Main page
[00:29] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[00:37] s55 (n=Scott@cpc1-livi1-0-0-cust180.sgyl.cable.virginmedia.com) left irc: "Leaving"
[00:42] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[00:43] brianmario (n=brianmar@c-24-7-56-53.hsd1.ca.comcast.net) joined #handbrake-dev.
[00:55] Rodeo-laptop (n=tim@adsl-84-227-221-220.adslplus.ch) joined #handbrake-dev.
[00:58] BradleyS (n=BradleyS@99-37-5-191.lightspeed.bcvloh.sbcglobal.net) left irc: Client Quit
[00:59] BradleyS (n=BradleyS@99-37-5-191.lightspeed.bcvloh.sbcglobal.net) joined #handbrake-dev.
[01:57] mduell (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) joined #handbrake-dev.
[02:25] Rodeo-laptop (n=tim@adsl-84-227-221-220.adslplus.ch) left #handbrake-dev.
[02:55] mduell_ (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) joined #handbrake-dev.
[02:59] mduell (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) left irc: Read error: 110 (Connection timed out)
[03:41] brianmario (n=brianmar@c-24-7-56-53.hsd1.ca.comcast.net) left irc: Client Quit
[03:50] mduell (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) joined #handbrake-dev.
[03:56] bogo_lode (n=bogo_lod@75.108.108.82) left irc: Client Quit
[04:25] HandBot joined #handbrake-dev.
[04:25] #handbrake-dev: mode change '+o HandBot' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[04:59] brianmario (n=brianmar@c-24-7-56-53.hsd1.ca.comcast.net) joined #handbrake-dev.
[05:13] [saintd3v] j45: they tell you how you got k-lined?
[05:16] [j45] not really. they suggested i clicked on one of the links. which i didn't do of coarse.
[05:47] mduell- (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) joined #handbrake-dev.
[06:07] mduell (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) left irc: Read error: 60 (Operation timed out)
[06:18] jbrjake-laptop (n=jbrjake@pool-72-91-188-93.tampfl.fios.verizon.net) left irc:
[06:38] mduell- (n=mduell@rrcs-24-153-202-98.sw.biz.rr.com) left irc: Read error: 110 (Connection timed out)
[07:43] barf (n=Dark_Shi@193.69.144.164) left irc: Excess Flood
[08:39] [saintd3v] j45: yeah, i was in #freenode shortly after it happened, had quite a few people in there
[08:39] [saintd3v] saying similar things
[08:39] [saintd3v] one guy even had 2 of his channel bots get k-lined
[08:40] [saintd3v] lots of other known afks, one guy was even playing quake3 at the time he got k-lined.
[09:55] ritsuka (n=Ritsuka@87.19.31.32) joined #handbrake-dev.
[09:55] brianmario (n=brianmar@c-24-7-56-53.hsd1.ca.comcast.net) left irc: Client Quit
[09:55] #handbrake-dev: mode change '+v ritsuka' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[10:01] superdump (n=rob@unaffiliated/superdump) joined #handbrake-dev.
[10:01] #handbrake-dev: mode change '+v superdump' by HandBot!n=HandBot@concorde.lapsus.org
[11:01] superdump (n=rob@unaffiliated/superdump) left irc: "WeeChat 0.3.2-dev"
[12:10] s55 (n=Scott@cpc1-livi1-0-0-cust180.sgyl.cable.virginmedia.com) joined #handbrake-dev.
[12:10] #handbrake-dev: mode change '+o s55' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[14:58] bogo_lode (n=bogo_lod@75.108.108.82) joined #handbrake-dev.
[14:58] berylium_work (n=john@69.15.172.241) left irc: Read error: 110 (Connection timed out)
[15:41] Rodeo (n=tim@lap34-2-82-237-95-151.fbx.proxad.net) joined #handbrake-dev.
[15:41] #handbrake-dev: mode change '+v Rodeo' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[16:40] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[16:50] refulgentis_ (n=jpo@pool-71-186-235-161.bflony.fios.verizon.net) left irc: Client Quit
[16:58] jbrjake (n=jbrjake@219-253.127-70.tampabay.res.rr.com) left irc: Read error: 60 (Operation timed out)
[17:01] jbrjake (n=jbrjake@219-253.127-70.tampabay.res.rr.com) joined #handbrake-dev.
