You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Video buffering currently does not seem to take into consideration which parts of the video will be skipped automatically by sponsorblock. This means that when a segment of the video is skipped due to sponsorblock, the buffer becomes partially or completely empty because part of the buffer was also "skipped over". This leads to a buffering pause immediately after the skip, or to buffer underruns being more likely after a segment is skipped by sponsorblock.
It would instead be ideal, if we already know that playback will skip over a segment of the video, to not buffer the parts of the video that will be skipped automatically, and instead jump ahead and start buffering the part of the video that follows the segment that will be skipped. This would allow, ideally, the buffer to be close to full even after a segment is skipped by sponsorblock, thereby reducing the risk/frequency of buffer underruns.
The text was updated successfully, but these errors were encountered:
Video buffering currently does not seem to take into consideration which parts of the video will be skipped automatically by sponsorblock. This means that when a segment of the video is skipped due to sponsorblock, the buffer becomes partially or completely empty because part of the buffer was also "skipped over". This leads to a buffering pause immediately after the skip, or to buffer underruns being more likely after a segment is skipped by sponsorblock.
It would instead be ideal, if we already know that playback will skip over a segment of the video, to not buffer the parts of the video that will be skipped automatically, and instead jump ahead and start buffering the part of the video that follows the segment that will be skipped. This would allow, ideally, the buffer to be close to full even after a segment is skipped by sponsorblock, thereby reducing the risk/frequency of buffer underruns.
The text was updated successfully, but these errors were encountered: