On Tue, Mar 09, 2021 at 11:07:21AM +0100, Ulf Zibis wrote: > > Am 09.03.21 um 10:43 schrieb Ulf Zibis: > > > > Am 09.03.21 um 09:57 schrieb Peter White: > > > The frame count *is* exact. The way to get there is just not that > > > pretty. If anybody on this list knows a better one, step up, please. > > > > I suspect that the last "progression" line shows the exact count, as it may > > only report the last progression state a few µs before the operation is > > completed. > > Can one please clarify this?
I don't know if I understood your request correctly but here goes: ffmpeg -i example.mkv -an -sn -c copy -f null - ... Stream mapping: Stream #0:0 -> #0:0 (copy) Press [q] to stop, [?] for help frame=31186 fps=0.0 q=-1.0 Lsize=N/A time=00:21:40.59 bitrate=N/A speed=2.67e+03x video:162174kB audio:0kB subtitle:0kB other streams:0kB global headers:2kB muxing overhead: unknown As you can see, in the second to last line in the first field is the frame count. Since the copy codec gets used, I can see no reason why this should not always show the exact total frame count of any video. Also, note that I used a matroska file as an example, since it is one of the mentioned special cases mentioned here: https://stackoverflow.com/a/28376817 I had the same question you posted once and think I saw that one or a similar answer. After some ado I came to the conclusion that the aforementioned way is the most robust and universal one. But if I am wrong, I would very much like to be corrected. Peter _______________________________________________ ffmpeg-user mailing list [email protected] https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