[17:02] #handbrake-dev: mode change '+o jbrjake' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[17:08] [CIA-23] sr55 * r3083 win/C#/frmMain.Designer.cs:
[17:08] [CIA-23] WinGui:
[17:08] [CIA-23] - Added keyboard shortcuts:
[17:08] [CIA-23] * Ctrl-O = Open File
[17:08] [CIA-23] * Ctrl-Shift-O = Open Folder
[17:08] [CIA-23] * Ctrl-A = Show Activity Log window
[17:08] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[17:15] [CIA-23] sr55 * r3084 win/C#/ (frmMain.Designer.cs frmMain.cs):
[17:15] [CIA-23] WinGui:
[17:15] [CIA-23] More Keyboard ShortCuts
[17:15] [CIA-23] - Ctrl-A remapped to add to queue
[17:15] [CIA-23] - Ctrl-L for log window
[17:15] [CIA-23] - Ctrl-S to start an encode
[17:25] Um_cara_qualquer (n=dontknow@201-89-244-77.cscgo702.dsl.brasiltelecom.net.br) joined #handbrake-dev.
[17:28] Um_cara_qualquer (n=dontknow@201-89-244-77.cscgo702.dsl.brasiltelecom.net.br) left #handbrake-dev.
[17:28] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[17:32] *** s55 wonders if we can integrate MIT code into HB without causing licence nightmare
[17:33] [Rodeo] what code do you want to integrate?
[17:33] [s55] http://www.codeproject.com/KB/database/CsvReader.aspx
[17:36] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[17:36] [Rodeo] looks like what the FSF calls the "Expat license" - they claim it's GPL-compatible
[17:36] [Rodeo] http://www.fsf.org/licensing/licenses/
[17:37] [s55] yay
[17:56] Bebbo (n=Bebbo@78.32.184.5) joined #handbrake-dev.
[17:59] Bebbo (n=Bebbo@78.32.184.5) left irc: Client Quit
[18:27] [j45] Dark_Shikari: hrm, VFR_maniac just moved his dts rearranging from mp4.c to encoder.c. it still causes the dts of the first frame to be used twice, once in the calculation of the first frames dts and once in the calculation of the 4th frames dts (when bframes && bpyramid). I just don't see the logic in that. but i don't want to be rude. it's not my place to critique x264 code.
[18:28] [j45] tho the change does make mp4.c correct (in my view)
[18:39] [CIA-23] sr55 * r3085 win/C#/Functions/ (Main.cs QueryGenerator.cs UpdateCheckInformation.cs):
[18:39] [CIA-23] WinGui:
[18:39] [CIA-23] - Quick patch to allow "," to be used in CSV files when importing into the gui.
[19:05] [Rodeo] j45: I'm trying to understand your example from http://handbrake.fr/irclogs/handbrake-dev/handbrake-dev20100122_pg1.html
[19:06] [Rodeo] so I attempted to represent it, and there's something weird about it: http://handbrake.fr/pastebin/pastebin.php?show=1151 - note that I only started looking into DTS and PTS last night so tbh I know very little about it
[19:06] [Rodeo] hope I'm not wasting your time
[19:09] [j45] that may have been the bad example i gave. and my method of generating dts doesn't match what x264 is doing. i need to do a different example to show where x264 would actually fail.
[19:10] [Rodeo] tbh, the way I understand DTS and PTS, DTS can't be > PTS
[19:11] [Rodeo] as ritsuka said the other day, frames can't be decoded after they're displayed
[19:11] [j45] it's not suppose to ever be. but the current way x264 is generating it, that can happen. which is what i'm trying to help fix
[19:11] [Rodeo] so if you get a DTS > PTS, something is wrong, and I'm not sure how you'd go about recovering from it
[19:12] [j45] did you see this:
[19:12] [j45] (01:23:52 PM) j45: kierank: i've been thinking about dts and running over a variety of examples of vfr frame timings and have an idea. I can't say it's completely baked yet, but the idea is to keep track of how many frames are decoded but not yet displayed at any time, and maintain that value at the necessary number of reference frames. here's some pseudo code that it think does the job http://pastebin.com/mb6bc1f1
[19:13] [Rodeo] hmm, wait I'm starting to understand something (I think) - let me think about it for a while
[19:15] [j45] it's a difficult thing to wrap your head around. I constantly loose my train of thought when trying to work out the details.
