Tuesday, 2015-04-21

*** tpb has joined #timvideos00:00
*** techman83 has quit IRC00:18
*** techman83 has joined #timvideos00:18
*** ChanServ sets mode: +v techman8300:18
cfeltonMaZderMind: might be easiest to route the I2C signals to a header (directly) versus cutting up a cable, break out as scl_i, sda_i, and sda_o01:22
CarlFKcfelton: low hanging fruit that is kinda needed https://github.com/timvideos/HDMI2USB/issues/10201:52
tpbTitle: Map 720p and 1024x768 to a hardware switch · Issue #102 · timvideos/HDMI2USB · GitHub (at github.com)01:52
CarlFKmithro: is this the feature? related to getting some diagnostics that might help with the edid problem?  https://github.com/timvideos/HDMI2USB/issues/4401:55
tpbTitle: Connect the USB-UART connector to the control system inside the firmware · Issue #44 · timvideos/HDMI2USB · GitHub (at github.com)01:55
mithrocfelton: they are already available via a header03:50
mithro(on the Atlys board)03:59
tpbTitle: Digilent Atlys Board: Loading Firmware · timvideos/HDMI2USB Wiki · GitHub (at github.com)04:01
*** Niharika has joined #timvideos05:18
*** Niharika has quit IRC06:16
*** Niharika has joined #timvideos06:18
*** hyades__ has joined #timvideos07:27
*** hyades__ has quit IRC07:27
*** hyades__ has joined #timvideos07:28
*** hyades__ is now known as hyades07:34
*** Niharika_ has joined #timvideos08:22
*** Niharika has quit IRC08:22
*** Niharika_ is now known as Niharika08:25
MaZderMindjust re-implemented the gst-switch video comparison routines - i think the tests are pretty readable now: https://github.com/MaZderMind/gst-switch/blob/int-tests-rewrite2/python-api/tests/integrationtests2/test_video.py#L8209:47
tpbTitle: gst-switch/test_video.py at int-tests-rewrite2 · MaZderMind/gst-switch · GitHub (at github.com)09:47
MaZderMindand they don't fail randomly on travis anymore09:47
MaZderMindsome parts are still missing but not more then a weekend worth of work09:47
*** hyades has quit IRC10:06
*** hyades has joined #timvideos10:06
mithroMaZderMind: I'll try and have a look at it tomorrow10:20
mithroMaZderMind: please ping me if I haven't responded by the end of the week10:20
mithroIf I could do Debconf, FrosCon and CC-Camp with the aim of working on the TimVideos stuff - that might be enough to get me over.10:22
mithrobblr, heading home10:23
*** Niharika has quit IRC10:33
MaZderMindmithro: I'll finish the Tests, reword the READMEs and the PR and come back to you then10:37
*** Niharika has joined #timvideos11:37
*** Niharika has quit IRC11:42
*** Niharika has joined #timvideos13:38
CarlFKmithro:  (bump) is this the feature? related to getting some diagnostics that might help with the edid problem?  https://github.com/timvideos/HDMI2USB/issues/4414:32
tpbTitle: Connect the USB-UART connector to the control system inside the firmware · Issue #44 · timvideos/HDMI2USB · GitHub (at github.com)14:32
cfeltonCarlFK: in general it should, you could include a parallel EDID dump to the UART14:45
CarlFKcfelton: good morning14:45
cfeltonCarlFK: hello14:45
cfeltonin this case, using the USB-UART would only be somewhat beneficial14:46
cfeltonit would be more of a test case for the diag port14:46
CarlFKEDID is just a lump of static bytes that get sent.  ... yeah never mind that.14:46
cfeltonyup, looking at the HDL my guess is an odd timing thing, like missing a i2c restart or something14:47
CarlFKI have a "LENOVO ThinkPad Twist 33474HU" with 2 external ports: mini hdmi and display port14:48
*** hyades has quit IRC14:49
CarlFKdisplay port to hdmi dongle, to atlys, no problem.  like I don't think it has ever malfunctioned14:49
cfeltonthe irony, more complicate - more reliable :)14:50
CarlFKmini hdmi to atlys, malfunctions about 1/2 the time, with what looks like 5 or 6 flavors14:50
CarlFKbut it also jams up my box so trying to collect stats and figure out what is going on gets difficult14:50
CarlFKit might help if I shut down X14:51
*** _1_playboy has joined #timvideos15:59
*** Niharika has quit IRC16:30
*** Niharika has joined #timvideos16:40
*** Niharika is now known as Niharika_away17:34
*** CarlFK has quit IRC18:05
*** CarlFK has joined #timvideos18:58
*** ChanServ sets mode: +v CarlFK18:58
MaZderMindcfelton: i thought also about a timing issue or clock jitter. should be very obvious on a scope20:08
cfeltonMaZderMind: i2c is slow and source sync, there more than likely will be jitter :)20:09
cfeltonwhat is probably the most interesting is the slew rates but you would need to look at the physical signal (not routed to a pin)20:10
cfeltoni2c is open-collection, pulled low by the FPGA driver and then high by external pull-up resistors20:10
MaZderMindhm that would require a hdmi interception board that pulls out the different signals20:10
MaZderMindor a dismanteled cable20:11
MaZderMindbut it's all thery until one measures it ^^20:12
*** tariq786 has quit IRC20:14
CarlFKMaZderMind: did you send mithro your address?20:16
cfeltonMaZderMind: SCL and SDA are not diff signals in the connector20:18
cfeltonMaZderMind: sorry, misread20:18
cfeltonyes, would require dismantle cable or a breakout connector20:19
MaZderMindCarlFK: yes, the package is on the road20:19
MaZderMindcfelton: i did that for ethernet too to measure the levels (just out of curiosity), will do that for hdmi, too, when the board arrived20:20
MaZderMindbut i guess i'll need some weeks to get into fpgas and the whole suit of tools required20:21
*** Niharika_away has quit IRC21:01
*** shenki has quit IRC21:36
*** shenki has joined #timvideos21:37
*** ChanServ sets mode: +v shenki21:37

Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!