Re: [FFmpeg-user] Fifo muxer with ISMV output results in corrupted video
Hello, It seems that using the argument "-flags +global_header" with the bitstream filter "dump_extra" fixes the issue. Why this isn't needed when directly muxing with ISMV instead of fifo is beyond me. On Tue, 23 Jun 2020 at 12:03, Louis Geuten wrote: > Hello, > > I'm using ffmpeg to transcode vod or live content using h264 through the > ISMV protocol. > I have had a few issues with networking recently, and decided to try to > use the fifo muxer to decouple the output I/O from the encoding. > > However it seems that the video packets produced by the muxer are not > valid. The audio seems fine. > Saving the file on disk allow me to probe it, and it fails with the > following error: > > [extract_extradata @ 0x48c9900] No start code is found. > bonjour/Stream(v-0-1280): Invalid data found when processing input > > I have attached the output logs for both commands. > > The *fifo* command, which produces corrupt video files > *(Job-7-Run-36.err)* > /usr/local/bin/ffmpeg -nostdin -loglevel trace -probesize 20M > -analyzeduration 10M -y -re -stream_loop -1 -i kuro_demo.mp4 -map 0:0 -vf > "scale=1280:720" -pix_fmt yuv420p -preset veryfast -tune zerolatency > -x264opts no-sliced-threads:nal-hrd=cbr:scenecut=0:qpmax=45:keyint=25 > -minrate 2600K -maxrate 2600K -bufsize 5200K -b:v:0 2600K -flags +cgop -g > 25 -keyint_min 25 -r 25.00 -c:v:0 libx264 -an -bsf:v dump_extra -f fifo > -fifo_format ismv -queue_size 200 -drop_pkts_on_overflow 1 > -attempt_recovery 1 -recovery_wait_time 1 -restart_with_keyframe 1 > -format_opts > frag_duration=192:movflags=isml+frag_keyframe:write_prft=wallclock:fflags=+genpts > bonjour/Stream(v-0-1280) -map 0:1 -c:a:0 aac -b:a:0 96K -ac 2 > -metadata:s:a:0 language=und -bsf:v dump_extra -f fifo -fifo_format ismv > -queue_size 200 -drop_pkts_on_overflow 1 -attempt_recovery 1 > -recovery_wait_time 1 -restart_with_keyframe 1 -format_opts > fflags=+genpts:frag_duration=192:movflags=isml+frag_keyframe:write_prft=wallclock > bonjour/Stream(a-0-und) > > The same command without using the fifo muxer: (*Job-7-Run-37.err*) > /usr/local/bin/ffmpeg -nostdin -loglevel trace -probesize 20M > -analyzeduration 10M -y -re -stream_loop -1 -i kuro_demo.mp4 -map 0:0 -vf > "scale=1280:720" -pix_fmt yuv420p -preset veryfast -tune zerolatency > -x264opts no-sliced-threads:nal-hrd=cbr:scenecut=0:qpmax=45:keyint=25 > -minrate 2600K -maxrate 2600K -bufsize 5200K -b:v:0 2600K -flags +cgop -g > 25 -keyint_min 25 -r 25.00 -c:v:0 libx264 -an -f ismv -movflags > isml+frag_keyframe -write_prft wallclock -fflags +genpts -frag_duration > 192 bonjour/Stream(v-0-1280) -map 0:1 -c:a:0 aac -b:a:0 96K -ac 2 > -metadata:s:a:0 language=und -f ismv -fflags +genpts -frag_duration 192 > -movflags isml+frag_keyframe -write_prft wallclock bonjour/Stream(a-0-und) > > I am using the latest git HEAD commit > (44ce333f03e5daf7883147938f74018cec287aec) built from source. > > Is there some option I am missing to make the output ISMV file correct ? > Or is this a bug in the fifo muxer ? > > Thanks for the help. > -- ___ Louis Geuten - Developper Skype: orygin9 *Freecaster - Belgium sprl* mobile: +32 494 08 82 80 15 rue des Phlox - 5100 Nannine - Namur, Belgium VAT: BE0467 231 677 *Freecaster - France sarl* 9 cité du Paradis 75010 Paris, France TVA: FR32 824 896 575 ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fifo muxer with ISMV output results in corrupted video
On 25-06-2020 05:17 pm, Louis Geuten wrote: Hello, It seems that using the argument "-flags +global_header" with the bitstream filter "dump_extra" fixes the issue. Why this isn't needed when directly muxing with ISMV instead of fifo is beyond me. Each muxer has a list of flags; those relevant to encoders are conveyed by ffmpeg to that encoder. The fifo muxer isn't a muxer per-se but an intermediary layer dressed up as a muxer. It won't necessarily share the relevant flags with the child muxer, so they have to be manually set by the user. Gyan ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fifo muxer with ISMV output results in corrupted video
