Optimierung

HEVC decoder plugin blocks AVC decoder

I'm using Intel® Media Server Studio 2015 R4. My application makes usage of the media sdk library (libmfxsw64.dll) for decoding AVC video.

As soon as I add the HEVC decoder plugin folder to my application (15dd936825ad475ea34e35f3f54217a6 plus content), the initialization of AVC decoding fails in MFXVideoDECODE_Init with error MFX_ERR_INVALID_VIDEO_PARAM.

I even do NOT load the plugin with MFXVideoUSER_Load (.. MFX_PLUGINID_HEVCD_SW ..), so I wonder what's the influence of the plugin folder here?

HEVC Decoder Multicore Performance

I'm running Intel® Media Server Studio 2015 R4 and use the HEVC decoder plugin.

As far as I understood, it will work CPU only. So I'd expect near 100% CPU usage, but I see only 50% and running two cores only.

I already tried to adjust mfxVideoParam.AsyncDepth and mfxInfoMFX.NumThread, but this does not make a difference.

Am I missing anything?

Sample Encode

Hello,

I am trying to use sample_encode to encode raw YCbCr frames captured from a webcam using EmguCV in a C# project. I understand that sample_encode expects a YUV/NV12 file as input in order to encode. Is there any way to modify the sample_encode in order to work with individual frames rather than an entire file, as I would like to be able to compress a live feed. 

The problem of " libiomp5md.dll!__kmp_suspend_64() " with Intel C++ Compiler XE 15.0 Update 4

After running  several hours, the code with OpenMP or MKL would lock at

         libiomp5md.dll!__kmp_suspend_64()
         libiomp5md.dll!__kmp_barrier()
         libiomp5md.dll!__kmp_join_barrier()
         libiomp5md.dll!__kmp_join_call()
         libiomp5md.dll!_vcomp_fork()

The code example with OpenMP could be:

for (long long n = 0; n < N; n++)  // N=2^50
{
#pragma omp parallel for
        for (int i = 0; i < Z; i++)
        {
                y[i] = a[i] * b[i];
        }
}

Xeon Phi - Error in installation

When installing the mpss-3.5.1 on CentOS 7 (kernel 3.10.0-229.el7.x86_64), I got an error when issue " micctrl --initdefaults" as below:

[Error] micrasrelmond: [Parse] START_WITH_SECURITY=false invalid config parameter

[Warning] micrasrelmond: Generating compatibility network config file /opt/intel/mic/filesystem/micrasrelmond/etc/sysconfig/network/ifcfg-mic0 for IDB.

clang-based frontend for linux?

Could you guys make your clang-based frontend (http://llvm.org/devmtg/2014-04/PDFs/Posters/ClangIntel.pdf) available for linux also?

Your EDG frontend has a lot of bugs with c++11/14 code and it would be much easier to use clang, which has better c++11/14 support.  I can't imagine it would be much harder than supporting OS X.

how to structure a flow graph with a blocking input source

What is a good way to modify Michael Voss' Feature Detection flow graph example when the source filter providing input images is blocking waiting for another image? This is a required modification if one wants to implement this graph for a real-time input source like a video camera.

Optimierung abonnieren