xfxf / av-foss-stack Goto Github PK

See if it copes, or if the disk writer craps out. Would be super useful to sync talks constantly.

This is after Ansible setup is done.
We are using (including 1x spare):

MIXING + CAMERA CAPTURE MACHINES

• 5x Lenovo W530 (Next Day Video)
• 1x Asus 17" (Next Day Video)
• 1x Desktop (Next Day Video)

PRESENTER LAPTOP CAPTURE

• 7x Lenovo i5 laptop (LA)

Each room will have 2 laptops, with possible exception of Keynote venue using 3 if we choose (extra camera).

We also have a spare desktop machine, Ryan's Lenovo Yoga (do not wish to reformat), a bunch of other Lenovo i5 laptops (LA) + Lenovo Core 2 Duo Laptops (LA).

Will be Sunday. @xfxf to draft an email for Chris to send out.

5PM in one of the rooms.

Previously I'd just go look in /var/log/

I don't really know how to do the same thing with systemd....

Here's a ticket to remind him that we're using a delay of 800ms, to make things stay in sync.

@mithro: You wanted to follow up on this later...

Is definitely confusing presenters, as expected. Approval email is a good idea, but needs better context.

During Wrest Point testing, the Opsis would not output successfully to the Blackmagic gear they have in the Plenary hall (ATEM TV Studio, and separately a Blackmagic HDMI to SDI converter they use for running long lines).

No video is detected / shown. Tried all 720p and 1080p resolutions available.

@xfxf has identical equipment with him at the Hackfest that can be used for testing.

We should be doing ~25fps everywhere, from the Opsis, to vocto ingests, to vocto, to saving, to encoding (veyepar melt options). Suspect we're doing 25<>30 conversion in a few places currently.

IF booting up one of the LCA laptops without the Opsis plugged in, it takes forever to boot, due to systemd trying to constantly start hdmi2usbd/hdmi2usbmond. Should be dependent on an actual Opsis device being plugged in.

Wrest Point uses Panasonic 16:10 (native res 1920x1200) projectors with a Panasonic ET-YFB200 box which takes various inputs (including HDMI).

This works fine with the Opsis at 720p50h, but 720p60 has significant chopped off pixels on the left.

AV guy tried to adjust settings try a second YFB200 box, nothing fixed it. I have observed the same ~2 years ago on early HDMI2USB firmware with an Atlys.

Do we fix this, or do we run 720p50 as the output resolution @mithro?

Soundies often play music between sessions which is commercially licenced, which causes YT streaming or videos to get canned with copyright notice.

Put together a nice playlist of appropriate CC-licenced music that can be used.

LCA is using Symposium this year, which isn't zookeeper.

https://github.com/timvideos/streaming-system/tree/master/website
streaming-system/website/frontend/xxx2internal.py

Consider smarter/less manual monitoring process of each room. Primary things to watch for:

• Audio (levels, feedback, mics are on properly, etc)
• Switching (appropriate laptop/camera swaps)
• Camera use (zoom, panning, etc).

Something like this would be good (need to enable preview streams in voctocore):
https://github.com/voc/multiview-monitor

We need an additional set of hands to help us test Opsis boards before the conference.

This would be following the instructions + filling in the spreadsheet listed in #11

Have made a few callouts for help, yet to get anybody.

Test new 'nextgen' firmware to ensure it will capture from required devices (various laptops, cameras).

@mithro to detail a plan for this.

• Milk

Allow teams to edit their talks after recording. Will provide a 'feedback loop' of what they do during recording translates to editing (so better cuts, better audio levels, etc).

Will require appropriate architectural changes to encoding/storage.

@CarlFK @stefanor please move this to 'done' once both machines appear to be working OK (ensure voctomix runs on these, capture works OK, audio output works, etc). and there's no issues rebooting them.

The desktop machine with the Gigabyte motherboard was not booting up properly during Pycon AU, freezing within minutes, and audio out was not working. I thought it was bad capacitors, but it might just be a faulty video card causing issues - I've removed it, it turned on fine (it wasn't turning on before).

The desktop machine with the Asus motherboard is solid, I use it daily, but it doesn't boot up the first time with 4x RAM sticks in it. I've removed 2x and it appears to be fine. Only test here is ensuring it always turns on.

Want to know this earlier rather than later, as I'll buy replacement motherboard(s) if required.

The last time we used voctomix (pycon australia), we had some A/V sync issues with the voctomix gui. Some remote sources also had to be delayed slightly (see pycon au scripts in xfxf/video-scripts).

This seemed slightly different on the various laptops (i.e. the i7 Lenovo's exhibited different behaviour than the Asus 17").

Once voctomix is running we REALLY need to setup a full room and leave a test recording going + ensure all sources (camera, laptop input, audio) + ensure nothing gets out of sync, including voctogui.

