GetHDL documentation bug

The documentation for mfxFrameAllocator::GetHDL states:

This function returns the OS-specific handle associated with a video frame. If the handle is a COM interface, the reference counter must increase. The SDK will release the interface afterward.

but all d3d samples simply return the handle without incrementing the ref count. I think the docs are wrong.


Can't undersatnd Intel vTune Amplifier XE 2016 license method

Hello, I'm Saar and I'm trying hard to find the way to activate Intel vTune Amplifier XE 2016 with a leagal registered serial number.

After reading those pages:



Target usage


What exactly does the target usage setting do? I'm noticing that if I use "BEST_QUALITY" with a bitrate of 4mbit, sometimes the encoder is not fast enough to keep up with a live stream (24 fps).

However, when I am using "BALANCED" with a bitrate of 50mbit, there is no such speed issue.

I'd like to know what exactly the targetusage setting is doing, since it is also not documented anywhere.


I'm using a

Intel(R) Core(TM) i7-4700EQ CPU @ 2.40GHz

if that matters.

VPP - Denoising is not working as expected

I run the VPP code in sample_vpp (sdk sample downloaded from https://software.intel.com/sites/default/files/MediaSamples_Windows_6.0.0.68.msi) by passing a noisy input nv12 for removing noise. The below is the command line I used:

sample_vpp.exe -sw 300 -sh 300 -scc nv12 -dw 300 -dh 300 -dcc nv12 -denoise 99 -i input.yuv -o denoised.yuv

The resultant output still has the noisy image and is exactly identical to the input image though no errors are reported in the console.

CentOS 7.2 + MLNX_OFED 3.1-1 + MPSS 3.6.1

I'm trying to get mpss running on our centOS 7.2 cluster.

We're using kernel 3.10.0-327.4.4.el7.x86_64.

MLNX_OFED_LINUX-3.1- (OFED-3.1-1.0.3).

On Mellanox Technologies MT27500 Family [ConnectX-3] adapters.

Installation of both MOFED and mpss run flawlessly, except when I try to use them together.

So I can install mpss. Setup ethernet networking. Ssh from/to xeon phi. Run code on xeon phi, all without a problem problem.

I can install Mellanox ofed, use infiniband (ON THE HOST),(ibv_*_pingpong ) without problems. 

simple_decode_vmem - GPU to CPU memory copy - ways to optimize

Hello All,

I profiled and found that copying memory from GPU to CPU is very expensive. I am looking for your inputs to alleviate this performance loss.

While the decoder using video memory gave a performance of 1653 FPS (here the output is not dealt after decoding), copying the decoder's output to system memory after decoding gave just 80FPS (in the simple_decode_vmem application). Such a fall it is, which leaves us with just 2 decodes per processor. I used MFX_GPUCOPY_ON but not to avail any performance benefits.

how to configure intel media sdk in android

posted for 

deepak k.



how to configure intel media sdk in android


Hi sir very good Morning to you i have gone throughout the stack overflow that you are providing an api for video editing ,cutting etc. i have visited to your website but unable to find proper solution to configure intel media sdk in android studio. I have download the source code from Githup but while i open that code there are many errors due to jar files are missing even i am not able to find the jar file also. Kindly help me for both of these. 

Poor speed in MIC

Dear All:

As learning purpose, i tried to code a program which find total number prime number for a given range. isprime function finds  if a number is prime or not. I added !$omp declare simd to vectorize that function. I do not know why, program perform three times slower in intel phi than host.


Host: 16 sec

MIC: 43 sec

Suscribirse a Optimización