On Thu, 15 Apr 2021 at 01:20, Gabriel Balaich wrote:
> This seems to work great on the surface, however, if my scene ever shifts
> from something high complexity to something low complexity or vice versa
> the FFmpeg instance crashes with error "av_interleaved_write_frame():
> Unknown error".
Am 12.04.21 um 08:53 schrieb Michael Koch:
Am 12.04.2021 um 00:39 schrieb Ulf Zibis:
My question is not about head-switching effects, it mainly is about temporal
noise. E.G., there is the hqdn3d filter, which can do both, spacial and
temporal denoise. But I'm missing some guidance about
Hi Christoph,
Am 03.04.21 um 00:42 schrieb Christoph Gerstbauer:
Hello, forget my last mail. The original video is destroyed in the first place.
If you SPLIT the fields from the original MPG you see PER FIELD interlacing
artefacts. That is a sign that the original video from tape was not
> I don't like it, I definitely do not doubt the relevance (I did look into
> the issue several times), I just want to point out that once the issue
> is fixed (from our pov), it is possible that the requested feature
> (remuxing of "interlaced" j2k) will still not work, and never will.
>
> Otoh,
Hello,
I have been trying to use udp multicast to send high speed low latency
video to multiple computers simultaneously. To do this, I have been using
mpegts as a container like follows:
ffmpeg -f lavfi -i testsrc2=s=1920x1080:d=30 -c:v dirac -f mpegts udp://
239.0.0.1:1234
ffmpeg -f lavfi -i
On Wed, 14 Apr 2021 16:03:31 +0200, Moritz Barsnick wrote:
>On Tue, Apr 13, 2021 at 17:49:55 +0200, Bo Berglund wrote:
>> Question:
>> Is there a quicker way to just go from ts to mp4 so the file format is OK
>> even
>> though the video size remains at 1280x720?
>>
>> (I could live with the
I'm attempting to send a 4K60 50Mbps stream from FFmpeg to YouTube, and
can't seem to get any system fully working.
I started by trying to send the stream directly from FFmpeg to YouTube:
ffmpeg `
-guess_layout_max 0 -thread_queue_size -indexmem -f dshow
-rtbufsize 2147.48M -video_size
