Macro-blocks artefacts when enconding H264 stream

[Daniel_E-B]

Hi,

I'm using the Media SDK to encode an H.264 stream and when i play the encoded file I see some unusual macroblocks artefacts in certain frames.

This only happens when I encode with a large number of B-frames compaired to the GOP size.

I've attached a GOP presenting the problem (starting from frame 11 in stream order ) as well as the logs using the media SDK tracer.

Thanks!

[Petter_L_Intel]

Hi Daniel,

can you please also share the details about your system configuration, such as your graphics driver version and type of Processor.

Regards,
Petter

[Daniel_E-B]

Hi Petter,

Here is the log from sys_analyzer containing the informations you requested.

Graphics Devices:
        Name                                         Version             State
        Intel(R) HD Graphics 4600                    9.18.10.3165        Active

System info:
        CPU:    Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
        OS:     Microsoft Windows 7 Enterprise
        Arch:   64-bit

Thanks,

[Petter_L_Intel]

Hi Daniel,

Thanks for sharing your system configuration.

Could you please try using the latest available graphics driver from here: https://downloadcenter.intel.com/default.aspx   . The most recent driver for your system has the version number 3345.  Let us know if the you see the same issues using this driver?

In the meantime I will also ask codec developer about the macroblock behavior around frame 11 to see if this is expected with that configuration.

Regards,
Petter

[Daniel_E-B]

Hi Petter,

I updated my drivers to the latest version and it seems to the fix my original bug but now i have another issue.

With the 3345 drivers update i have an unusual hi CPU usage, even when i don't do anything on the machine. I have an 8 cores system and one of those cores goes up to more than 80% usage, by doing nothing!

After investigation i found out that it was due to the latest update. With the 3257 drivers i don't have this problem.

Any idea on what could be the cause? Performance is essential for us, which makes the behavior of the 3345 version kind of disturbing.

Thanks!

Daniel

[Petter_L_Intel]

Note: Related to the following forum post: http://software.intel.com/comment/1775146#comment-1775146