Tuesday, 2014-05-27

*** tpb has joined #timvideos00:00
aps-sidsmparuszewski: This site.url here should be changed to blog_url or whatever suits it.  https://github.com/pinoss/getting-started/blob/master/feed.xml#L1600:07
tpbTitle: getting-started/feed.xml at master · pinoss/getting-started · GitHub (at github.com)00:07
aps-sidsfeed gives wrong links.00:07
aps-sidsmithro: I know how to create stream of test producers using gui only and I'm not able to figure out or find relevant documentation about using flumotion-admin-text. Do you have any idea about using flumotion-admin-text? This is what I get when I use it. Doing startall, nothing happens.  http://picpaste.com/pics/flu-admin-screen-YKWpy8um.1401150994.png00:37
CarlFKaps-sids: I have never looked into how flumotion-admin/text interacts with the server.  I am pretty sure it does it via a network connection, so you should be able to run the client on another box or vm ?00:45
aps-sidsCarlFK: while porting some modules, the flumotion-admin (ui client) breaks as the modules used for ui are not compatible and they need to be ported as well. I want to setup a basic test stream but i can't use the gui since its broken. I'm talking about this - http://www.flumotion.net/doc/flumotion/manual/en/trunk/html/section-quick-start-starting.html00:55
tpbTitle: Starting Flumotion (at www.flumotion.net)00:55
aps-sidsbasically i want to do the steps done in this link using cmd line or some configuration file.00:56
mithroaps-sids: you need to learn to manipulate the xml config files directly01:36
aps-sidsmithro: oh, I get it now. you mean like the planet.xml file we use in manager?01:40
mithromanager.xml and worker.xml too I think?01:40
tpbTitle: streaming-system/tools/flumotion-config at master · timvideos/streaming-system · GitHub (at github.com)01:40
mithro /usr/local/etc/flumotion/managers/default/planet.xml01:41
mithro /usr/local/etc/flumotion/workers/default.xml01:41
tpbTitle: Configuration Files (at www.flumotion.net)01:42
mithroCarlFK: I got the screen from amazon02:29
mithro~seen ajit02:42
tpbmithro: ajit was last seen in #timvideos 10 weeks, 0 days, 11 hours, 57 minutes, and 54 seconds ago: <Ajit> https://docs.google.com/document/d/1akF6lBb78OWTGFosnYUC5-MxuNZXCrCob70ga3uQB8A/edit02:42
mithro~seen tija02:42
tpbmithro: tija was last seen in #timvideos 3 days, 12 hours, 22 minutes, and 15 seconds ago: <tija> Can anyone suggest me a way to measure the FPS of video streaming from HDMI2USB?02:42
mithroaps-sids: btw what timezone are you?03:16
aps-sidsmithro: GMT+5:3003:17
mithroDid you just get up very early or not gone to bed yet?03:17
aps-sidsDidn't sleep.03:19
mithroaps-sids: I'd recommend trying to keep decent hours, it's fairly easy to get into the habit of sleeping strange hours and it's not good for your health03:19
aps-sidsmithro: Actually, I slept last evening so wasn't feeling too sleepy. I'll keep this is mind and thanks for the concern :)03:23
mithroI have horrible sleeping habits myself03:28
*** CarlFK has quit IRC04:21
*** CarlFK has joined #timvideos04:22
*** ChanServ sets mode: +v CarlFK04:22
*** Niharika has joined #timvideos04:28
*** Niharika has quit IRC05:04
*** Niharika has joined #timvideos05:08
*** Niharika has left #timvideos05:10
*** Niharika has quit IRC05:11
*** slomo has joined #timvideos06:41
*** rohitksingh has joined #timvideos07:01
rohitksinghmithro: are you in VC? could you also send that invite to my gmail ID. Google is not letting me accept the invte07:02
mithrorohitksingh: Hi!07:03
mithrorohitksingh: I assume you were not going to make the meeting today07:03
mithroAyush is running late07:03
rohitksinghI can attend today and, I'm sitting ready for it! I'd resume my studies after the VC. No problem in that!07:04
mithrookay, give me a second07:04
rohitksinghGot the mail! Should we start or wait for Ayush?07:10
mithrorohitksingh: nah, we can start now07:11
mithroif we finish early, that is fine07:11
mithrojust trying to see if we can have a permanent URL for the hangout07:11
rohitksinghthis one is the URL right? https://plus.google.com/hangouts/_/google.com/gsoc-mentor07:12
tpbTitle: Google+ (at plus.google.com)07:12
mithrorohitksingh: no07:15
mithrorohitksingh: see private message07:17
*** tija has joined #timvideos07:55
mithroHi tija!07:59
mithrotija: I saw somewhere that it was taking quite a while for you to compile the firmware?08:00
*** ayush3504 has joined #timvideos08:00
tijathis is happening because I have changed the place and route strategy. So the ISE is putting "extra effort" meet all timing.08:01
mithrotija: how long is it taking you?08:02
tijamithro: Around 10 -12 mins. I am using a i5 4Gb ram laptop08:03
rohitksinghtija: mine is also i5 4GB 32-bit...took longer than 1 hours from synthesis to bitgen :( :O08:04