pretty messy command with a few redundancies, but the funniest one -an and -c:a combo Le jeu. 25 juin 2020 à 14:35, Gyan Doshi a écrit : > > > On 25-06-2020 05:17 pm, Louis Geuten wrote: > > Hello, > > > > It seems that using the argument "-flags +global_header" with the > bitstream > > filter "dump_extra" fixes the issue. > > Why this isn't needed when directly muxing with ISMV instead of fifo is > > beyond me. > > Each muxer has a list of flags; those relevant to encoders are conveyed > by ffmpeg to that encoder. > > The fifo muxer isn't a muxer per-se but an intermediary layer dressed up > as a muxer. It won't necessarily share the relevant flags with the child > muxer, so they have to be manually set by the user. > > Gyan > ___ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fifo muxer with ISMV output results in corrupted video
Hello andrei, I know the command is a bit messy, but this is not a contradiction. There are two separate outputs, one for video with the -an, and another with the audio mapping and an audio codec specified. AFAIK they are not used together. These commands are generated "on the fly" from my software, and they can be used quite a lot of use-cases. If you have any other useful critique of the command here, please go ahead and I'll improve it. Regards, Louis On Thu, 25 Jun 2020 at 15:05, andrei ka wrote: > pretty messy command with a few redundancies, but the funniest one -an and > -c:a combo > > Le jeu. 25 juin 2020 à 14:35, Gyan Doshi a écrit : > > > > > > > On 25-06-2020 05:17 pm, Louis Geuten wrote: > > > Hello, > > > > > > It seems that using the argument "-flags +global_header" with the > > bitstream > > > filter "dump_extra" fixes the issue. > > > Why this isn't needed when directly muxing with ISMV instead of fifo is > > > beyond me. > > > > Each muxer has a list of flags; those relevant to encoders are conveyed > > by ffmpeg to that encoder. > > > > The fifo muxer isn't a muxer per-se but an intermediary layer dressed up > > as a muxer. It won't necessarily share the relevant flags with the child > > muxer, so they have to be manually set by the user. > > > > Gyan > > ___ > > ffmpeg-user mailing list > > ffmpeg-user@ffmpeg.org > > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > > > To unsubscribe, visit link above, or email > > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". > ___ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fifo muxer with ISMV output results in corrupted video
Hello, Gyan, I understand the fifo pseudo-muxer doesn't forward everything to the muxer behind it, but shouldn't this flag be set too when not using fifo ? Or is it detected somewhere and added directly ? I had to dive into the source code to find why there wasn't any extradata populated that the muxer could use properly. Anyway thanks for your reply. Louis On Thu, 25 Jun 2020 at 14:35, Gyan Doshi wrote: > > > On 25-06-2020 05:17 pm, Louis Geuten wrote: > > Hello, > > > > It seems that using the argument "-flags +global_header" with the > bitstream > > filter "dump_extra" fixes the issue. > > Why this isn't needed when directly muxing with ISMV instead of fifo is > > beyond me. > > Each muxer has a list of flags; those relevant to encoders are conveyed > by ffmpeg to that encoder. > > The fifo muxer isn't a muxer per-se but an intermediary layer dressed up > as a muxer. It won't necessarily share the relevant flags with the child > muxer, so they have to be manually set by the user. > > Gyan > ___ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] How to merge several segments of a video to the original video?
