| 01:26 | BogdanXor | joined the channel |
| 01:30 | BogdanXor | left the channel |
| 02:26 | BogdanXor | joined the channel |
| 02:31 | BogdanXor | left the channel |
| 03:00 | Spirit532 | left the channel |
| 03:27 | BogdanXor | joined the channel |
| 03:31 | BogdanXor | left the channel |
| 04:24 | jucar | left the channel |
| 04:28 | BogdanXor | joined the channel |
| 04:32 | BogdanXor | left the channel |
| 05:29 | BogdanXor | joined the channel |
| 05:33 | BogdanXor | left the channel |
| 05:49 | DGMurdockIII | joined the channel |
| 06:29 | BogdanXor | joined the channel |
| 06:34 | BogdanXor | left the channel |
| 06:42 | DGMurdockIII | left the channel |
| 07:22 | Alex_Chooks_ | joined the channel |
| 07:30 | BogdanXor | joined the channel |
| 07:34 | BogdanXor | left the channel |
| 07:45 | Bertl_oO | off to bed now ... have a good one everyone!
|
| 07:45 | Bertl_oO | changed nick to: Bertl_zZ
|
| 07:51 | sebix | joined the channel |
| 08:05 | se6astian|away | changed nick to: se6astian
|
| 08:06 | se6astian | good morning
|
| 08:31 | BogdanXor | joined the channel |
| 08:36 | BogdanXor | left the channel |
| 08:39 | sebix | se6astian, yesterday around noon we decided to stop the experiments unsuccessful experiments with the axiom beta
|
| 08:39 | sebix | I'll give a summary of what we did (not) achieve in the end via mail
|
| 08:55 | se6astian | right!
|
| 09:06 | sebix | Just found this: https://wiki.apertus.org/index.php/HDMI/SDI_Capture_PCIe_Card_Evaluation
|
| 09:06 | sebix | Are these *tested* or just a list of cards someone found in the internet?
|
| 09:07 | sebix | And I see the alpha supported YCbCr 4:2:2? https://wiki.apertus.org/index.php/AXIOM_Alpha_Software#HDMI_Output_Parameters
|
| 09:22 | se6astian | these are just found on the internet
|
| 09:22 | se6astian | correct, the alpha had a dedicated HDMI interface chip
|
| 09:23 | se6astian | now we drive hdmi directly from the fpga
|
| 09:23 | sebix | so there's no code which could be ported to the beta?
|
| 09:32 | BogdanXor | joined the channel |
| 09:32 | se6astian | correct
|
| 09:36 | BogdanXor | left the channel |
| 09:53 | sebix | left the channel |
| 10:13 | BogdanXor | joined the channel |
| 11:10 | sebix | joined the channel |
| 11:13 | Bertl_zZ | changed nick to: Bertl
|
| 11:13 | Bertl | morning folks!
|
| 11:16 | sebix | left the channel |
| 11:19 | Alex_Chooks_ | left the channel |
| 11:20 | Alex_Chooks_ | joined the channel |
| 11:25 | Alex_Chooks_ | left the channel |
| 11:31 | sebix | joined the channel |
| 12:12 | sebix | the atom feeds of the wiki are not valid, there are too much newlines: https://validator.w3.org/feed/check.cgi?url=https%3A%2F%2Fwiki.apertus.org%2Findex.php%3Ftitle%3DSpecial%253ARecentChanges%26feed%3Datom
|
| 12:43 | BogdanXor | left the channel |
| 12:44 | BogdanXor | joined the channel |
| 12:58 | Spirit532 | joined the channel |
| 13:10 | philippej | left the channel |
| 13:10 | philippej|away | joined the channel |
| 13:10 | philippej|away | changed nick to: philippej
|
| 13:20 | danieel | left the channel |
| 13:36 | danieel | joined the channel |
| 14:19 | Alex_Chooks_ | joined the channel |
| 14:42 | kaiserlich | joined the channel |
| 14:45 | sebix | left the channel |
| 15:33 | Bertl | off for now ... bbl
|
| 15:33 | Bertl | changed nick to: Bertl_oO
|
| 15:55 | sebix | joined the channel |
| 16:04 | se6astian | changed nick to: se6astian|away
|
| 16:20 | jucar | joined the channel |
| 16:34 | sebix | are github issues a good way to file bugs in e.g. mimg?
