<cone-268>
ffmpeg Andreas Rheinhardt master:5935423e1e8b: avcodec/aactab: Deduplicate swb_offset_960 tabs
<cone-268>
ffmpeg Andreas Rheinhardt master:a02670ded71e: avfilter/f_drawgraph: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:a40f833bac72: avfilter/f_graphmonitor: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:1d33a310df4e: avfilter/vf_aspect: Deduplicate inputs
<cone-268>
ffmpeg Andreas Rheinhardt master:2e2c28119ff1: avfilter/vf_blend: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:e557d89ac109: avfilter/vf_chromakey: Deduplicate inputs and outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:50d3c5bd8ca5: avfilter/vf_convolve: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:93abb9b560e8: avfilter/vf_hsvkey: Deduplicate inputs and outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:748c168f8eab: avfilter/vf_xmedian: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:c32c1a18b9bb: avfilter/vsrc_testsrc: Deduplicate outputs
<cone-268>
ffmpeg Andreas Rheinhardt master:8661b5e8f9e6: avfilter/vf_format: Deduplicate inputs
<cone-268>
ffmpeg Andreas Rheinhardt master:5b85ca5317f2: avutil/x86/pixelutils: Empty MMX state in ff_pixelutils_sad_8x8_mmxext
<cone-268>
ffmpeg Andreas Rheinhardt master:392ab35db130: avfilter/vf_mpdecimate: Remove emms_c
<cone-268>
ffmpeg Andreas Rheinhardt master:2fdaeec41bef: avfilter/vf_deshake: Remove unnecessary emms_c
<cone-268>
ffmpeg Andreas Rheinhardt master:de4846dd184b: avfilter/deshake: Merge header into its only user
<cone-268>
ffmpeg Andreas Rheinhardt master:3f890fbfd901: avcodec/cbs_h2645: Fix leak of SPS VUI extension data
MrZeus__ has quit [Ping timeout: 255 seconds]
kurosu has quit [Quit: Connection closed for inactivity]
Anuj has joined #ffmpeg-devel
<cone-268>
ffmpeg Andreas Rheinhardt release/6.1:868aa88d83c8: avcodec/cbs_h2645: Fix leak of SPS VUI extension data
mkver has quit [Ping timeout: 260 seconds]
thilo has quit [Ping timeout: 240 seconds]
Anuj has quit [Quit: Client closed]
navi has quit [Quit: WeeChat 4.0.4]
thilo has joined #ffmpeg-devel
thilo has joined #ffmpeg-devel
thilo has quit [Changing host]
rcctl has joined #ffmpeg-devel
lemourin has quit [Read error: Connection reset by peer]
lemourin has joined #ffmpeg-devel
Anuj has joined #ffmpeg-devel
jamrial has quit []
microchip_ has quit [Remote host closed the connection]
microchip_ has joined #ffmpeg-devel
Anuj has quit [Quit: Client closed]
cone-268 has quit [Quit: transmission timeout]
durandal_1707 has quit [Ping timeout: 240 seconds]
durandal_1707 has joined #ffmpeg-devel
MisterMinister has joined #ffmpeg-devel
Compnn has quit [Ping timeout: 248 seconds]
Anuj has joined #ffmpeg-devel
Anuj has quit [Quit: Client closed]
Art has joined #ffmpeg-devel
AbleBacon has quit [Read error: Connection reset by peer]
kurosu has joined #ffmpeg-devel
Art has quit [Quit: Connection closed]
<elenril>
Lynne: on it
Vedaa58193 has quit [Read error: Connection reset by peer]
Vedaa58193 has joined #ffmpeg-devel
___nick___ has joined #ffmpeg-devel
MisterMinister has quit [Ping timeout: 248 seconds]
___nick___ has quit [Client Quit]
___nick___ has joined #ffmpeg-devel
<Lynne>
thanks
<haasn>
michaelni: are you happy with my swscale series?
<haasn>
I want to get csp negotiation onto the ML as soon as possible
<haasn>
but it depends on this series
<haasn>
elenril: \o/
<JEEB>
Lynne: I actually was surprised to not have a midi track identifier in QTFF
<JEEB>
although you can't really have each note in a separate packet since overlapping packets are not a thing (which is why webvtt etc have "subpackets" in mp4)
<elenril>
Lynne: seems reasonable
<elenril>
but I didn't follow your discussion with Daemon404 in detail
<Lynne>
the issue is that there are multiple layers of offsets that have to be added, depending on the codec, container, container preroll signalling, container edit list signalling, and phase of the moon
<courmisch>
isn't yuu2 supposed to be aligned to at least 32 bits always?