On 6/25/2020 5:50 AM, mikelee164396 wrote: When I tried to merge, there was an error. See the attachment error.txt for the error content and command line. Please place those in the email body. Very few people will download and open the 7z file. Also, what do the .key and .m3u8 files look like? z! ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] RTMP handshaking behind a proxy
I have given this command another try on another machine today, and read a
bit more about the handshaking process.
It does not seem to open any other port, just exchanging specific packets
on the 1935 port...
So something is fishy on my redsocks configuration I suppose...
Here is part of the strace command:
socket(AF_INET, SOCK_DGRAM|SOCK_CLOEXEC, IPPROTO_IP) = 5
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.194")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(51463),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.146")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(45181),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.205")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(49529),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.119")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(33848),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.25")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(59504),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.28")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(45114),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.211")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(42766),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.169")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(43587),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.236")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(33059),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.9")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(40109),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.95")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(46689),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
connect(5, {sa_family=AF_UNSPEC, sa_data="\0\0\0\0\0\0\0\0\0\0\0\0\0\0"},
16) = 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.16")}, 16) = 0
getsockname(5, {sa_family=AF_INET, sin_port=htons(4),
sin_addr=inet_addr("192.168.0.217")}, [28->16]) = 0
close(5)= 0
socket(AF_INET, SOCK_STREAM|SOCK_CLOEXEC, IPPROTO_TCP) = 5
fcntl(5, F_GETFL) = 0x2 (flags O_RDWR)
fcntl(5, F_SETFL, O_RDWR|O_NONBLOCK)= 0
connect(5, {sa_family=AF_INET, sin_port=htons(1935),
sin_addr=inet_addr("52.223.195.194")}, 16) = -1 EINPROGRESS (Operation now
in progress)
poll([{fd=5, events=POLLOUT}], 1, 100) = 1 ([{fd=5, revents=POLLOUT}])
getsockopt(5, SOL_SOCKET, SO_ERROR, [0], [4]) = 0
write(2, "\33[0;39m[rtmp @ 0x1f512c0] \33[0m", 30[rtmp @ 0x1f512c0] ) = 30
write(2, "\33[0;32mHandshaking...\n\33[0m", 26Handshaking...
) = 26
poll([{fd=5, events=POLLOUT}], 1, 100) = 1 ([{fd=5, revents=POLLOUT}])
sendto(5,
"\3\0\0\0\0\t\0|\2\367xU\36\316\253\216\0366/\7\305\206\212p\262f\324\2
\345\10a"..., 1537, MSG_NOSIGNAL, NULL, 0) = 1537
poll([{fd=5, events=POLLIN}], 1, 100) = 0 (Timeout)
On Tue, Dec 10, 2019 at 6:47 PM Verachten Bruno wrote:
> Hi there,
>
> I unfortunately have to stream from behind a proxy. It works on one of my
> machines thanks to redsocks. The RTMP handshake goes well, and I can stream
> to twitch or youtube.
> On another machine, with the same redsocks configuration, the handshake
> never finishes, be
Re: [FFmpeg-user] FFMPEG endlessly increases memory usage over time with HLS packaging.