|
| 16:49 | kaiserlich | left the channel |
| 16:52 | se6astian|away | changed nick to: se6astian
|
| 17:09 | se6astian | changed nick to: se6astian|away
|
| 17:13 | se6astian|away | changed nick to: se6astian
|
| 17:19 | jucar | left the channel |
| 17:25 | sebix | se6astian, i have problems with the mimg 1.8 version you uploaded two days ago
|
| 17:25 | sebix | is this meant for the beta at all?
|
| 17:29 | se6astian | its widely untested so far
|
| 17:30 | se6astian | but I applied transparent overlays with it successfully
|
| 17:31 | sebix | x
|
| 17:31 | sebix | /opt/IMG/overlay_0*.rgb does not work
|
| 17:31 | rgon | joined the channel |
| 17:32 | rgon | It's been a long time since I last visited the IRC, so, first of all hi!
|
| 17:33 | rgon | I would like to know if it's late for an idea for the Gamma's form factor.
|
| 17:34 | rgon | Please tell me if this is not the place to post it, and if it would be a better idea to post it on the lab/Phabricator or another place
|
| 17:36 | rgon | Anyways, first of all congrats for your great work! Seeing all the development team has made is inspiring to say the least!
|
| 17:38 | RexOrCine | joined the channel |
| 17:38 | rgon | The idea is a simple concept, but having read/watched many camera reviews, a common trend I see is that given the great image quality cameras are achieving nowadays, form factor has become a deciding/important factor for many people who have had experience with proffesional video cameras before.
|
| 17:39 | mithro | left the channel |
| 17:42 | rgon | What I come to say is that the Apertus cameras do not need to be reserved to the "cine" market. If a simple consideration can push them into the ENG/TV world, that's a win-win and a huge customer base (as the Apertus cameras can essentially have a "controllable color science", live productions can really benefit from that, with custom in-camera output luts for example).
|
| 17:43 | sebix | rgon, we (me and some others) are trying to use the axiom beta for tv and similar use cases :)
|
| 17:44 | rgon | This is essentially one of my gripes with the Gamma's design. I know the primary objective for the Gamma is to have a powerhouse of a camera that can do everything. But if it's hard to carry on your shoulders, many will overlook it in favour of others.
|
| 17:45 | rgon | The only thing that I'm trying to say is that it would be a good idea to make the design a bit less tall (nearing a 1:1 aspect ratio height-length wise)
|
| 17:45 | mithro | joined the channel |
| 17:46 | sebix | See also https://wiki.apertus.org/index.php/Build_your_dream_camera#ajaxpoll-post-answer-FD2DC7B7977F99AD5592E19A6E7A63CD-3
|
| 17:47 | rgon | And thus, reduce the weight of the product. Take one of the Phantom Flex cameras as an example, they are boxy and utilitarian, but nevertheless, that doesn't influence in their usability at all
|
| 17:48 | rgon | @sebix That's cool! I always looked at the Apertus cameras as the ultimate in live video / Also thanks for the link, I'll post it right after I finish getting my thougts into words (to see what people think, rather than if they agree)
|
| 17:49 | se6astian | sebix: it worked with new images created via "convert input.png rgba:output.rgb"
|
| 17:49 | se6astian | but its possible the old images do not work anymore
|
| 17:50 | se6astian | hi rgon
|
| 17:50 | rgon | Another thing that would be interesting would be to add a "display" module. What do I mean with this, you may ask. I don't know if the beta body will have SDI/HDMI output without modules, but the idea is to create a "display module" that can be screwed to either the module or the body (getting power) so that the camera can easily be operated in an "unrigged" state
|
| 17:51 | rgon | I had a quick sketchup model testing the concept, whenever I get Sketchup running on my linux machine I'll send a picture if anybody's interested on what I mean by that
|
| 17:51 | sebix | rgon, by display you mean a small monitor?