Including schedule import, ensure all data OK.

http://pro.jvc.com/pro/attributes/4k/manual/GY-HM200_170EN_basic.pdf

Currently https://docs.google.com/spreadsheets/d/1uc_QICAHcZtQpQxt8bpTy2hKk7BN11h375hLQG0wcoA/edit#gid=441420403 says we need 17 opsis boards.

I think we have ~10 with cases + power supplies....

Test a new Macbook with USB-C dongles (3 types).

Seems fine on laptops, desync's a few times a day on machine used in Plenary (older i5)

Spreadsheet (results): https://docs.google.com/spreadsheets/d/1W3p1J-Lb6l-sWmh6n9-YJ_U5o7a0pDmWAR6QIKfCzI0/edit#gid=0

For #6, we wish to use hdmi2usbd. @deeprave to ensure this works for the usecase of sending messages (such as status/debug) and reading data back to parse using a python client.

https://keep.google.com/#LIST/159767e5ec5.a4d56a007c584930

Sometimes the uvcvideo driver gets stuck waiting for the HDMI2USB firmware to respond but it never will. Removing the module and adding it back will reset these timeouts getting things working again faster.

This also works around the problem where the serial port is opened before the video is opened.

If MSY don't have 6, get up to 6 (whatever stock they have). Need by early arvo Sunday (for setup - these are for front laptops to be used to display LCA logo/video/whatever between sessions). This will go into the LA kit (they're for the LA i5 laptops to do video out).

Full male displayport > female HDMI.

Need short dongle/adapter, not full cable. Cost should be <$16 each.

http://www.msy.com.au/tas/glenorchy/cables/16937-unitek-y-6332-displayport-to-hdmi-female-support-4k-converter.html

or

http://www.msy.com.au/tas/glenorchy/pc-accessories/16735-volans-vl-dphm-displayport-to-hdmi-m-f-converter.html

End goal is to PXE boot + install Ubuntu 16.04 LTS (with all updates).

Should easily be able to associate a new laptop to a particular room role.

Don't erase the existing imaging we have working on the server (so we have a backup option).

Should include:

• All 16.04 LTS updates (latest gstreamer updates)
• Latest voctomix
• Ability to run all required voctomix components (1x remote laptop w/ HDMI2USB, local laptop w/HDMI2USB+ALSA device or Blackmagic, cut script, video capture script)
• Support for blackmagic cards (drive installed)
• HDMI2USB flashing tools (so @mithro can ssh in and reflash Opsis's if required)
• Encoding tools + configured against veyepar,'s db so can be used as encode stations

Should include @joeladdison's monitoring software too later in the week once this is ready.

@CarlFK to print recording sheets on Sunday before training.

This spec is split out into a super basic spec (minimum spec), and then a second spec of the tool we'd like to be using. The latter is not meant to be a full 'wishlist' spec, just the minimum of what's realistic for v1.

The problem this is meant to solve: during conferences, occasionally we have issues with HDMI2USB + laptop compatibility. We rarely get useful debug information to solve this as the priority during the event is just get the laptop working (which might end up being swapping with a different laptop, use a different dongle, etc). We need to log this information.

Secondly, we'd love for this information to be central, displayed on a single dashboard we can show in AV NOC, to be able to solve problems faster. Statuses should dynamically update using JS - i.e. we can see when people plug in / unplug laptops, the current version of the firmware flashed on it, the current mode the device is configured in, the status of the recording machines (is voctomix running, hard drive space left, etc) with appropriate alerts occuring.

MINIMUM SPEC

Write a daemon that connects to the Opsis (via serial using HDMI2USBd, or via a TCP connection if @mithro prefers this), enables debug mode, and dumps this with timestamps to a local file. Ensure new V4L connections to the webcam device (i.e. voctomix starting a capture) isn't interfered by this.

SECOND SPEC

• Daemon should be able to parse events (list events we care about below? including getting current firmware info + status of input/outputs/encoder), and understand what's happening.

• Daemon should push these events via REST back to a app running on a central AV server (with an appropriate standarised scheme that describes the source machine, etc).

• A simple dashboard should be developed that reads these events and displays them.

• Dashboard should understand various 'states' of components, so if an error /warning condition occurs, it can appropriately display this on the dashboard.

• Dashboard should be extended to report things such as hard drive space, status of various voctomix components, machine reboot events, etc (list other things on the machine we care about as a minimum/)

https://docs.google.com/spreadsheets/d/1uc_QICAHcZtQpQxt8bpTy2hKk7BN11h375hLQG0wcoA/edit?disco=AAAAA1WOp8Y&usp=comment_email_document

Cutting to a file + recording sheets is a little messy. Integrate this so it can be automated as part of the UI.

• correct licence
• allow embed

Add to this if we need to locate/buy things before LCA.

Back/front real-time comms, left/right podium

potentially using a Behringer X32 desk, or another desk that can be probed remotely using automation.