Apparently you think that the constantly increasing amount of "Used" memory in my case falls in the case described as: Memory that is: used, but can be made available You’d call it: Free (or Available) Linux calls it: Used (and Available) Let's see what happens if I run 20 processes for few days... Cheers, Alex -Original Message- From: ffmpeg-user On Behalf Of Carl Eugen Hoyos Sent: 25 June 2020 07:28 To: FFmpeg user questions Subject: Re: [FFmpeg-user] FFMPEG endlessly increases memory usage over time with HLS packaging. > Am 25.06.2020 um 02:38 schrieb Alessandro Molon : > > I would like to use ffmpeg as live HLS ABR packager and use /dev/shm to store > playlist and chunks in RAMDISK. > But unfortunately I’ve noticed that ffmpeg constantly increase the memory > used until it saturates it. No. At least both your „free“ and „valgrind“ output indicate the opposite (as does the word „saturate“ above). Please read up about the columns „free“, „used“ and „available“ of free console output. https://www.linuxatemyram.com/ Carl Eugen ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-user] difficulties with time stamps?
For a soft telecined 3/1001 FPS movie. ffprobe reports frames.frame.0.pkt_duration_time="0:00:00.033367" ...1.001/30 frames.frame.1.pkt_duration_time="0:00:00.050044" should be 0:00:00.050050...1.5*1.001/30 The duration of frame 1 is too short by 0.06. Could this error (or errors like this) be the cause of some of the difficulties with time stamps? ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] HEVC acceleration
> You should probably also try with older FFmpeg versions... I tried with an older version of ffmpeg (3.4.7), and it works. The issue is that I don't see any difference when using ffmpeg with and without the "-c:v hevc_v4l2m2m" option. This means that the HEVC accelerator is not enabled. Am I right? Am I missing something in the command line of ffmpeg? Mohammed > > ___ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] difficulties with time stamps?
Am Do., 25. Juni 2020 um 21:15 Uhr schrieb Mark Filipak : > Could this error (or errors like this) be the cause of some of the > difficulties with time stamps? If you experience difficulties, be it with timestamps or otherwise, please provide the (ffmpeg) command line you tested together with the complete, uncut console output and an input sample file. Carl Eugen ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] How to merge several segments of a video to the original video?
Am Do., 25. Juni 2020 um 14:51 Uhr schrieb mikelee164396 : > A video was downloaded in an android app. It is > found that the video is divided into several > segments and a .M3u8 file. Apart from what was said: If the m3u8 segments are transport streams (as I expect), the right tool to "merge" them is cat (or copy /b on windows). Carl Eugen ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] FFMPEG endlessly increases memory usage over time with HLS packaging.
Am Do., 25. Juni 2020 um 20:39 Uhr schrieb Alessandro Molon : > Let's see what happens if I run 20 processes for few days... On most hardware, you cannot run 20 (real-life) FFmpeg processes for a few days but if there is constantly increasing memory footprint or a memory leak, one process should be sufficient to show it. Please find out what top-posting means and avoid it here, Carl Eugen ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] HEVC acceleration
On Thu, 25 Jun 2020 at 15:27, Mohammed Bey Ahmed Khernache < mohbeyi...@gmail.com> wrote: > > You should probably also try with older FFmpeg versions... > I tried with an older version of ffmpeg (3.4.7), and it works. > The issue is that I don't see any difference when using ffmpeg with and > without the "-c:v hevc_v4l2m2m" option. > This means that the HEVC accelerator is not enabled. Am I right? > Am I missing something in the command line of ffmpeg? > > Mohammed > It's probably working :) Please create a ticket on trac.ffmpeg.org . Could you run git bisect to identify which commit caused the problem? Thanks, Andriy ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-user] Producing smooth video from images