<elenril>
Lynne: yeah, that's why i'd prefer to wait before putting it in 6.1
<elenril>
push it into git, see if it breaks anything, then it can go in 6.1.1
<elenril>
also, FATE tests would be nice
<elenril>
for as much of these combinations as we can get
<Daemon404>
sorry, ive been down sick
<Daemon404>
i will send a review today on both
<Daemon404>
(i am still sick, but i have some comments)
<Daemon404>
5 days of rona, and i got symptoms day 5 -_-
HarshK23 has joined #ffmpeg-devel
<Daemon404>
Lynne, gimme a few hrs
<michaelni>
haasn, i think it was ok but iam not 100% sure i tested the last 2 patches, ill take another look later today
<Lynne>
elenril: I agree
<Lynne>
fate is utterly broken with a default skip
<Lynne>
I have no idea why yet
<Lynne>
I get afade/filter failures despite AAC not being anywhere involved in the chain
novaphoenix has quit [Quit: i quit]
novaphoenix has joined #ffmpeg-devel
dellas has joined #ffmpeg-devel
navi has joined #ffmpeg-devel
Anuj has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
Krowl has joined #ffmpeg-devel
Anuj has quit [Quit: Client closed]
tmm1 has joined #ffmpeg-devel
tmm1_ has quit [Ping timeout: 255 seconds]
jamrial has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<Lynne>
how am I supposed to treat your writing as anything serious now?
<Lynne>
you spend a whole chapter upfront talking about my terminology, calling me silly and stubborn for using it
<durandal_1707>
drama?!
<durandal_1707>
link?
<Lynne>
when you yourself just have not spend any time with codec researchers, but have been drinking the mp4/isobmff cool-aid
<Lynne>
you iterate over and over how I'm wrong when everything has been what I've been stubbornly trying to tell you for two days
<Lynne>
then you make up your own terms to justify you're right, and completely ignore anything codec-related underneath
<Daemon404>
you are free to believe what you wish, but the insult are not necessary. this is why i was putting off replying.
Daemon404 has left #ffmpeg-devel [#ffmpeg-devel]
<Lynne>
...
Krowl has quit [Read error: Connection reset by peer]
<Lynne>
fuck aac to the ends of the earth
<durandal_1707>
i missed this serious discussion drama entirely
* microchip_
sells popcorn
AbleBacon has joined #ffmpeg-devel
<JEEB>
not sure this stuff is only relevant to AAC :P it is just the thing which is most common around us and which does not essentially hard-code various parameters in the format either via specification or due to having only a single implementation :P
<JEEB>
also I don't think it's mp4 cool-aid when mp4 pretty much just happens to be one of the few specified containers which has facilities to deal with both priming samples as well as how much extra you need to decode in order to get proper output at your seek point
<Lynne>
matroska has that as well
<Lynne>
and I disagree with this too - the amount of samples to strip while seeking is mathematically defined
<Lynne>
in all cases but opus
<JEEB>
oh, so technically matroska could also support intra refresh seeking info delivery? good to know, never knew.
<JEEB>
and I did not do this on purpose, but writing that previous line does make me think whether the same information for such seek-delay should be utilized both audio and video. since on a high level it's a similar thing that needs to be done in both cases. number of packets or samples to have to take into account when seeking and then discarded after decode
mkver has quit [Ping timeout: 255 seconds]
Krowl has joined #ffmpeg-devel
<durandal_1707>
Lynne: for seeking you do not drop samples, you just set overlap history buffer to 0
<Lynne>
to reject any notion of mathematically defined values...
<Lynne>
to reject any attempt at defining exact terminology...
<Lynne>
to gaslight everyone into using an inaccurate undefined value for mathematical concepts
<Lynne>
to call the mathematical defined value as cargo cult hack value
<Lynne>
to call me silly and stubborn
<Lynne>
to leave immediately when I pointed this out
<Lynne>
to call the act of pointing it out as being an insult
<Lynne>
is he a spanish inquizitioner from the middle ages who discovered time travel and decided to put his talents to use on free software?
<durandal_1707>
who?
<durandal_1707>
drama again?!
<Lynne>
what drama?
<Lynne>
I am pointing out raw, proven facts
<Lynne>
*mathematically proven facts no less
<Lynne>
anyone that argues with that is not free to disagree
<durandal_1707>
where is math proof of whatever here is issue?
<Lynne>
1+1=2, 2-1=1
<durandal_1707>
rotfl
<durandal_1707>
i expected patch/code
<JEEB>
teh cargo cult references were with regards to raw ADTS, I think? objecting just putting some value there is IMHO valid.
<Lynne>
overlap + input signal = overlap + input signal, overlap + input signal - overlap = input signal
<Lynne>
it literally is 1+1=2 and 2-1 =1
<durandal_1707>
you are free to define 2*2 to whatever you wish
<durandal_1707>
same for 1+1 and/or 2-1
<JEEB>
there is a thing called algorithmic delay most certainly, but I think the thing is that the seeking delay can be something else and thus having it dynamic would make sense since this would then allow for also utilizing the information for general seeking related discarding
dellas8386 has joined #ffmpeg-devel
dellas83 has quit [Ping timeout: 240 seconds]
<Lynne>
I did agree to this
<Lynne>
no one reads what I'm writing, though!
Krowl has quit [Read error: Connection reset by peer]
Krowl has joined #ffmpeg-devel
<Lynne>
but there is an exactly defined value of seeking delay
<Lynne>
mathematically, as it is all decoder-based
Krowl has quit [Read error: Connection reset by peer]
<Lynne>
encoders may have hints, but if those hints violate mathematics, they shall be disregarded
Krowl has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
Krowl has joined #ffmpeg-devel
___nick___ has quit [Ping timeout: 255 seconds]
Krowl has quit [Read error: Connection reset by peer]
durandal_1707 has quit [Ping timeout: 255 seconds]