doc
https://docs.nvidia.com/video-technologies/video-codec-sdk/ffmpeg-with-nvidia-gpu/
example :
ffmpeg -y -vsync 0 -hwaccel cuda -hwaccel_output_format cuda -i input.mp4
-vf scale_npp=1920:1080 -c:a copy -c:v h264_nvenc -b:v 5M output1.mp4
-vf scale_npp=1280:720 -c:a copy -c:v h264_nvenc -b:v 8M
On Tue, Nov 9, 2021 at 12:07 AM Steven Kan wrote:
> Hmm. I hadn’t considered that. What I’m actually doing is combining two
> 1920 x 1080 streams into a 3840 x 1080 stream. Would an NVidia 1030 be
> able to do that, twice?
>
>
ppl are right, 1030 cant encode (i couldn't imagine that !), only dec
you could simply plug a recent low profile nvidia (e.g. 1030) into pcie
slot of your hpe micro and nvenc would do 2 fhd h264 encodes like a charm
&rei
On Sun, Nov 7, 2021 at 11:40 PM Steven Kan wrote:
> > On Jan 18, 2021, at 10:42 PM, Carl Eugen Hoyos
> wrote:
> >
> > Am Mo., 18. Jan. 2021 um 2
6:17 PM Viacheslav Dubrovskyi
wrote:
> Hi,
>
> Try add --enable-cuda-nvcc or --enable-cuda-llvm
>
> It depends from it. Look to ffbuild/config.log
> yadif_cuda_filter_deps_any='cuda_nvcc cuda_llvm'
>
>
> 20.01.2021 00:53, andrei ka пишет:
> > hi all,
imho, you should first just test your case on a clean input on lan, make a
ts file, play it in loop with ffmpeg to multicast locally,
retransmit udp to srt locally with srt-live-transmit and test the rest of
your chain. chances are your pb is not your in your commands
but it your network / inputs..
>
> > Am 02.02.2021 um 16:16 schrieb Alessandro Molon <
> alex.mo...@vision247.com>:
> > I normally transcode using ffmpeg with h264_nvenc, that produces a NV12
> output
>
allesandro, post some ffmpeg's cmdline exemple of your nv12 nvenc output :-)
&rei
hi all,
i can't figure out how to get yadif_cuda in my ffmpeg...
i compile more or less ok ffmpeg w/ cuda (ff 4.3.1 + cuda 11.2 on ubuntu
20.4), i've just dropped fontconfig in ffmpeg, cuz in my first try it
screwed up a few cuda's explected fontconfig's symbolds. so
i configured :
./configure
well, intel mentions h264/h265 9/1 perfs ratio on older cpus here:
https://www.intel.com/content/dam/www/public/us/en/documents/white-papers/cloud-computing-quicksync-video-ffmpeg-white-paper.pdf
so, chances are this is how it is, veven if you're supposed to have enough
vid memory to do 2hd & and 4
a bit crazy stuff all these pythons :-P guess it must be useful for
something for some...
jrottenberg's docker is much simpler and pretty universal... and w/o cuda
makes <90mb... his dockerfile
is very easy do mod, i made my custom docker img based on his one with all
the latest libs + latest
cuda
>
> Start with https://trac.ffmpeg.org/wiki/Creating%20multiple%20outputs
add to the recipes in this wiki another one: you can output ffmpeg results
to a local multicast and then plug other things (ffmpeg et al) using that
multicast as a source
&rei
__
btw, i really wonder thy why want to bechnmark fflmpeg with your tiffs if
your "software will output to jpegs" :-) internally ffmpeg with convert
your tiffs anyway. convert 20s of your tiffs to jpegs (imagemagic batch),
leave threads to 0 and si if ffmpeg will be faster. and next try mjpeg
codec on
20k frames are a mere 800s of a 25fps movie... how slow is cpu process ?
afaik nvenc can only do mjpeg_cuvid for stitching images, it won't work
with tiffs...
e.g. ffmpeg -vsync 0 -hwaccel cuvid -c:v mjpeg_cuvid -framerate NN -i
image_%04d.jpg -c:v h264_nvenc x.mp4...