tijarohitksingh: I am pretty sure mine doesn't take that long08:05
rohitksinghtija: does Graphics Card accelerate the process? Because I have my ATI card disabled. Would it affect?08:07
tijarohitksingh: Don't know. You can try.08:09
ayush3504I'm yet to compile it. I'll let you guys know how much time it takes for me.08:09
rohitksinghalso PAR takes longest time...around 90% of the time...Let me try if I can increase speed!08:09
ayush3504tija, rohitksingh: which version/distro of Windows/linux are you using it on?08:12
rohitksinghUbuntu 12.04 has Xilinx 14.7 Installed, and Windows 7 has both 14.2 and 14.7 installed. Both are 32-bit08:13
JoelwI don't think ISE supports GPU acceleration08:14
JoelwThe tighter your timing constraints, the longer it'll take08:14
JoelwChipScope Pro can also make the process loads slower, and it'll make timing harder to achieve08:15
tijaDoes anyone know how I can force the jpeg core to encode at lower quality. In the documentation there are commands for the same.08:18
tijaJust complied the firmware. It took me 13 mins.08:25
rohitksinghtija: you have 32-bit or 64-bit os and toolchain?08:28
tija64 bit08:28
tijaI use ubuntu 13.1008:28
ayush3504rohitksingh: maybe its just the settings and timing constraints as joelw mentioned08:29
rohitksinghhmm...maybe, 1 reason is also 64-bit...and i5 also probably 3rd gen? I've 2nd gen i5...I'm also currently compiling. It took 5min 43 seconds for sysnthesis....1 min 29 sec for Translate...and currently doing map08:31
rohitksinghmap took around 18 minutes....currently doing par08:49
tijarohitksingh: are you using ISE GUI or makefile?08:52
rohitksinghwoah! par done...around 5 minutes...08:53
rohitksinghISE GUI...which one do you use?08:53
rohitksinghtotal time 30 minutes 8 seconds....atleast better improvement from last time :)08:54
rohitksinghtija: ISE GUI...which one do you use?08:54
rohitksinghI think i can bring it down further to around 15-20 minutes by changing process priority...going to try it08:58
*** tija has quit IRC09:04
*** rohitksingh has quit IRC09:05
mithrogah, they both left09:11
mithroshenki: ping?09:22
*** rohitksingh has joined #timvideos09:50
shenkimithro: pong10:19
*** ayush3504 has quit IRC11:02
*** ayush3504 has joined #timvideos11:19
*** ayush3504 has quit IRC11:29
mithroshenki: ha, Rusty's blog mentions you :)12:42
mithroshenki: you'll be all famous soon :P12:44
shenkimithro: i just fird up my feed reader to see if he'd published that blog post yet. he was writing it thisfternoon :)12:45
shenkimithro: i wrote a little android app for him to do the tests on native 3G (without having to teather)12:45
shenkinothing puts the pressure on like having rusty watch over your shoulder while you'r coding12:45
mithroha, start talking about ipchains, that'll distract him12:46
shenkimithro: oh, so i didn't realise tija had written another blog post12:48
mithroNeed to set up a RSS->email gateway? :P12:48
shenkimithro: given the frame rate goes up when he turns the compression up, i strongly suspect we are limited post-jpeg core12:48
shenkii signed up for newsblur after everyone was talking about it the other day12:48
shenkiso he's either being slowed down transfering from the core into ram, or ram into cypress12:49
shenkieither way, it's on his list for this week to measure the 'frame done' signal from the jpeg core, and see if it corrosponds to the frame rate12:49
mithroshenki: I don't understand the flow enough to know12:50
shenkii'll suggest he draws a diagram of what's going on12:50
mithrobut yeah, measuring usb FPS output seems like a bad signal to be using12:50
shenkiulitmately it's what we care about, but yeah, need to remove some of the layers12:51
shenkimithro: any news on the hdmi2usb boards?12:52
mithroLooks like we'll be sending off end of this week12:52
*** techdragon has quit IRC12:52
mithroI could use some help getting the firmware into a useful state to support both the Atlys and this board12:53
shenkiwill we require changes to both the FPGA and fx2?12:54
mithroI don't think we need any changes to the FX2 at all12:55
shenkithat's good news12:55
shenkiso just changing the ucf for the new pad configuration?12:56
mithroLet me finish publish this event notice and then I'll share with you what I have so far12:56
*** rohitksingh has quit IRC12:57
mithroshenki: the master spreadsheet is at https://docs.google.com/a/mithis.com/spreadsheets/d/10vNcsOAxnuiwc5diespjIepMySxhR0iVZfYxouq4p-E/edit13:02
tpbTitle: Google Sheets - create and edit spreadsheets online, for free. (at docs.google.com)13:02
mithroI've divided the firmware into "Modules"13:07
mithroThen for each NET that was externally visible I've mapped to a PIN13:08
