I'll take a look at KCP (thanks for mentioning it!).
I'm really taking SRT to the limit for when problems or connection
interruptions occur. Apparently the implementation of SRT within FFMPEG has
some problems that do not happen when I use SRT-live-transmit with output
to udp while reading this
On 2/18/2019 4:03 PM, Ulf Zibis wrote:
I note, that those composed PNG bitmaps consume 10 % of the footprint of
the video. Is there a possibility to compress them to e.g. JPEG?
PNG _is_ compressed, it's just a different algorithm from JPEG (or GIF).
z!
Thank for the detailed information. I'm trying to make full use of MPEG-2,
because it's less complex. It's mainly disgned for MP@ML set-top boxes , just
hope to keep compliance with DVD standards.
After I corrected flags, it works pretty well. I know a little about
sc_threshold, b_strategy and
On 2/18/19, 6:26 PM, "ffmpeg-user on behalf of Ted Park"
wrote:
Is this a problem? When you link it to the original, everything is offset
by the difference?
Exactly, we have a workflow where a user can mark an in and out against the
lores version, and when we cut this out of the
Am 19.02.19 um 00:28 schrieb Carl Eugen Hoyos:
> 2019-02-18 22:27 GMT+01:00, Ulf Zibis :
>
>> Stream #0:6 -> #0:6 (dvd_subtitle (dvdsub) -> ass (ssa))
>> Stream #0:7 -> #0:7 (dvd_subtitle (dvdsub) -> ass (ssa))
>> Stream #0:8 -> #0:8 (dvd_subtitle (dvdsub) -> ass (ssa))
>> Stream #0:9 ->
2019-02-18 22:27 GMT+01:00, Ulf Zibis :
> Stream #0:6 -> #0:6 (dvd_subtitle (dvdsub) -> ass (ssa))
> Stream #0:7 -> #0:7 (dvd_subtitle (dvdsub) -> ass (ssa))
> Stream #0:8 -> #0:8 (dvd_subtitle (dvdsub) -> ass (ssa))
> Stream #0:9 -> #0:9 (dvd_subtitle (dvdsub) -> ass (ssa))
> Stream
Is this a problem? When you link it to the original, everything is offset by
the difference?
___
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user
To unsubscribe, visit link above, or email
Hi Moritz,
Am 18.02.19 um 22:27 schrieb Ulf Zibis:
>> What's in these files? Are they really text? (I hadn't heard of srtx so
>> far.)
> Yes, they are really text. Seem to be an intermediate data file for
> dvd::rip. I guess, they record the timestamps for the bitmap subtitles.
Oops, I'm sorry, I
On Mon, 18 Feb 2019, DeBacker, Bart wrote:
On 2/17/19, 5:57 PM, "ffmpeg-user on behalf of Carl Eugen Hoyos"
wrote:
2019-02-17 23:26 GMT+01:00, DeBacker, Bart :
> ffmpeg -i hires.mxf -s 960x540 lores.mp4
>
> ffmpeg version 4.0.1
This is soon one year old.
Carl
> I guess you mean the files named "pic.srtx". But what should I do with
> them? The content looks like:
> 1
> 00:00:31,160 --> 00:00:34,429
> /media/Sicherung/Filme/VLuN/dvdrip/tmp/subtitles/002-C001/00/pic0001.pgm.txt
Those look like srt.. Maybe it is doing OCR on the vobsubs and producing text
Hi Moritz,
Am 18.02.19 um 22:02 schrieb Moritz Barsnick:
> On Mon, Feb 18, 2019 at 21:13:29 +0100, Ulf Zibis wrote:
>> over 6 GB memory, so my machine heavily swapped
> Probably okay if you reduce the -probesize and -analyzeduration to fit
> your file. Sorry, it's a bit hit and miss.
Yes, it was
On Mon, Feb 18, 2019 at 21:13:29 +0100, Ulf Zibis wrote:
> over 6 GB memory, so my machine heavily swapped
Probably okay if you reduce the -probesize and -analyzeduration to fit
your file. Sorry, it's a bit hit and miss.