&rei
either do capture with multicat + ffrpobe (or mediainfo), but better try a
couple: multicat + tsanalyze (from tsduck) :
timeout --foreground 5 multicat -X -u -U @239.192.18.17:1234 /dev/null |
tsanalyze --service-analysis
&rei
___
ffmpeg-user mailing lis
>
> Re-scaling all images with FFmpeg:
> I think that's only possible with a loop in a script. Too complicated.
>
i think he could pipe it...
ffmpeg -i %%02d.JPG -i audio -vf fps=fps=1,scale=H:W -f nut - | ffmpeg -f
nut -i - -vf zoompan=... ...
&rei
___
even if there's no config options, one can aways edit sbr routines in some
way in (fdk)aac sources to castrate support of HE profile...
&rei
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscrib
oki, cool. i dont know how ffmpeg deals x11 capture, i thought probing is
used only on mpeg streams.
ak
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
ffmpeg
ffmpeg -video_size "$resolution"\
> ¦ ¦ -loglevel warning \
> ¦ ¦ -probesize 100M \
> ¦
> ¦ ¦ -f x11grab -i ":0.0+$xpos,$ypos" \
> ¦ ¦ ..
> ¦ ¦ -level:v 5.1\
>
> Where resolution is in the format e.g. 1920x1080; xpos, ypos are
> integers --
you mean the names of segments are the same, not "program time"... would
make sense to me if the idea was related to kicking the last couple of ts
out from caches (you may need to read src code for understanding why it's
happening)... what if you delete all m3u8 before restart... ?
&rei
__
>
> You could check those videos with FFprobe. May be they have no audio
> stream, or something else is different.
>
ffmpeg would be enuf, ffmpeg would list all the tracks in input files
before starting transco, on failure dude should just check his first
20lines of log to see if mp4 have snd trac
hi all,
could plz someone wise me up, in some google searches p7 preset showed up
in ffmpeg some time ago (at the same time with slow, so, today's ffmpeg
"slow" may and may not by a rename of p7), nvidia still publish p7 preset
results for cuda 11.* sdk, there's NvenC on git (chinise encoder based
btw, today on aws it's cheaper+faster to encode on nvidias (mean, not on
elemental,on ec2+gpu)...
" Gains are variable, and usually quite small."
https://trac.ffmpeg.org/wiki/CompilationGuide -> performance tips...
&rei
___
ffmpeg-user mailing list
ffmp
and if you try with a bit simpler to read command, like :
ffmpeg -re -i 'udp://127.0.0.1:12007?fifo_size=5000' -t 10 \
-map 0 -map 0 -map 0 -map 0:a:0 \
-r 25 -c:v libx264 -sc_threshold 0 -g 50 -profile:v high -preset:v fast \
-c:a libfdk_aac -ac 2 -ar 48000 \
-b:v:0 2M \
-b:v:1 365k \
-b:v:2
1) use 30 fps, who cares about ntsc fps in web universe today ? u'd
set segment_time = N*hls_time and not ask to do crazy 6.06 & 6 at the
same time ;
2) use more understandable -force_key_frames
"expr:eq(mod(n,CHUNK_SIZE_IN_FRAMES),0)" (you seem to segment to 2s
chunks = > CHUNK_SIZE_IN_FRAMES = 6
google 'ffmpeg multicast' or 'vlc multicast'. oh, fun stuff, today you can
broadcast your videos on lan via rabbitmq as well :-P
&rei
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visi
guess you should first check a bunch of subtitling programs. from you
description, you're simply doing subtiting of your vids. text placement is
easier done by specialised subtitlers,
and once you're done, you can burn 'em with ffmpeg (the only reasonable
usage of ffmpeg i can image for what you're
fyi, this is example of ffmpeg docker file with gpu encoding capabilities :
https://hub.docker.com/r/jrottenberg/ffmpeg/dockerfile
you take it, mod it for taking 4.1.2 ( FFMPEG_VERSION=4.2.1) , and make
patch mod in docker file
DIR=/tmp/ffmpeg && mkdir -p ${DIR} && cd ${DIR} && \
./configu
you should join the dockerfile you're executing, i can't know what commands
you're executing and from where.
