| *** tpb has joined #timvideos | 00:00 | |
| mithro-work | CarlFK / p2-mate: mjpeg works - it's just a bit slow, IE we can't do a full 30fps | 00:37 |
|---|---|---|
| mithro-work | p2-mate, there is a bunch of opensource firmware which replicates the fxusb firmware functionality - we just haven't gotten around to figuring it out yet | 00:39 |
| *** parx1 has joined #timvideos | 02:13 | |
| *** parx has quit IRC | 02:15 | |
| *** iiie has quit IRC | 02:25 | |
| *** parx1 is now known as parx | 02:30 | |
| *** hyades has joined #timvideos | 03:04 | |
| mithro-work | hello hyades - you where after me on the weekend? | 03:05 |
| *** hyades has quit IRC | 03:23 | |
| *** hyades has joined #timvideos | 03:24 | |
| *** hyades has quit IRC | 04:17 | |
| *** mithro has joined #timvideos | 04:42 | |
| *** hyades has joined #timvideos | 04:46 | |
| hyades | hi mithro-work | 04:53 |
| hyades | i wanted to know about integration tests | 04:54 |
| mithro | hyades, what about them? | 06:46 |
| hyades | suppose i am testing things like doing a switch, how will i confirm if it actually does? | 06:47 |
| mithro | hyades, good question | 07:20 |
| mithro | hyades, I'm not sure? maybe look at the output using gstreamer? | 07:23 |
| hyades | mithro: meeting! | 07:30 |
| mithro | hyades, you are correct it is meeting time, but I'm afraid I can't give you my full attention at the moment | 07:31 |
| hyades | should we have it some other time? | 07:32 |
| mithro | hyades, might be best | 07:32 |
| hyades | today or tomorrow? | 07:33 |
| mithro | probably tomorrow | 07:34 |
| hyades | ok cool | 07:34 |
| p2-mate | mithro: ah, so the problem with reaching full 30fps is in the fxusb fw? | 08:51 |
| mithro | p2-mate, no | 09:04 |
| mithro | p2-mate, the problem is with the mjpeg core we are using | 09:04 |
| mithro | it could easily be optimised but we haven't had the funds to do so | 09:04 |
| p2-mate | ah ok | 09:06 |
| p2-mate | I guess for slides it doesn't matter that much | 09:10 |
| p2-mate | mithro: any reason you do all the command parsing in the FPGA? wouldn't it be easier to parse the commands on the fxusb and just transmit the usb_cmd, jpeg_encoder_cmd, hdmi_cmd and selector_cmd? | 09:14 |
| mithro | p2-mate, the fxusb will be replaced with ah UTMI chip in rev2 | 09:15 |
| p2-mate | aha | 09:16 |
| p2-mate | I see... | 09:16 |
| p2-mate | and then the whole usb processing moves to the FPGA, but I guess you will add a CPU core then as well? | 09:16 |
| mithro | p2-mate, probably not | 09:18 |
| mithro | p2-mate, doing USB in hardware is pretty well understood | 09:19 |
| p2-mate | hmm, the entire usb stack in hdl? | 09:19 |
| mithro | yes, a couple already exist | 09:21 |
| p2-mate | I'm more wondering if it's efficient use of FPGA slices vs softcore + code | 09:22 |
| mithro | p2-mate, I guess we'll come to that issue when we need to | 09:23 |
| p2-mate | I see. I'm still playing with the idea of streaming to ethernet as that would make a lot of sense for us | 09:24 |
| p2-mate | but that would require a CPU, I don't think doing dhcp etc in hdl is a good idea :) | 09:25 |
| mithro | p2-mate, I think your probably better off just pairing the device with a Raspberry PI or similar | 09:26 |
| p2-mate | maybe, RPI is kind of crappy though when it comes to networking | 09:28 |
| p2-mate | it's USB implementation is slow and unreliable | 09:29 |
| p2-mate | the requirement would be proper USB HS support and gige, must sustain 40MB/s or so | 09:30 |
| p2-mate | but then you end up with chips which can also capture 12+Mpixels images and you wonder what the FPGA and USB is doing there :) | 09:31 |
| *** bananadine has joined #timvideos | 09:43 | |
| p2-mate | mithro: at 30fps you would need about 38MB/s USB bandwidth right? | 09:54 |
| mithro | https://github.com/timvideos/HDMI2USB/wiki/Resolutions | 09:55 |
| tpb | <http://ln-s.net/-k$b> (at github.com) | 09:55 |
| bananadine | mithro, are we going to have a video chat? | 09:56 |
| mithro | bananadine, I'm afraid I don't have time today | 09:57 |
| mithro | maybe tomorrow? | 09:57 |
| mithro | p2-mate, https://github.com/timvideos/hdmi2usb/wiki | 09:58 |
| tpb | Title: Home · timvideos/HDMI2USB Wiki · GitHub (at github.com) | 09:58 |
| bananadine | alright | 09:58 |
| bananadine | same time | 09:58 |
| *** bananadine has quit IRC | 10:03 | |
| *** mithro has quit IRC | 10:45 | |
| *** mithro has joined #timvideos | 11:34 | |
| *** hyades has quit IRC | 12:12 | |
| *** hyades has joined #timvideos | 13:16 | |
| *** hyades has quit IRC | 13:57 | |
| *** shenki_ has joined #timvideos | 14:10 | |
| *** shenki has quit IRC | 14:10 | |
| *** mithro has quit IRC | 14:10 | |
| *** mithro has joined #timvideos | 14:23 | |
| *** mithro has quit IRC | 16:03 | |
| *** iiie has joined #timvideos | 17:02 | |
| *** hyades has joined #timvideos | 17:44 | |
| *** iiie has quit IRC | 18:03 | |
| *** hyades has quit IRC | 18:04 | |
| *** iiie has joined #timvideos | 18:19 | |
| *** rihnapstor has joined #timvideos | 18:38 | |
| *** rihnapstor has joined #timvideos | 18:38 | |
| *** rihnapstor has quit IRC | 19:39 | |
| *** parx has left #timvideos | 20:41 | |
| *** parx has joined #timvideos | 21:41 | |
| CarlFK | iiie up for some streaming? | 22:34 |
| CarlFK | which is pretty much nothing for you now i think? | 22:35 |
| CarlFK | show starts in 2 hours | 22:35 |
| *** mithro-work has quit IRC | 22:35 | |
| iiie | sure! | 22:35 |
| CarlFK | yay. | 22:36 |
| iiie | CarlFK, what's the show / room name? | 22:36 |
| CarlFK | ps1 | 22:36 |
| CarlFK | psone ? | 22:36 |
| iiie | psone | 22:36 |
| iiie | that one is already setup : ) | 22:36 |
| iiie | I'll spin up mission an hour before ( an hour from now ) | 22:37 |
| CarlFK | k - I'll get dressed. | 22:37 |
| *** parx has quit IRC | 22:41 | |
| *** CarlFK has quit IRC | 23:20 | |
| *** CarlFK has joined #timvideos | 23:39 | |
| iiie | welcome back CarlFK | 23:42 |
| iiie | when the time comes the answer is ec2-54-227-83-142.compute-1.amazonaws.com | 23:43 |
| CarlFK | heh | 23:46 |
| CarlFK | do we have a command to test the collector yet? | 23:46 |
| iiie | I don't think so. Test what of the collector? We could swap in the test config. But that wouldn't be testing dv conversion. | 23:47 |
Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!