Hello everybody,<br><br>I try to undersand why the CRF+VBV is preferable to the CBR mode for live streaming applications.<br><br>I read in the post <a href="http://mailman.videolan.org/pipermail/x264-devel/2010-February/006945.html">"Making sense out of x264 rate control methods"</a>, that CRF+VBV performs better quality than CBR with the following configuration : <br>
<pre style="font-family:arial,helvetica,sans-serif"><font>CBR : x264 --vbv-bufsize 2000 --vbv-maxrate 1000 --bitrate 1000 -o <output> <input><br>--> here we constraint the bitrate and the quality fluctuates with the frame complexity<br>
<br>CRF+VBV : x264 --vbv-bufsize 2000 --vbv-maxrate 1000 --crf 18 -o <output> <input><br>--> here we constraint the quality and the maximum bitrate, and the final bitrate of the sequence depends on its complexity but can be superior to the </font><font>vbv-maxrate.</font>
</pre>Here, the both modes have the same vbv-maxrate, so the CRF+VBV mode cannot allocate more bits than the CBR mode to a frame(?)<span style="background-color:rgb(234,234,234)" id="src0"><br></span>So if we compare one frame encoded with the both modes, the quality cannot be better with the CRF+VBV mode.<br>
<span style="background-color:rgb(234,234,234)" id="src0">So the CRF+VBV mode is preferable</span> only because the quality is constant over all the frames of the sequence (No flickering effect).<br><br>Are My assumptions true?<br>
<br>Thanks for your help<br><br>Eloļse<br>