./configure is the conf command in ffmpeg src uncompressed dir, your
commands batch (shell or docker) is not located in that folder at
the time it issues the command, so, cd ffmpeg_unzipped_
go to /tmp and check what you've got where, like (ls is likz dir) :
cd /tmp
ls
if you patch* is whre but not .txt rename it : /bin/mv patch* patch.txt
&rei
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-u
or grab this :
https://drive.google.com/file/d/1ByK7u-uOyPpUaIjVx7Ilq5tnY4rEad5e/view?usp=sharing
:-)
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
ffmpeg-us
save into patch.txt the part of URL
http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20180316/739542e4/attachment.obj
starting from --- (till the end) (rather do wget URL && vi to del the
useles mail header part, if you edit in win editor make sure you save in it
unix format, txt files in win
you're doing well, almost there :-) just leave 4.1.2 w/o $ sign, $ is for
var's values in unix sh...
&rei
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
ffmp
first type 'multucast address' in google... :-)
&rei
___
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"
oups, correction: dts patch works ok 1:1 up to 4.2.1 ...
.
say; if patch file is in /tmp/patch.txt, in docker (bask) commands (for
getting to ffmpeg's configure & make are like (patch is command, guess it'd
be patch.exe for win) :
FFMPEG_VERSION=4.2.1
curl -sLO https://ffmpeg.org/releases/ffmpeg
personally I wouldn't bother with windoze recompiling on a modern machine
unless you're badly dependent windoze's gui etc. installing ubuntu in vm
is a matter of 5-10 mins today
(or 2 mins on aws, aws's free tiny instances can do sooo much stuff)...
after your get linux you'd install docker in 3 c
hi,
try this patch :
http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20180316/739542e4/attachment.obj
(explained here:
https://ffmpeg.org/pipermail/ffmpeg-devel/2018-March/226706.html)
but it works up to 4.1.3 .
&rei
On Wed, Oct 7, 2020 at 1:46 PM Philip Orrill
wrote:
> Hi there
>
> I have
t; ffmpeg -re -i http://localhost:5055/dash.mpd dashvideo.mp4
>
>
> I have set up a webserver into IIS with a dash manifest and the segment
> chunks so I would like to know to do that?
>
>
> but it is not working.
>
>
> A lot of thanks in advance
>
> _____
On Tue, Oct 6, 2020 at 8:25 AM Mar Andrés López
wrote:
> I need to stream the video with dash over UDP. Is posible to do that in
> just one command
afaik you can't, it sends chunks and updates manifests at the same time,
you'd need rewrite ffmpeg to serialise it and encapsulate. and yet, if yo
:-) e.g. : https://github.com/leandromoreira/live-stream-from-desktop
&rei
On Tue, Oct 6, 2020 at 8:25 AM Mar Andrés López
wrote:
> Dear All,
>
>
> I am new in ffmpeg, I was Reading the documentation that ffmpeg has
> available for dash. The given command example is the following:
>
>
> ffmpeg -
did you try to transwrap it ( bmxtranswrap ) ? bmx suite and not ffmpeg
should be the first candidate to try to fix weird mxfs
&rei
On Mon, Oct 5, 2020 at 6:13 PM Schampignon Cristobal via ffmpeg-user <
ffmpeg-user@ffmpeg.org> wrote:
>
> Hi Carl Eugen,
>
> could I send it privately for example vi
>
> > I may be wrong, but try to put double quote on server url in your script
> file.
>
+ i may be wrong aslo, but it may be easier to manage line drops (which
may always happen) if
you encode w/ ffmpeg to multicast and next read that multicast and srt send
it with srt-live-transmit (and do the
hi,
what is "absolute timestamp", a security cam's time timecode burned into
video ? pts ?
i'd keep guessing, " Play the files on web using Nginx and Video.js" means
you want to encode to hls, hand it to nginx for streaming and play that
m3u8 url with ffmpeg giving "-ss hh:mm.ss..." parameter ?
ak
so, just run vlc in cmdline & no display mode ;-)
&rei
On Thu, Sep 24, 2020 at 5:31 PM Carl Zwanzig wrote:
> On 9/24/2020 6:29 AM, Firas Shahin via ffmpeg-user wrote:
> > But this tells me that it must be the software versions, since you
> didn’t have any issues.