Hello, I have a sequence of images, image001.png to image945.png, and I am using this command to produce a video: ffmpeg -framerate 60 -i image%03d.png -c:v libx264 -pix_fmt yuv420p -vf scale=1920:1080,fps=60 -r 60 video.mp4 The video is fine but I would like to improve it. I was thinking in some kind of "motion blur" or a better sense of motion. The filters minterpolate and tblend seem to be what I am looking for, but I don't know how to use nor which is better. Any tips? Kind regards,Leonardo ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Producing smooth video from images
Am Fr., 26. Juni 2020 um 01:11 Uhr schrieb Leonardo via ffmpeg-user : > I have a sequence of images, image001.png to image945.png, > and I am using this command to produce a video: > ffmpeg -framerate 60 -i image%03d.png -c:v libx264 -pix_fmt yuv420p -vf > scale=1920:1080,fps=60 -r 60 video.mp4 Remove "fps=60" and "-r 60", provide the complete, uncut console output and explain what you want to improve about the output file. (The minterpolate filter is powerful but you should first explain why you want to use it) Remember that "-pix_fmt yuv420p" which you probably need for your use-case heavily affects output quality which you cannot circumvent. Carl Eugen ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fwd: OBS and OBS Portable Recording Into Same File Simultaneously
David Bayles writes: > Corrupted video file, FFprobe output text file, and FFmpeg output text file > are all available to download here: > https://drive.google.com/drive/folders/1R0Y5plgkhTWWFNNKgSNJ5yitpKQbIKgg?usp=sharing > > I'm hoping there is a way to extract good video and audio from this video > file, this was for a very important job that is impossible to do again. The > video linked is a 1 minute recreation of the exact same problem... the > actual video file is confidential (for a legal court case) and is much > longer/larger. I used the exact same settings, the only difference was the > window OBS was capturing (I used a YouTube video as opposed to the > videoconference software window used for the deposition). I am no expert, but for the video, I would try using the select filter to capture the I-frames. https://ffmpeg.org/ffmpeg-all.html#toc-select_002c-aselect See the 4th example in the examples section. Leo ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fwd: OBS and OBS Portable Recording Into Same File Simultaneously
Trying this now! Thanks! On Thu, Jun 25, 2020 at 11:03 PM Leo Butler via ffmpeg-user < ffmpeg-user@ffmpeg.org> wrote: > David Bayles writes: > > > Corrupted video file, FFprobe output text file, and FFmpeg output text > file > > are all available to download here: > > > https://drive.google.com/drive/folders/1R0Y5plgkhTWWFNNKgSNJ5yitpKQbIKgg?usp=sharing > > > > I'm hoping there is a way to extract good video and audio from this video > > file, this was for a very important job that is impossible to do again. > The > > video linked is a 1 minute recreation of the exact same problem... the > > actual video file is confidential (for a legal court case) and is much > > longer/larger. I used the exact same settings, the only difference was > the > > window OBS was capturing (I used a YouTube video as opposed to the > > videoconference software window used for the deposition). > > I am no expert, but for the video, I would try using the select filter > to capture the I-frames. > > https://ffmpeg.org/ffmpeg-all.html#toc-select_002c-aselect > > See the 4th example in the examples section. > > Leo > ___ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fwd: OBS and OBS Portable Recording Into Same File Simultaneously