> Stream #0:6[0x26]: Subtitle: dvd_subtitle
> Stream #0:7[0x20]:
Thanks Ted.
Am 18.02.19 um 19:59 schrieb Ted Park:
>> I got the same results with that. Does it matter at which position I put
>> these arguments and does it last to use only one of the two?
> I’m pretty sure probe size and analyze duration are per file, put it before
> the input. And they are
Ted Park wrote:
>Also, raw files from camera sensors in general aren’t just
>decoded, they’re rendered to a colorspace. Wouldn’t you want to
>review the original (for grading, etc) before doing anything
>with it using ffmpeg?
Yep!
___
ffmpeg-user
>>> Not directly, Ari needs to be debarred first at least.
>
> de-Bayered?
Thanks..
> Arri’s converter is command line and multi-platform, so it may well
> coordinate nicely with FFmpeg.
That’s just the command line tool though. They do have an SDK, but you need to
apply become a partner to
> I got the same results with that. Does it matter at which position I put
> these arguments and does it last to use only one of the two?
I’m pretty sure probe size and analyze duration are per file, put it before the
input. And they are both limits, so I think you need to use both (if you
Hi Moritz,
Am 18.02.19 um 17:06 schrieb Moritz Barsnick:
> ffmpeg needs to detect them *before* demuxing. That's why I told you to
> use large values for "-probesize" and "-analyzeduration". They tell
> ffmpeg to look deeply into the video before deciding which input
> streams are available. Then
Hi Moritz,
thanks for your additional hints.
Am 18.02.19 um 17:06 schrieb Moritz Barsnick:
> ffmpeg needs to detect them *before* demuxing. That's why I told you to
> use large values for "-probesize" and "-analyzeduration". They tell
> ffmpeg to look deeply into the video before deciding which
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
On Mon, Feb 18, 2019 at 16:29:01 +0100, Ulf Zibis wrote:
> >> Also there should be 13 subtitle streams in this video, but I don't find
> >> them here.
> > In VOBs, subtitles tend to appear later in the files. Use large values
> > for "-probesize" and "-analyzeduration" for seeing them (not only
Am 18.02.19 um 14:56 schrieb Moritz Barsnick:
> You can use "-map 0" to get *all* streams from the input. This is at
> the risk of also mapping e.g. data streams, which the output format may
> not support.
Great, this works fine now.
>> Also there should be 13 subtitle streams in this video,
Hello,
I'm trying to grab live stream from youtube to process it with opencv in
python, in order to do so I run subprocess './ffmpeg.exe -i
*Stream_Url_GoesHere* -f image2pipe -pix_fmt bgr24 -vcodec rawvideo -' and
read decoded image from pipe. This works, however, m3u8 playlist on youtube
On Mon, Feb 18, 2019 at 11:56:42 +0100, Ulf Zibis wrote:
> I have a ripped vob stream with 5 audio streams. I want all these audio
> streams in the mp4 file. What to do for this?
Oh, another thing: It depends on how these VOBs were ripped. In most
cases, ffmpeg won't we able to handle them as a
On Mon, Feb 18, 2019 at 11:56:42 +0100, Ulf Zibis wrote:
> I have a ripped vob stream with 5 audio streams. I want all these audio
> streams in the mp4 file. What to do for this?
Read here:
https://ffmpeg.org/ffmpeg.html#Automatic-stream-selection
"In the absence of any map options for a
Hi Dave.
No, I started investigating in using Avid’s own SDK. Haven’t gotten it working
yet.
Best, Kasper
> On 16 Feb 2019, at 18.19, David Prosser wrote:
>
> Hi all,
> I'm late to the conversation but, did either of you have any luck in
> getting ffmpeg to create Avid readable MXFs? I'm
Hi,
I have a ripped vob stream with 5 audio streams. I want all these audio
streams in the mp4 file. What to do for this?
Also there should be 13 subtitle streams in this video, but I don't find
them here.
$ ffmpeg -i
26 matches
Mail list logo