>
> Always try with the current
(disk io check means use dd utility (cygwin) - write some bytes in parallel
with your ts creation (with dd if=/dev/random of=Y... in windoze speak), or
do it somehow else, check if your drops are only cpu induced, not but dd io)
On Tue, Sep 22, 2020 at 12:26 PM andrei ka wrote:
> hi
> som
hi
sometimes it helps to separate encoding & packaging - encode to a multicast
and read that multicast from another process and repackage it. color
conversions can eat cpu. what for do you need 2nd scaling to 1080 after
defining that you're grabbing at that resolution, your input should be
already
btw, tsduck suite can extract pts of scte35, chances are you could make
filter_complex_script with these pos to insert kframes
&rei
On Wed, Sep 2, 2020 at 6:36 AM MediaStream wrote:
> Looking for SCTE-35 pass through implementation:
>
> 1. Extract SCTE-35 from MPEG-TS.
> 2. Translate timing of
hi,
a side note, afaik, hls spec advises: "...for broadest compatibility,
Variant Streams SHOULD contain the same encoded audio bitstream..." ;
&rei
On Mon, Aug 24, 2020 at 8:03 AM KRISHNAKUMAR N K
wrote:
> On Fri, 21 Aug 2020 at 11:57, Moritz Barsnick wrote:
>
> > On Fri, Aug 21, 2020 at 0
n Wed, Jul 29, 2020 at 11:26 AM Valentin Schweitzer
wrote:
> On 28/07/2020 18:37, andrei ka wrote:
> > hi all
> >
> > i have multicast sources that sometimes actually switch to just static
> > images encoded at very low bitrates, bitrate drops from 2Mbps when it'
chances are big bmx mxf suite should be able do to that
https://sourceforge.net/projects/bmxlib/
On Wed, Jul 29, 2020 at 1:18 PM Carles Vila wrote:
> Hi, I've been asked to deliver an "MXF file (OP1A) XDCAM 422 HD (1080p)
> 25fps , 50 Mbps with 16 audio channels" with a certain audio mapping
>
>
hi all
i have multicast sources that sometimes actually switch to just static
images encoded at very low bitrates, bitrate drops from 2Mbps when it's a
decent signal to 500kbps when it's static. so, I don't know how to detect
it on a fly, ffmpeg keeps happily encoding. Wish we had some input filt
hi all,
afaik ffmpeg timeouts (~4secs max) if input udp multicast stops
broadcasting. is where any known way (with production build, i saw one ~2
yrs old patch for switching source to secondary if primary dies) to keep
ffmpeg doing dups or sending anything else, black, smtpe bars, etc, till
multic
guess it'll make jumps in pts, maybe genps on output would be useful
On Fri, Jul 10, 2020 at 1:46 PM Leonardo via ffmpeg-user <
ffmpeg-user@ffmpeg.org> wrote:
>
>
> On Friday, July 10, 2020, 1:37:39 AM GMT-3, Gyan Doshi <
> ffm...@gyani.pro> wrote:
>
>
> Yes, using the concat filter.
>
>
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
wget -O - -o /dev/null
http://wycutters.wyscout.com/box/argentina/velez_test_g1234567.mp4 |
ffprobe.new -i -
On Mon, Mar 2, 2020 at 4:01 PM Gabriele Greco
wrote:
> Hi guys,
>
> there is a way to probe a fragmented mp4 available through FTP faster? The
> operation is almost instant on a local fil
curious, why are you dying to get prores ? qt has at best 8 bits h264
encoding, you're just resampble colours to 10 bits. may be nvenc's lossless
10 bits h264 coud fit your needs as well, i'd go ~16x real time...
On Wed, Feb 26, 2020 at 5:34 PM Carl Eugen Hoyos wrote:
> Am Mi., 26. Feb. 2020 um
and why don't you do it all (almost) in ffmpeg ? make :
1) ffmpeg playlist input with your several files, transcode it the all to
required resolution / br / gops and stream it to local multicast ;
2) play that multicast in vlc (or ffplay) if you need visual control ;
3) repackage that multicast to
you don't need groups here, just make a master index.m3u8 with :
index.m3u8 :
#
#EXTM3U
#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH=128000
index_low.m3u8
#EXT-X-STREAM-INF:PROGRAM-ID=1,BANDWIDTH= 256000
index_high.m3u8
#
and
ffmpeg -re -i aac_file -vn -filter_compl
run 2 ffmpegs, one sending to local multicast, 2nd running in bash
infinite loop copying from that multicast to nfs, if nfs fais, restart
ffmpeg after the loop sees nfs coming back
Le mar. 1 oct. 2019 à 18:50, Dennis Mungai a écrit :
> Hello there,
>
> Is it possible to keep FFmpeg running when
https://trac.ffmpeg.org/wiki/Map
On Wed, Aug 21, 2019 at 3:12 PM Rudi C
wrote:
> I like to extract the English audio stream of a video file using ffmpeg.
