BtbN changed the topic of #ffmpeg to: Welcome to the FFmpeg USER support channel | Development channel: #ffmpeg-devel | Bug reports: https://ffmpeg.org/bugreports.html | Wiki: https://trac.ffmpeg.org/ | This channel is publically logged | FFmpeg 7.0 is released
Mister_Magister has quit [Quit: bye]
Mister_Magister has joined #ffmpeg
Juest has quit [Ping timeout: 264 seconds]
Juest has joined #ffmpeg
mven971 has joined #ffmpeg
mven97 has quit [Ping timeout: 255 seconds]
mven971 is now known as mven97
liana has joined #ffmpeg
minimal has quit [Quit: Leaving]
Muimi__ has joined #ffmpeg
hightower2 has quit [Remote host closed the connection]
hightower2 has joined #ffmpeg
MightyBOB has quit [Remote host closed the connection]
<mightysands>
So I have a video file in 1080p but it's visibly in a 4:3 aspect ratio - just naturally - and it seems to have black bars on either side of the video. Top and Bottom, but on each side as well
<mightysands>
I tried running crop detect, but all it seemed to do was remove the leftmost black bar, leaving the top, bottom and right side black bars intact, with the exception of the right hand size, whose black bar greatly increased in size, to the point where the video window itself shrunk and shifted to the left half of the monitor
<mightysands>
Oh...nvm. It just didn't adjust the aspect when I changed my window to fullscreen. If I open ffplay with the -fs option, it's perfect
<mightysands>
haha
<mightysands>
still weird, but hey, I got something I wanted
sgm has quit [Remote host closed the connection]
sgm has joined #ffmpeg
Livio has quit [Quit: leaving]
mightysands has left #ffmpeg [ERC (IRC client for Emacs 27.2)]
iliv has joined #ffmpeg
waleee has joined #ffmpeg
beaver has quit [Ping timeout: 260 seconds]
beaver has joined #ffmpeg
emanuele6 has quit [Read error: Connection reset by peer]
emanuele6 has joined #ffmpeg
emanuele6 has quit [Read error: Connection reset by peer]
dreamon has joined #ffmpeg
emanuele6 has joined #ffmpeg
emanuele6 has quit [Read error: Connection reset by peer]
emanuele6 has joined #ffmpeg
<rajivharlalka>
I am having a confusion, I read an HDR image on MATLAB and it tells the image has 3 dimensions(maybe RGB) and each value is in the range of uint16_t. But when I read the image through the AVframe data structure on ffmpeg, I get one value of range uint8_t. How are the two things related 🤔
<BtbN>
sounds like you told it to convert to 8bit somehow?
<rajivharlalka>
what do mean by it here? ffmpeg? I am tinkering with making a new filter and fetched the pixel values from AVframe and just printed it.
<rajivharlalka>
So I don't see an option of getting a RGB distinct values each of uint16_t.
<BtbN>
If your filter only advertises 8 bit formats, ffmpeg/lavfi will automatically convert
Haxxa has quit [Quit: Haxxa flies away.]
Haxxa has joined #ffmpeg
YuGiOhJCJ has joined #ffmpeg
vampi__ has quit [Remote host closed the connection]
vampi__ has joined #ffmpeg
coldfeet has quit [Remote host closed the connection]
rv1sr has quit []
zmt00 has joined #ffmpeg
tm512 has joined #ffmpeg
<tm512>
so with these RTMP frame drops that were plaguing me with ffmpeg, I started getting pretty suspicious when they seemed to be happening just as often at bitrates that used to cause no issue, so I switched back from the 7.0 branch over to the 6.1 branch, and that appears to have resolved things
<tm512>
hoping I don't end up needing to bisect things to figure out where the regression was introduced, since it's not exactly trivial to actually reproduce things. sometimes the frame drops occur within a few minutes, other times they take the better part of an hour before showing up
bitbinge has joined #ffmpeg
jagannatharjun has quit [Quit: Connection closed for inactivity]
<BtbN>
well, sounds like bisect time to me at least
jtgd has quit [Quit: WeeChat 4.3.1]
jtgd has joined #ffmpeg
chandash has joined #ffmpeg
<tm512>
bisecting a regression/bug that is so nondeterministic is just incredibly aggravating. if I try to go quickly, I risk marking a commit incorrectly, so ideally, I have to let each commit run for at least a couple hours, in order to have any degree of confidence
<BtbN>
well, it's just not gonna get fixed/found then
emanuele6 has quit [Read error: Connection reset by peer]
e^pi-1 has joined #ffmpeg
<tm512>
I mean, not necessarily. definitely the case that I, without any real familiarity with ffmpeg's source code, wouldn't have much chance in knowing what is causing this issue without bisecting, but not everyone is as dumb :p
<BtbN>
Well, ffmpegs rtmp code has always worked fine for me, for very extended periods of time.
<BtbN>
Without a clear reproducer you put in a bug report, or a bisect and issue report, it's very unlikely anything will happen
<tm512>
what's weird is that there don't seem to be any significant changes in the RTMP code between 6.1 and 7.0, at least nothing that would change behavior in my use case
emanuele6 has joined #ffmpeg
SuicideShow has quit [Ping timeout: 255 seconds]
lavaball has quit [Remote host closed the connection]
System_Error has quit [Ping timeout: 260 seconds]
SuicideShow has joined #ffmpeg
chandas2 has joined #ffmpeg
chandash has quit [Ping timeout: 246 seconds]
<BtbN>
It could also very well be related to anything else
<BtbN>
the threading in ffmpeg.c, changes in the tcp code, something completely random
Sakura`Kinomoto has quit [Remote host closed the connection]
chandas2 has quit [Ping timeout: 252 seconds]
chandash has joined #ffmpeg
System_Error has joined #ffmpeg
Sakura`Kinomoto has joined #ffmpeg
finsternis has quit [Ping timeout: 252 seconds]
finsternis has joined #ffmpeg
<tm512>
I could maybe test with some other protocols. I think I also need to test with x264, though that was having serious issues when I tested it most recently. even recording to a local file, not involving the network at all, within a couple minutes of recording, x264 starts dropping so many frames that the framerate effectively drops from 60 to 30 or 20fps
<tm512>
not sure if it's a related issue. the end result is kinda similar, with frames dropping constantly
talismanick has quit [Remote host closed the connection]
lec_thege8042726 has quit [Remote host closed the connection]