mithroThe idea is that the Pins - ATLYS and Pins - NUMATO will be loaded from the schematics (I already have a tool which generates the required output from the KiCad schematic)13:08
mithroIt's the script which loads the other megaspreadsheet13:09
*** FeltonChris has quit IRC13:09
mithroWhich I'd like to split out into multiple parts13:10
mithroIE The section which does the comparisons between the LX16, LX25, LX45, LX25T, LX45T would be generally useful for anyone who wants to support all the chips13:10
mithroAlready chatted to Bunnie about it13:11
mithroSeeing if there is any possibility of the next revision of the laptop board supporting a T series13:11
shenkiT series?13:11
mithroHe said he'd look into it if I prove it works on my board13:12
mithroThe Spartan 6 have a "T variant" which has inbuilt high speed GTP transceivers13:12
mithroIt would allow us to do 1080p and other interesting things13:12
shenkiso our fpga is a t series?13:14
JoelwDoes the Artix 7 have normal I/O fast enough? Ugh, cheapest one is $103 on Digikey :)13:15
Joelwshenki: I don't think any of the reasonably priced dev boards have the 'T'13:15
mithroJoelw: yeah that is a big problem too13:16
shenkiso our board doesn't have a T series?13:16
mithroshenki: our board should support any of the LX16, LX25, LX45, LX25T and LX45T depending on which resistors are populated13:16
shenkimithro: ok. so the spreadsheet says the conference model will have the T, and the others don't13:17
shenkii understand now13:17
mithroshenki: that is the plan13:17
mithroshenki: which should mean that the conference model of HDMI2USB is also a decent board for doing prototyping with the GTP ports13:19
shenkimithro: i thought we could do 1080p with our board?13:20
shenkioh, no, it could almost do it. not quite though. i recall testing it now13:20
mithroshenki: no, the pins don't twiddle fast enough13:20
shenkihrm, if you open our ise project, it doesn't know where all of the files are13:20
shenkisurely other people have come across that in the past few weeks?13:21
shenkiWARNING:ProjectMgmt:430 - DesignStrategies - Strategy file C:/Xilinx/14.2/ISE_DS/ISE/spartan6/data/spartan6_performance_with_iobpacking.xds doesn't exist or is empty, skipping.13:21
shenkiyes ISE, that path doesn't exist. good work.13:21
mithroI haven't been opening it in ISE13:22
shenkineither had i, until just now13:22
shenkii wanted to see what xilinx's visualisation tools were like these days13:22
shenkithe last project i was on i used the altera tools to look at the schematic view a lot13:23
shenkiit was very useful for working out what hooks into where13:23
mithrooh, that reminds me13:24
mithroJahanzeb's brother was interested in seeing if he could port the HDMI2USB stuff to Altera13:25
shenkidoes he have a dev board in mind?13:25
mithroI'm not sure13:25
mithroHe is fairly new13:26
shenkiwhat does the system use the SPI pins for?13:31
shenkithe hdmi2usb system has some pads for a SPI port13:32
mithroThere is SPI for reading the config from flash13:32
mithroThere will be a SPI interface for talking to the PMIC on the Numato board13:33
*** rohitksingh has joined #timvideos13:34
mithroPower Management IC13:37
mithroIt's a PIC13:37
mithroit also has a bunch of peripherals and LEDs attached to it13:37
mithrowe'll need to write some firmware for it eventually13:38
shenkiPIC booo13:38
mithroit requires literally *zero* supporting components13:39
mithroand has a very stable voltage reference internally13:39
mithrothose where the two main reasons13:39
shenkiyeah. no arguments there. the toolchain is not ideal though, unless things have changed?13:40
mithrojust sdcc13:40
mithromost of the PIC firmware will probably be generated from a different spreadsheet14:02
shenkiyou should have been a civil engineer14:06
shenkithe only people i know who love spreadsheets as much as you are civil engineers14:06
mithroI find them a reasonable way to represent data14:09
mithrorefactoring code with no external differences in output is hard :/14:18
*** slomo has quit IRC16:04
*** Niharika has joined #timvideos16:13
*** Niharika has left #timvideos17:17
*** tija has joined #timvideos17:55
*** tija has quit IRC19:00
*** tija has joined #timvideos19:07
tariq786_labcomments and suggestions are welcome at hdmi2ethernet.blogspot.com19:08
*** slomo has joined #timvideos20:10
*** slomo has joined #timvideos20:10
*** aps-sids has quit IRC20:31
*** aps-sids has joined #timvideos20:40
*** tija has quit IRC20:51
*** tija has joined #timvideos21:18
*** aboutGod has joined #timvideos21:38
*** aboutGod has left #timvideos21:44
*** rohitksingh has left #timvideos22:02
*** slomo has quit IRC22:41
*** CarlFK has quit IRC23:14
*** tija_ has joined #timvideos23:19
*** tija has quit IRC23:22

Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!