1,433
edits
Line 434: | Line 434: | ||
=== <span style="color:#0931C6">Memory bandwidth consumption</span><br> === | === <span style="color:#0931C6">Memory bandwidth consumption</span><br> === | ||
'''''Test pipeline ( | '''''Test pipeline (ducatimpeg4dec):''''' | ||
<pre style="background:#d6e4f1"> | <pre style="background:#d6e4f1"> | ||
gst-launch-1.0 filesrc location=/am5728-gst-tests/video-samples/Wreck- | gst-launch-1.0 filesrc location=/am5728-gst-tests/video-samples/Wreck-It_Ralph_MPEG4.mp4 ! qtdemux name=demux demux.video_0 ! queue ! mpeg4videoparse ! ducatimpeg4dec ! fakesink sync=true -e | ||
</pre> | </pre> | ||
'''''Test pipeline ( | '''''Test pipeline (avdec_mpeg4):''''' | ||
<pre style="background:#d6e4f1"> | <pre style="background:#d6e4f1"> | ||
gst-launch-1.0 filesrc location=/am5728-gst-tests/video-samples/Wreck- | gst-launch-1.0 filesrc location=/am5728-gst-tests/video-samples/Wreck-It_Ralph_MPEG4.mp4 ! qtdemux name=demux demux.video_0 ! queue ! mpeg4videoparse ! avdec_mpeg4 ! fakesink sync=true -e | ||
</pre> | </pre> | ||
Line 447: | Line 447: | ||
'''''Memory bandwidth consumption by memory readings obtained results:''''' | '''''Memory bandwidth consumption by memory readings obtained results:''''' | ||
[[Image:AM572x-testbench- | [[Image:AM572x-testbench-MPEG4-dec-readsbandwidth.png|center|700px|AM572x-testbench-MPEG4-dec-readbandwidth.png]]<br> | ||
In the chart above, it can be seen that when using hardware acceleration, | In the chart above, it can be seen that when using hardware acceleration, less memory bandwidth consumption by memory readings is obtained. The average difference is 185.2 MB/s for sequential reads and 214.3 MB/s for aleatory reads. | ||
'''''Memory bandwidth consumption by memory writings obtained results:''''' | '''''Memory bandwidth consumption by memory writings obtained results:''''' | ||
[[Image:AM572x-testbench- | [[Image:AM572x-testbench-MPEG4-dec-writebandwidth.png|center|700px|AM572x-testbench-MPEG4-dec-writebandwidth.png]]<br> | ||
In the chart above, it can be seen that when using hardware acceleration, less memory bandwidth consumption by memory writings is obtained. The average difference is | In the chart above, it can be seen that when using hardware acceleration, less memory bandwidth consumption by memory writings is obtained. The average difference is 1313.2 MB/s for sequential writes and 157.6 MB/s for aleatory writes. |
edits