|
| 17:52 | rgon | Yes, kinda. The hardware can be simple (well, in comparison). Just an FPGA receiving the TDMS HDMI signal (a Spartan can do) or an Artix (which has got transceivers) for an SDI input, and then a MIPI DSI LCD (cheap and readily available in 1080p+ resolutions)
|
| 17:53 | rgon | It's pretty much inspired from the Blackmagic URSA/URSA Mini cameras, but imagine that if it were a "module"
|
| 17:53 | rgon | A module that screws on top of another module, if you will
|
| 17:54 | rgon | The reasoning behind that is the same, added usability for "guerrilla"/live/TV stuff
|
| 17:55 | rgon | But the monitor would serve the simple purpose of being an in-camera remote control
|
| 17:56 | rgon | Sorry for writing in a block of text in such an anti-climatic way. Anyways, I just wanted to hear what people thought of that.
|
| 17:58 | sebix | se6astian, the parameter -D is mentioned in the help page but not implemented
|
| 17:59 | sebix | And the old images can be reused when converting them once back and forth
|
| 18:03 | rgon | ** When i said "i don't know if the beta body will have SDI/HDMI output" i mean the Gamma body, just to clear up.
|
| 18:05 | rgon | left the channel |
| 18:05 | rgon | joined the channel |
| 18:11 | Spirit532 | left the channel |
| 18:11 | Spirit532 | joined the channel |
| 18:50 | jucar | joined the channel |
| 19:24 | BogdanXor | left the channel |
| 19:32 | se6astian | rgon sorry for not replying here
|
| 19:32 | se6astian | I think the best place to make suggestions is the lab in deed
|
| 19:32 | se6astian | *indeed
|
| 19:32 | kaiserlich | joined the channel |
| 19:53 | Bertl_oO | sebix: yeah, the mimg needs a rework, it was quickly adapted for the overlays, but that broke some compatibility and features
|
| 20:03 | kaiserli_ | joined the channel |
| 20:05 | kaiserlich | left the channel |
| 20:07 | kaiserli_ | left the channel |
| 20:07 | kaiserlich | joined the channel |
| 20:07 | kaiserlich | left the channel |
| 20:07 | kaiserlich | joined the channel |
| 20:19 | kaiserlich | left the channel |
| 20:19 | slikdigit | joined the channel |
| 20:21 | BogdanXor | joined the channel |
| 20:26 | BogdanXor | left the channel |
| 20:27 | sebix | Yay, I managed to display a histogram as overlay :)
|
| 20:29 | alexML | nice :)
|
| 20:30 | sebix | Pretty useless tough, because I use the output of cmv_hist3 in a python3 script with matplotlib and then imagemagick to convert it to rgba... It takes 14s all in all
|
| 20:35 | sebix | Time to learn vhdl
|
| 20:54 | se6astian | still cool :)
|
| 20:57 | rgon | Oh, alright, I'll publish this in the lab ASAP
|
| 20:58 | rgon | BTW, I've listened to the latest team talk and noticed a developer "in the UK" is developing the Apertus Remote. Is he here?
|
| 21:00 | rgon | I was just about to suggest creating the UI of the remote in plain and simple HTML5+CSS+JS. This way, creating the "web emulator" wouldn't need any "emulation" at all, and the development of an Android app (or iOS/Ubuntu Phone/WP) would be quite easy.
|
| 21:00 | se6astian | rgon: no he doesnt use IRC
|
| 21:01 | rgon | ah, alright
|
| 21:01 | rgon | Anyways, I'll also post it in the lab just in case
|
| 21:01 | se6astian | sebix: I think you can get the histogram to be pretty much realtime without FPGA implementation
|
| 21:01 | se6astian | by checking the code how mimg writes overlay data
|
| 21:02 | se6astian | if you change just a small area of the image it will be much faster than writing 2 million pixel (fullhd)
|
| 21:10 | slikdigit2 | joined the channel |
| 21:10 | slikdigit3 | joined the channel |
| 21:11 | slikdigit2 | left the channel |
| 21:12 | slikdigit | left the channel |
| 21:13 | sebix | of course. And I also can "extend" cmv_hist3 to read the data from memory, process it and write the overlay.