[19:15] [Rodeo] DTS and PTS are integers, right? what are they representing, a time (e.g. in milliseconds) or something more abstract?
[19:15] [j45] they represent a time. the timebase can be anything. we happen to use 1/90000 sec
[19:16] [Rodeo] do DTS and PTS necessarily use the same timebase?
[19:16] [j45] yes
[19:16] [Rodeo] OK
[19:16] [Rodeo] and are they absolute times or relative times?
[19:18] [j45] in our output, they start at 0 for the first frame and are representations of the absolute time from the beginning of the video. but input sources can have dts and pts that don't start at 0 and have discontinuities during the sequence
[19:18] [Rodeo] oh OK - makes it more fun I guess
[19:19] [j45] thats where PCR's come in. they tell you where the discontinuities are and how to adjust for them
[19:19] [Rodeo] oh
[19:19] [Rodeo] didn't know that was related
[19:21] [Rodeo] but by the time we send the decoded input to x264, HB has stripped any discontinuities, right?
[19:21] [j45] correct. we rebase the start to 0 and remove discontinuities
[19:21] [Rodeo] i.e. the input PTS it send to x264 is monotonous (hopefully I'm getting the term right)?
[19:21] [Rodeo] OK
[19:34] [j45] so x264 takes a sequence in coded order and assignes pts's that are in presentation order to the dts's of the sequence then subtracts init_delay to generate dts. here's an example of how that breaks. presentation order I B1 B2 P with durations 1 1 4 1 implies pts 0 1 2 6. coding order is I P B2 B1 and pts (in coded order) 0 6 2 1. dts is pts (in presentation order) - init_delay -2 -1 0 4. note that frame 4 (coded order) is presented at
[19:34] [j45] the solution is to shorten the dts duration of the problem frame. this can be done without affecting the presentation time
[19:36] [Rodeo] let me try to represent it - BTW did my previous representation seem correct?
[19:36] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[19:36] [j45] i need to give it a good read. haven't yet. just glanced at it
[19:37] [Rodeo] OK
[19:38] [j45] if bpyramid, coding order could be i0 p3 b2 b1
[19:38] [Rodeo] OK your example assumes b-pyramid?
[19:38] [j45] yes
[19:38] [Rodeo] I see
[19:39] [j45] i could probably generate one that doesn't assume that. i kind of blindly plugged in numbers that my gut told me would work
[19:40] [Rodeo] well, let me think about it first
[19:40] [Rodeo] if you don't mind
[19:43] [Rodeo] OK in your example, PTS (coding order) is 0 6 2 1, and "DTS is PTS in presentation order" -> so DTS in coding order is 0 1 2 6?
[19:44] [j45] minus init_delay which is 2 frames when bframe && bpyramid, so -2 -1 0 4
[19:45] [Rodeo] ah, so those values are DTS, I see
[19:46] [Rodeo] and so the last offset is negative
[19:48] [j45] correct. but it's not necessary for the last dts to be 4. so long as it is > 0, decoding will work. so an adjustment can be made to fix the offset.
