#timvideos

CarlFK: the thumb is running correctly?

correct

okay

then the queue is between the composite and the UI

mithro, there's no queue in the preview socket pipeline in git (tcpclientsrc ! gdpdepay ! videoconvert ! cairooverlay ! videoconvert ! xvimagesink)

thaytan: great!

so must be in the server somewhere?

before the compose/videomix, or else everything would be equally delayed

thaytan: everyone is equally delayed I think?

I thought it would be somewhere around https://github.com/timvideos/gst-switch/blob/ma...

Title: gst-switch/gstswitchserver.c at master · timvideos/gst-switch · GitHub (at github.com)

does tcpserversink have an internal queue or something?

hrm, it shouldn't matter because that pipeline is only created on a new connection?

nope - that last comment was totally wrong

mithro, lunch - bbiab

actually, I can test this

what is your test pipeline?

mithro: mine?

CarlFK: yeah the thing which does the numbers

demo... gst-launch-1.0 videotestsrc pattern=smpte is-live=1 \

! clockoverlay time-format="%S" font-desc="Sans 240" \

CarlFK: thats not all of it?

mithro: add the clockoverlay line to run-demo.sh

mithro: http://paste.ubuntu.com/8829161/

Title: Ubuntu Pastebin (at paste.ubuntu.com)

I'll be back in about an hour

k ill prollly be off to bed

back now

thaytan: any idea why http://paste.ubuntu.com/8830204/ displays the first frame and then stops for like 10 seconds before continuing?

Title: Ubuntu Pastebin (at paste.ubuntu.com)

mithro, hmm, no - not sure

thaytan: it's almost like gdppay transfers the first buffer and then sits around for a while to get it's shit together or something?

10 seconds is a long thumb-twiddle

oh

the freeze is equal to the time between starting the tcpserversink and the tcpserversrc command

ah, so it's about segments

what are segments?

segment events, describing what the timestamp range and playback offset should be

thaytan: do things need to be retimed or something?

yeah, but I'm not sure that the current behaviour is optimal

I can see why it does it

gdppay gets a segment event with start=0sec and sends it to gdpdepay

but tcpserversink drops the first n seconds of data, so gdpdepay never gets those

which means gdpdepay generates a segment event with start=0sec, but a first buffer with time=n secs

which is exactly how you describe a sequence that is 'this first buffer gets played after n secs'

after n seconds it doesn't jump though, it just starts playing delayed...

yes, that's right

thaytan: so what is the work around?

as long as the delay is smaller than what can be buffered in tcpserversink/tcp socket and friends

so how do we make it buffer nothing?

is the value of n same on both ends?

mithro, not sure - it might be something to add to gdpdepay to adjust the segment

hyades, the value of n is how many buffers tcpserversink discards before the client connects

thaytan: can we adjust the output of gdpdepay?

mithro, with a pad probe or a custom element, yes

thaytan: no inbuilt "gst-plugin-futzer"

thaytan: actually, this kind of makes sense - there is a bunch of code in flumotion which fuzts with the gddepay stuff

I think...

thaytan: would this be the timestamp stuff?

which timestamp stuff? (sounds likely, yes)

Title: flumotion/feedcomponent.py at 921550fa45ae42b2a9f51e4e522d5558fcc9a579 · timvideos/flumotion · GitHub (at github.com)

mithro, ah no - that will just make warnings about timestamp gaps in the input stream

it won't modify any timestamps

for flumotion purposes it wants to keep the segments intact anyway

because they keeps the synch between separately captured audio and video streams when muxing

I need to think about it

mithro, I think the adjustment has to be explicitly done

because both audio and video need to move by the same amount, or else they'll get out of synch

[streaming-system] aps-sids pushed 1 new commit to check-signal: http://git.io/8MLmvg

streaming-system/check-signal d12e063 Amanpreet Singh: Use HEAD request instead of GET

shenki: ping?

hyades__: ping?

mithro: pong

hyades: so you saw that ere are more comments on that pull request right?

a min

travis not building is hugely because of the pylint errors

every time a new update to pylint comes in, the make lint breaks

so, looking at https://github.com/timvideos/gst-switch/pull/26

Title: Updated Readme, fixes by hyades · Pull Request #26 · timvideos/gst-switch · GitHub (at github.com)

each place I've commented on, you should reply to the comment with if you have fixed the issue or not

mithro: I will make the build succeed on travis and come back

hyades: I said

Looks like its also broken on head, so I'm happy to merge before fixing that.

hyades: but there is a bunch of other small things which need to be fixed

hyades: should take you like 5 minutes to fix them, and then I'll do the merge

mithro: probably two things are left - the checkouts and the pylint issues

mithro: correct me if I missed anything

hyades: are you looking at the web interface?

mithro: yup, what should i be looking at?

hyades: hrm, it looks github didn't email me when you added those last two comments

hyades: you should also start sentences with capital letters

specially in comment messages

hyades: okay it looks good to merge I think

thaytan: why can't a leaky queue be good enough?

[gst-switch] mithro pushed 19 new commits to master: http://git.io/-foiPQ

gst-switch/master 18caf85 Aayush Ahuja: Update .travis-setup.sh

gst-switch/master bdf23c6 Aayush Ahuja: Update README.md [ci skip]

gst-switch/master a7a707d hyades: Merge remote-tracking branch 'upstream/master'...

mithro: yea there is a possible problem with the github. It was sending me emails for your comments the entire night

mithro, for the tcpserversink case, you don't need a leaky queue - tcpserversink should drop packets according to its soft limits already

it might be sufficient to use sync=false on the receiver and just display packets as they arrive from the network

you might get a bit more display jitter, but that's fine for the preview thumbnail

what do I put the sync=false on?

mithro, xvimagesink

good morning folks

run-demo.sh .. change 300x200 to ! video/x-raw, width=740, height=480 \

Title: 740x480 causes broken video · Issue #32 · timvideos/gst-switch · GitHub (at github.com)

CarlFK: probably its built like this