512) and Intel® Advanced Vector Extensions 2 (Intel® AVX2) instruction sets.
Improved performance of LZO data compression functions on Intel® AVX2 and Intel® Streaming SIMD Extensions 4.2 (Intel® SSE4.2).
Removed support for Intel® Pentium® III processor. The minimal supported instruction set is Intel® Streaming SIMD Extensions 2 (Intel® SSE2).
Removed support for the Intel® Xeon Phi™ x100 product family coprocessor (formerly code name Knights Corner) in this release: The Intel® Xeon Phi™ x100 product family coprocessor (formerly code named Knights Corner) was officially announced end of life in January 2017. As part of the end of life process, the support for this family will only be available in the Intel® Parallel Studio XE 2017 version. Intel® Parallel Studio XE 2017 will be supported for a period of 3 years ending in January 2020 for the Intel® Xeon Phi™ x100 product family. Support will be provided for those customers with active support.
Threading Notes: To support the internal threading in the Intel® IPP functions, Intel® IPP provides the Threading Layer APIs in the platform-aware functions. These APIs can support both 64-bit object sizes (for large size images and signal data) and internal threading in Intel® IPP functions. Check the “Threading Layer Functions” pat in the Intel® IPP Developer Reference to get more information on these APIs. Your feedback on extending the new threading functions is welcome. The legacy Intel IPP threaded libraries are available by the custom installation, and the code written with these libraries will still work as before. However, the threaded library will not expand its threading functions, and the new threading will be developed only in the new Intel® IPP threading layer APIs. User’s application is recommended to use the new Intel® IPP threading layer APIs or implement the external threading in their applications.
We need many guides and examples of how utilize the function in Multi Threaded scenarios.
The forum Cookies life span is really short (If you don't visit for 3 days you need to login again).
Could you please extend that to a month or something?
There is no reason for this short life span of the cookie.
The data isn't that sensitive.
It seems 2018 Update 1 is released.
Could you reconsider and bring back the Matrix Library (Optimized for small matrices)?
It was one of the most useful features of IPP.
Have you got chance to check Intel MKL libraries, which is focusing on mathematical computation? Matrix functions are now provided in that libraries now. In the recent MKL releases, it also improves the performance on small matrix computation (for example, by direct_call APIs).
I haven't checked MKL specifically.
We would be happy to integrate only one library (Even one is too much trouble).
I really think a library optimized for small matrices should be included in IPP.
Even you won't optimize it to AVX 512 (Which is by itself not showing any improvement on CPU's).
It depends on domain: in ipps they are threaded only for cpus that have 2 cores on die (not more!) and shared cache. In ippi - 2d FFT is threaded for images > 256x256.
I cannot download 2018 Update 3.
Click "w_ipp_2018.3.210_online.exe" to download on registrationcenter.intel.com page, occur page not found.
w_ipp_2018.3.210.exe file is too.