|
| 21:15 | se6astian | exactly
|
| 21:15 | se6astian | https://github.com/apertus-open-source-cinema/beta-software/tree/master/cmv_hist3
|
| 21:15 | slikdigit3 | left the channel |
| 21:15 | sebix | jep, I'm aware of it
|
| 21:15 | se6astian | https://github.com/apertus-open-source-cinema/beta-software/tree/master/mimg
|
| 21:15 | se6astian | commited all Beta internal software to github just 2 days ago
|
| 21:16 | sebix | mimg has a bug (-D is not implemented, but listed in the manpag)
|
| 21:18 | Bertl_oO | patches are welcome
|
| 21:18 | sebix | I've no idea how to fix it yet^^
|
| 21:19 | sebix | but working on it
|
| 21:20 | slikdigit | joined the channel |
| 21:20 | slikdigit | left the channel |
| 21:20 | slikdigit | joined the channel |
| 21:22 | BogdanXor | joined the channel |
| 21:23 | sebix | concerning the interface for setting gain/exposure/overlay/mat4 etc, do I understand correctly, that the goal is to create library which can handle this for c, python etc. and also via ssh etc?
|
| 21:27 | BogdanXor | left the channel |
| 21:27 | Bertl_oO | yup, at some point
|
| 21:28 | sebix | the c code could then also provide the python bindings
|
| 21:30 | slikdigit2 | joined the channel |
| 21:32 | sebix | are there docs about the used registers and there purpose?
|
| 21:33 | Bertl_oO | the wiki has some information, but the only accurate source is the VHDL code
|
| 21:33 | slikdigit2 | left the channel |
| 21:34 | slikdigit | left the channel |
| 21:36 | sebix | e.g. for the overlay, there's the scn_addr = 0x00000000. Which one of the maps is the relevant one here: https://wiki.apertus.org/index.php/CMV12000_Register_Blocks
|
| 21:40 | aombk | left the channel |
| 21:40 | Bertl_oO | the base addresses can be seen in the bash functions
|
| 21:41 | Bertl_oO | set | grep -A2 reg
|
| 21:41 | Bertl_oO | for the registers themselves, it largely depends on what kind of registers they are
|
| 21:43 | Bertl_oO | if there is a separation between read and write, then the first addresses are typically read-only and the read-write ones start at offset 0x100
|
| 21:44 | Bertl_oO | the register size can be 16 or 32bits wide, (for LUTs and Matrix a special bitsize applies)
|
| 21:44 | aombk | joined the channel |
| 21:46 | Bertl_oO | https://github.com/apertus-open-source-cinema/beta-software/blob/master/cmv_hdmi3/top.vhd
|
| 22:13 | sebix | left the channel |
| 22:23 | BogdanXor | joined the channel |
| 22:24 | se6astian | I am off to bed
|
| 22:24 | se6astian | good night
|
| 22:25 | se6astian | changed nick to: se6astian|away
|
| 22:28 | BogdanXor | left the channel |
| 22:44 | rgon | left the channel |
| 22:53 | slikdigit | joined the channel |
| 22:53 | slikdigit2 | joined the channel |
| 22:54 | slikdigit | left the channel |
| 22:55 | slikdigit | joined the channel |
| 22:55 | slikdigit2 | left the channel |
| 23:00 | slikdigit2 | joined the channel |
| 23:04 | slikdigit | left the channel |
| 23:19 | slikdigit2 | left the channel |
| 23:25 | BogdanXor | joined the channel |
| 23:29 | BogdanXor | left the channel |
| 23:33 | arpu | joined the channel |
| 00:25 | BogdanXor | joined the channel |
| 00:30 | Bertl_oO | off to bed now ... have a good one everyone!
|
| 00:30 | Bertl_oO | changed nick to: Bertl_zZ
|
| 00:30 | BogdanXor | left the channel |