Dear Stuart,
I could see video normally and audio was noticeably pitched down,
since there was no time code: [dv @ 01c72fc2f840] Detected timecode is
invalid
I cannot know if people were moving faster, or it was just an after party :)
1. change pcm codec to pcm_s24le in ffmpeg line.
2. tr
>From my experience with dv and dvcam tapes I would recommend capturing DV
with Enosoft DV Processor in a raw dv stream (extension *.dv)
and then process it with ffmpeg or ffmbc.
However, the problem that you are experiencing might be related to the
audio resampling, since video goes normally as
- *tbn* = the time base in AVStream that has come from the container.
- *tbc* = the time base in AVCodecContext for the codec used for a
particular stream.
- *tbr* = *tbr* is guessed from the video stream and is the value users
want to see when they look for the video frame rate.
O
No.
On Tue, 28 Aug 2018, 05:03 Carl Zwanzig, wrote:
> Hello,
>
> Are there any readily-available & recent windows builds with BMD decklink
> enabled? The 4.0.x builds from Zeranoe do not and I'd rather not set up a
> windows build environment for what should (otherwise) be a quick
> experiment.
Recommendation:
-several ssd drives in raid 0 configuration ex. Samsung EVO PRO ( allows
for capacity and speed, and longer read/write life-cycle) Calculate the
space you need based on codec selection or god forbid based on that long 4K
raw video :)
-use thunderbolt port straight from camera or c
before the
part of the command where tee pseudomuxer is being overrided by fifo, and
inheritance is effective for any slave f=mpegts muxer,
so that you don't have to repeat so many switches for multiple mpegts
instances being created
Thanks.
On Tue, Jul 24, 2018 at 8:22 PM, Carl Eugen Hoyos
w
-L -p 1234
Untill next time,
On Sat, Jul 21, 2018 at 12:35 AM, Dragan Randjelovic
wrote:
> I am using ffmpeg on Windows, [ffmpeg version N-91217-g2bd24d4a37]
>
> 1. How can I pipe output of ffmpeg without saving it to a file to three
> different processes?
>
> 2. Is it possibl
-L -p 1234
Untill next time,
On Sun, Jul 15, 2018 at 9:46 PM, Dragan Randjelovic
wrote:
> Hi,
>
> I am trying to achieve the following under windows and I am wondering what
> would be the right syntax for it,
> while explicitly I do not want to invoke another ffmpeg process.
>
00|[f=rtp_mpegts]rtp://ip_address:11000"
| tee > (nc64 -L -p 9000 )
but I still have no test results, which I will report offcourse.
Regards.
On Sat, Jul 21, 2018 at 12:35 AM, Dragan Randjelovic
wrote:
> I am using ffmpeg on Windows, [ffmpeg version N-91217-g2bd24d4a37]
>
> 1. Ho
00|[f=rtp_mpegts]rtp://ip_address:11000"
| tee > (nc64 -L -p 9000 )
but I still have no test results, which I will report.
Regards.
On Sun, Jul 15, 2018 at 9:46 PM, Dragan Randjelovic
wrote:
> Hi,
>
> I am trying to achieve the following under windows and I am wondering what
> w
It is about one input stream - mpets that has already been muxed properly
including metadata, that I want to multiply and deliver to different
processes, and/or use different transport protocols like udp or rtp over
tcp/udp, ex: one for multicast streaming , one for unicast connection to a
streamin
I am using ffmpeg on Windows, [ffmpeg version N-91217-g2bd24d4a37]
1. How can I pipe output of ffmpeg without saving it to a file to three
different processes?
2. Is it possible to use encoded stream with different muxers ex. -f mpegts
> udp and -f rtp_mpegts > rtp
and pipe first stram to pro
w days ago there was an email about missing metadata support
in rtp_mpetgts muxer
Regards.
Dragan
On Sun, Jul 15, 2018 at 10:25 PM, Moritz Barsnick wrote:
> On Sun, Jul 15, 2018 at 21:46:58 +0200, Dragan Randjelovic wrote:
> > I am trying to achieve the following under windows and
Hi,
I am trying to achieve the following under windows and I am wondering what
would be the right syntax for it,
while explicitly I do not want to invoke another ffmpeg process.
-f tee "[onfail=ignore][f=mpegts] pipe:1>nc64 -L -p
1234|[f=mpegts]udp://IP:1234/"
Regards.
__
I am experiencing similar issues once I want to name tracks and set
language code.
Regards.
Dragan R.
On Mon, Jul 2, 2018 at 11:24 AM, Moritz Barsnick wrote:
> On Wed, Jun 27, 2018 at 16:28:26 +0200, Christian Knoefel | Ilm-Provider
> wrote:
> > Hello,
>
> Hi Christian,
> thanks for providing t
Here is an interesting article:
https://avpres.net/FFmpeg/SD_HD.html
as far as scaling and re-sampling algo goes I would invoke avisynth and try
thin raster with
Spline64Resize(1440,1080)
as it usually gives better Y-PSNR metric.
Try also sinc re-sampler and see if it suits your case. Onc
could you please provide complete ffmpeg command line including output.
please include the following outputs:
ffmpeg -f decklink -list_devices 1 -i dummy
you will get the list of installed decklink adapters:
ex: [decklink @ 0x2cad400] 'DeckLink Quad (1)'
then issue the following command and pro
use nal-hrd option:
-x264opts nal-hrd=cbr:force-cfr=1 ( -b:v 25M -minrate 25M maxrate 25M
-bufsize 25M )
*hrd=cbr*
Regards.
On Thu, Aug 10, 2017 at 3:28 PM, jmkhoo1
wrote:
> If you mean a constant frame rate, use -r:v 25. If you actually mean
> bitrate, use -b:v and -b:a. I think you might
18 matches
Mail list logo