I tried this: C:\Users\David>ffmpeg -i "C:\Users\David\Downloads\2020-06-23 09-02-44 corrupt.mkv" -vf select='eq(pict_type\,I)' "C:\Users\David\Downloads\2020-06-23 09-02-44 corrupt SELECT I-FRAMES.mkv" 2> "C:\Users\David\Downloads\2020-06-23 09-02-44 corrupt SELECT I-FRAMES.txt" And got this (video and log file): https://drive.google.com/drive/folders/1OMf6oeLIkYU1d05dPqO1tfOxI1_KnYGK?usp=sharing Is that what you meant to do? Any other ideas? Thanks! On Thu, Jun 25, 2020 at 11:43 PM David Bayles wrote: > Trying this now! Thanks! > > On Thu, Jun 25, 2020 at 11:03 PM Leo Butler via ffmpeg-user < > ffmpeg-user@ffmpeg.org> wrote: > >> David Bayles writes: >> >> > Corrupted video file, FFprobe output text file, and FFmpeg output text >> file >> > are all available to download here: >> > >> https://drive.google.com/drive/folders/1R0Y5plgkhTWWFNNKgSNJ5yitpKQbIKgg?usp=sharing >> > >> > I'm hoping there is a way to extract good video and audio from this >> video >> > file, this was for a very important job that is impossible to do again. >> The >> > video linked is a 1 minute recreation of the exact same problem... the >> > actual video file is confidential (for a legal court case) and is much >> > longer/larger. I used the exact same settings, the only difference was >> the >> > window OBS was capturing (I used a YouTube video as opposed to the >> > videoconference software window used for the deposition). >> >> I am no expert, but for the video, I would try using the select filter >> to capture the I-frames. >> >> https://ffmpeg.org/ffmpeg-all.html#toc-select_002c-aselect >> >> See the 4th example in the examples section. >> >> Leo >> ___ >> ffmpeg-user mailing list >> ffmpeg-user@ffmpeg.org >> https://ffmpeg.org/mailman/listinfo/ffmpeg-user >> >> To unsubscribe, visit link above, or email >> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". > > ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-user] 360 degree VR video streaming support in ffmpeg+HLS
Hello ffmpeg team Does ffmpeg support 360 degree VR video streaming through HLS/DASH? I could see its being supported in libx264 under unofficial/experimental , but I do not see this support in HLS. when transcoding to HLS, looks like ffmpeg drops spatial metadata. please confirm. -- Best Regards *Rama krishna Thelagathoti* ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-user] Fwd: OBS and OBS Portable Recording Into Same File Simultaneously
Perhaps another question: if I extract each and every frame as an image file, then go through one by one and there is no still image resembling what I filmed (i.e. every frame/picture is just solid green), does that mean that the video is definitely unrecoverable? Or might there be another way to recover the video? On Fri, Jun 26, 2020 at 12:32 AM David Bayles wrote: > I tried this: > C:\Users\David>ffmpeg -i "C:\Users\David\Downloads\2020-06-23 09-02-44 > corrupt.mkv" -vf select='eq(pict_type\,I)' > "C:\Users\David\Downloads\2020-06-23 09-02-44 corrupt SELECT I-FRAMES.mkv" > 2> "C:\Users\David\Downloads\2020-06-23 09-02-44 corrupt SELECT > I-FRAMES.txt" > > And got this (video and log file): > > https://drive.google.com/drive/folders/1OMf6oeLIkYU1d05dPqO1tfOxI1_KnYGK?usp=sharing > > Is that what you meant to do? Any other ideas? > > Thanks! > > On Thu, Jun 25, 2020 at 11:43 PM David Bayles > wrote: > >> Trying this now! Thanks! >> >> On Thu, Jun 25, 2020 at 11:03 PM Leo Butler via ffmpeg-user < >> ffmpeg-user@ffmpeg.org> wrote: >> >>> David Bayles writes: >>> >>> > Corrupted video file, FFprobe output text file, and FFmpeg output text >>> file >>> > are all available to download here: >>> > >>> https://drive.google.com/drive/folders/1R0Y5plgkhTWWFNNKgSNJ5yitpKQbIKgg?usp=sharing >>> > >>> > I'm hoping there is a way to extract good video and audio from this >>> video >>> > file, this was for a very important job that is impossible to do >>> again. The >>> > video linked is a 1 minute recreation of the exact same problem... the >>> > actual video file is confidential (for a legal court case) and is much >>> > longer/larger. I used the exact same settings, the only difference was >>> the >>> > window OBS was capturing (I used a YouTube video as opposed to the >>> > videoconference software window used for the deposition). >>> >>> I am no expert, but for the video, I would try using the select filter >>> to capture the I-frames. >>> >>> https://ffmpeg.org/ffmpeg-all.html#toc-select_002c-aselect >>> >>> See the 4th example in the examples section. >>> >>> Leo >>> ___ >>> ffmpeg-user mailing list >>> ffmpeg-user@ffmpeg.org >>> https://ffmpeg.org/mailman/listinfo/ffmpeg-user >>> >>> To unsubscribe, visit link above, or email >>> ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". >> >> ___ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