> Also, if the file doesn’t have an English audio stream, I want to extract
> all audio streams. Preferably with a language suffix in their na
nt_size+second_level_segment_duration+program_date_time:\
hls_start_number_source=datetime:hls_time=6:hls_list_size=10:hls_segment_filename=%v_stream2_%Y%m%d%H%M%S_%%04d_%%08s_%%013t.ts:master_pl_name=playlist2.m3u8]ariant2_%v_manifest.m3u8"
On Tue, Jul 23, 2019 at 4:50 PM andrei ka wrote:
> i tried :
>
yah, guys, if someone could dump us the results of
nvEncGetEncodePresetConfig api call, that would be insteresting...
https://superuser.com/questions/1447122/what-are-the-values-iside-hevc-nvenc-presets
On Wed, Jul 24, 2019 at 2:23 AM andrew
wrote:
> I have been experimenting with GPU encoding
x27; for writing
[hls @ 0x27cefc0] Opening 'variant_1_manifest.m3u8.tmp' for writing
[hls @ 0x27cefc0] Opening 'variant_2_manifest.m3u8.tmp' for writing
[hls @ 0x27cefc0] Opening 'variant_3_manifest.m3u8.tmp' for writing
[hls @ 0x27cefc0] Opening 'playlist.m3u8' for
> It's really hard for anyone to help if you don't show us the errors.
> ;-) Preferably the complete, uncut console output.
>
> Moritz
>
it was the question for Vorlel, his scrip which he sez (in prevous msg)
might be useful for some doesn't work
___
>
also, here you'll end up getting abr bitrate control (check your "Encoding
settings" fom mediainfo) which may not be that you'd expect, for cbr you'd
set maxrate=minrate, for vbr add bufsize, for crf - crf + buffsize
On Mon, Jul 22, 2019 at 2:38 PM andrei ka wrote:
gt;> superfast -b:v:3 450k -maxrate:v:3 450k -minrate:v:3 150k \
> >>> -map a:0 -c:a:0 libfdk_aac -b:a:0 128k -metadata:s:a:0
> language=ron \
> >>> -f hls \
> >>> -var_stream_map "v:0,agroup:audio v:1,agroup:audio v:2,agroup:audio
> &
oups, i had extra space in command line, actually it works fine with 2nd
stream_map
___
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.
i tried for live, it took *only the first* var_stream_map...
ffmpeg -v verbose -y -vstats_file stats.txt -re -i 'http://URL.m3u8'
-map 0:v:1 -c:v copy -copyts -start_at_zero -metadata:s:0 language=eng
-map 0:v:2 -c:v copy -copyts -start_at_zero -metadata:s:0 language=eng
-map 0:v:3 -c:v copy -copy
hmm, i think one could extract frames with ffmpeg and pipe 'em into some
gpu encoging software. or one can invent some sort of chain processing,
like encoding video in gpu at 1 fps and then pipe into ffmpeg to split
video into frames in cpu ?