[19:48] [Rodeo] I see
[19:48] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[19:48] [j45] 0 here because the previous dts was 0
[19:49] [Rodeo] yes
[19:49] [j45] the trick is that there should never be more than 2 decoded frame waiting to be displayed. so whatever adjustments you make have to be balanced
[19:50] [Rodeo] hmm
[20:02] [Dark_Shikari] j45: critiquing is always good
[20:02] [Dark_Shikari] that's why #x264dev exists
[20:02] [Dark_Shikari] we want you to
[20:06] [Rodeo] j45: hopefully I got this right (at least, makes sense now) http://handbrake.fr/pastebin/pastebin.php?show=1153
[20:09] [j45] Rodeo: yes. of coarse you skimmed over the details of how the dts adjustment of b1 is made ;)
[20:09] [Rodeo] I guess
[20:10] [Rodeo] I'm only representing the results, not the actual adjustment algorithms
[20:34] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[20:37] ritsuka (n=Ritsuka@87.19.31.32) left irc:
[20:55] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[20:59] [Rodeo] lol, he's going to run a "big video site"
[20:59] [s55] :/
[20:59] [s55] seems to be that time of month again
[21:00] [Rodeo] yep
[21:00] [Rodeo] BTW, j45, your codes does make sense now (I think)
[21:01] [s55] arrg, just bought a tv show on iTunes thinking it might be something worth watching. First 30 seconds and I realise I've made a mistake
[21:01] [s55] :/
[21:01] [j45] looking at VFR_maniacs code, i'm starting to think it makes sense as well. just need to wrap my head around one last question
[21:02] [Rodeo] s55: write to sjobs@apple.com asking for a refund (don't forget to promise you'll buy the tablet as soon as it hits Apple retail)
[21:03] [s55] I'll probably want the damned thing anyway
[21:03] [Rodeo] yeah, me too
[21:03] [Rodeo] whether I'll actually buy it remains to be seen though
[21:03] [s55] I just forked out for a 3gs, so I'm not wanting to spend money on one :/
[21:05] [s55] Was going to grab a nexus one, but they still having problems with it :(
[21:05] *** Rodeo wonders what iPhone OS 4.0 will look like on a 3GS (assuming it's announced)
[21:05] [s55] yeh
[21:06] [s55] hopefully and updated safari with faster rendering :)
[21:07] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[21:07] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[21:09] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[21:11] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[21:14] barf (n=Dark_Shi@193.69.144.164) joined #handbrake-dev.
[21:14] barf (n=Dark_Shi@193.69.144.164) left irc: Remote closed the connection
[21:14] barf (n=Dark_Shi@193.69.144.164) joined #handbrake-dev.
[21:17] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[21:21] [s55] Rodeo, that guy you banned earlier is posting trying to claim that a bt download is "Over the air" recording
[21:21] [s55] found it quite amusing
[21:22] [Rodeo] hmm, I have 800 posts
[21:22] [s55] Yeh, the number starts racking up pretty quick ;)
[21:22] [s55] hence HBA ;)
[21:23] [s55] every thursday
[21:23] [s55] when dyna isn't klined
[21:25] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[22:13] jbrjake (n=jbrjake@219-253.127-70.tampabay.res.rr.com) left irc: Read error: 60 (Operation timed out)
[22:31] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) joined #handbrake-dev.
[22:34] superdump (n=rob@unaffiliated/superdump) joined #handbrake-dev.
[22:34] #handbrake-dev: mode change '+v superdump' by HandBot!n=HandBot@concorde.lapsus.org
[22:35] mduell (n=mduell@rrcs-24-153-202-109.sw.biz.rr.com) left irc: Client Quit
[22:57] s55 (n=Scott@cpc1-livi1-0-0-cust180.sgyl.cable.virginmedia.com) left irc: "Leaving"
[23:14] niamu_ (n=niamu@ipgateway.quadro.net) joined #handbrake-dev.
[23:15] niamu (n=niamu@ipgateway.quadro.net) left irc: Read error: 60 (Operation timed out)
[23:16] niamu_ (n=niamu@ipgateway.quadro.net) left irc: Client Quit
[23:16] niamu (n=niamu@ipgateway.quadro.net) joined #handbrake-dev.
[23:21] ritsuka (n=Ritsuka@87.19.31.32) joined #handbrake-dev.
[23:22] #handbrake-dev: mode change '+v ritsuka' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[23:29] NightStorm (n=NightSto@adsl-76-233-131-96.dsl.toldoh.sbcglobal.net) left irc:
[23:39] superdump (n=rob@unaffiliated/superdump) left irc: "WeeChat 0.3.2-dev"
[23:43] NightStorm (n=NightSto@adsl-76-233-131-96.dsl.toldoh.sbcglobal.net) joined #handbrake-dev.
[23:43] #handbrake-dev: mode change '+v NightStorm' by BrakeBot!n=BrakeBot@rocky.lapsus.org
[23:45] ritsuka (n=Ritsuka@87.19.31.32) left irc:
up
Generated by logs2html module for eggdrop v.2.3.4
Find latest version at http://sourceforge.net/projects/logs2html or http://shmupsik.osetia.org