====== ffmpeg - Hardware Acceleration using GPU - Benchmark ======

===== Null muxer =====

A simple way to benchmark decoding is to use the null muxer:

<code bash>
ffmpeg -i input -f null -
</code>

<WRAP info>
**NOTE:**  The following parts of the command forces a null file: <code bash>
-f null -
</code>

</WRAP>

----

===== Timing decoding =====

Linux and macOS users can add the time command:

<code bash>
time ffmpeg -i input -f null -
</code>

returns:

<code bash>
[...]
real	0m3.115s
user	0m0.223s
sys	0m0.244s
</code>

<WRAP info>
**NOTE:**  See **man time** for more info.
</WRAP>

----

===== Benchmark option =====

The **-benchmark** option can be added to output CPU time and maximum memory consumption:

<code bash>
time ffmpeg -i input -benchmark -f null -
</code>

returns:

<code bash>
[...]
bench: utime=0.154s stime=0.196s rtime=3.000s
bench: maxrss=365256kB
</code>

----

===== Choosing streams =====

If you want to just decode a particular stream then use the map option:

<code bash>
ffmpeg -i input -map 0:a:0 -f null -
</code>

<WRAP info>
**NOTE:**  Another approach to ignore streams is to use:

  * **-an** to ignore audio streams.
  * **-vn** to ignore video streams.

</WRAP>

----

===== Threads =====

You can decode with one thread if you want to:

<code bash>
ffmpeg -threads 1 -i input -f null -
</code>

<WRAP info>
**NOTE:**  Not all decoders have threading capabilities and some have several.

You can check decoder details such as <code bash>ffmpeg -h decoder=h264</code>

</WRAP>

----

===== Choosing a decoder =====

There can be several decoders available for a format.

You can name the decoder if you do not want to rely on the default:

<code bash>
ffmpeg -c:v vp8 -i input -f null -
ffmpeg -c:v libvpx -i input -f null -
</code>

----

===== Example =====

<code bash>
ffmpeg -hwaccel vaapi -hwaccel_device /dev/dri/renderD128 -hwaccel_output_format vaapi -i input.mp4 -c:v hevc_vaapi -q:v 28 -f null - -benchmark
</code>

returns:

<code bash>
...
bench: utime=0.156s stime=0.195s rtime=3.003s
bench: maxrss=358944kB
</code>

<WRAP info>
**NOTE:**  **maxrss** indicates the maximum RAM used during the ffmpeg execution.

  * **utime**: user time.
    * utime is the sum of processing time across all threads. Since they may operate in parallel, utime can exceed rtime.
  * **stime**: system time.
  * **rtime**: real time.

</WRAP>