On Thu, Apr 25, 2019 at 3:38 PM Moritz Barsnick wrote:
and why not to put it on google drive :-P
On Fri, Apr 5, 2019 at 2:19 PM Yufei He wrote:
> On 04/04/2019 02:47 PM, Carl Eugen Hoyos wrote:
> >
> >> Am 04.04.2019 um 20:06 schrieb Yufei He :
> >>
> >> I'm trying to trancode a 1080i xavc long gop mxf file with my codec, the
> >> file's frame rate
with nginx you can recieve rtmp and convert it into hls without transcoding
(nginx-rtmp-module, almost no cpu wasted on it). then you can run ffmpeg
script taking that hls as a source and do whatever you need
nginx/nginx-rtmp-module worked better on linux for me, on windoze it'd get
weird freez
hi,
try aggregartp + reordertp (that is not ffmpeg, works great for me) couple
or fec option of prompeg (https://ffmpeg.org/ffmpeg-protocols.html#prompeg)
&rei
On Mon, Mar 25, 2019 at 6:44 AM Vasanth M.Vasanth
wrote:
> Hi,
>
> I am new for ffmpeg I do some transcoding on my blade servers using
you can try -copyts -start_at_zero... your input sound is 5.1, you ouput
ac=2, you'd need to do a proper downmix first :
https://superuser.com/questions/852400/properly-downmix-5-1-to-stereo-using-ffmpeg...
you can also leave ar=48000, aac will squash high frecs anyway, rates
conversions are neve
i'd try open broadcast studio for screen capture till ppl find you 100%
ffmpeg command line.. 768p should be a joke for 2500g.
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link
btw, did you set client_max_body_size in nginx.conf ? something like
100M :-)
On Wed, Mar 20, 2019 at 6:03 PM Glenn W wrote:
> Thanks Moritz,
>
> > File share host. ;-)
>
> So true. Please see below link for more full example log from my local
> tests:
>
>
> https://drive.google.com/file/d/
remarque: ppl who send ismv via http to unified streaming servers
(apache/nginx) just put http url as output, it works.
On Wed, Mar 20, 2019 at 1:20 PM Ted Park wrote:
> > In my trace which I had analyzed as a proof of concept, HTTP was
> > perfectly recognized. I had used port , while my Wi
cygwin has tail with -c N -F
On Tue, Feb 26, 2019 at 11:00 AM wrote:
> Hello,
>
> Is there a way to transcode growing files with fmmpeg ?
>
> Here are the tests I did, with 2 instances of ffmpeg running
> simultaneously :
>
> 1 - Growing file generation faster than processing
> ffmpeg.exe -i inp
arri's cmd line exporter doesn't mention piping possibility
https://www.arri.com/en/learn-help/learn-help-camera-system/tools/arriraw-converter
On Sun, Feb 17, 2019 at 5:59 PM Carl Zwanzig wrote:
> On 2/16/2019 11:57 PM, Harvey Pikelberger wrote:
> > Arri’s converter is command line and multi-pl
> Sorry for posting without testing the head version, usually I do it!
>
>
actually your crazy stuff was kinda fun to read :-P
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link a
i'd think about adding scene change detection to your fixed number of
frames extact, lot's of things may happen during 4 secs. i bet you can add
this just as an extra filter to that simple mod 240 filter, so everything
will be done in 1 pass
On Fri, Jan 25, 2019 at 3:55 PM Dennis Mungai wrote:
interesting, looks like ffv1 did zero compression here, stream size of .mov
& ffv1 avi is the same. or the file size changed, it's only mediainfo
computed decompressed stream size ?
On Mon, Jan 21, 2019 at 10:11 PM Jim Shupert
wrote:
> On 1/21/2019 2:45 PM, Moritz Barsnick wrote:
> > On Mon, Ja
I saw the bmx tool, but I couldn't see how to make it concatenate MXF
> files... do you know how I could do this? Then I will be happy to try it
>
> On Mon, 21 Jan 2019 at 22:36, andrei ka wrote:
>
> > did try doing this by bmx ? https://sourceforge.net/p/bmxlib/home/Home/
>
did try doing this by bmx ? https://sourceforge.net/p/bmxlib/home/Home/
On Mon, Jan 21, 2019 at 5:50 PM Swami Kevala <
swami.kev...@ishafoundation.org> wrote:
> Hello,
>
> I take a Sony XAVC-I MXF file and run the following ffmpeg command:
>
> ffmpeg -i samples/fs7/card2/XDROOT/Clip/A002C001_1901
hello,
i use ffmpeg (4.1) to stream live hls using unified streamnig.
if in my abr encoding i mix baseline with main/high, live hls generated by
unified (USP) would
have #EXT-X-DISCONTINUITY on each segment :
encoding :
ffmpeg-y -re -i 'udp://239.255.255.1:21007?localaddr=127.0.0.1' \
-filter_c
82 matches
Mail list logo
