Abstract

FOSDEM rocks and the Debian Dev room has a lot of interesting talks, but one cannot possible attend all of them. But there is the videoteam which tries to provided good quality coverage, and possibly a live stream! Again we will be using a multiple camera setup and Ben Hutchings' DVSwitch software.

This page is basic "scratch space" for the organisation of this, as we won't have anywhere near the amount of gear or people we had at DebConf.

FosdemVideo2008 has useful historic info.

Volunteers

Please add yourself to the table below if you wish to help out. Also make sure you are subscribed to the debconf-video@lists.debconf.org mailinglist.

Name (irc nick)

dates available

[:HolgerLevsen:Holger Levsen] (h01ger)

saturday and sunday

Miguel Gea (xerakko)

saturday and sunday

[:DannyCautaert:Danny Cautaert] (DaCa)

saturday and sunday

Sebastian Harl (tokkee)

saturday and sunday

Ben Hutchings (Womble2)

saturday and sunday

Schedule

Saturday 2009-02-07

Time

Event

Main camera

Audience cam

dvswitcher

audio mixer

Sat 13:00-14:00

Outside broadcast on a budget - the DebConf video team and DVswitch Ben Hutchings, Holger Levsen

DaCa

Sat 14:00-15:00

Ultimate Debian Database: datamining Debian made easy! Stefano Zacchiroli, Lucas Nussbaum

Womble2

h01ger

Sat 15:00-16:00

Introducing DDE, Debian Data Export Enrico Zini

DaCa

Womble2

Sat 16:00-17:00

The Debian status quo on the Openmoko Neo Freerunner Luca Capello

DaCa

h01ger

Sat 17:00-17:30

Running Debian on Inexpensive Network Storage Devices Martin Michlmayr

DaCa

h01ger

Womble2

Sat 17:30-18:15

Grid Computing with Debian, Globus and ARC Anders Wäänänen, Steffen Möller, Mattias Ellert

Womble2

Sat 18:15-19:00

What does the DPL do? Steve McIntyre

h01ger

DaCa

Womble2

Sunday 2009-02-08

Time

Event

Main camera

Audience cam

dvswitcher

audio mixer

Sun 09:30-10:00

Debian Font Task Force: overview and impact on the open font community Nicolas Spalinger

h01ger

Sun 10:00-11:00

TDebs Neil Williams

h01ger

Sun 11:00-12:00

Internationalization in Debian: How to improve further? Nicolas François

DaCa

Womble2

Sun 12:00-13:00

The Common Debian Build System (CDBS) Peter Eisentraut

Womble2

Sun 13:00-14:00

Release management in Debian - can we do better? Frans Pop

h01ger

Sun 14:00-15:00

Lenny - the road to release Neil McGovern

h01ger

Womble2

Sun 15:00-16:00

The long road to KDE4 in Debian Sune Vuorela

h01ger

Womble2

Sun 16:00-17:00

Debian and Google Summer of Code 2008: wrap-up and insights Obey Arthur Liu

Womble2

Hardware

Required:

Available:

Network

Assuming that our room will be the same as last year (AW125), this means that there /probably/ will not be any FOSDEM network equipment in the Debian room. This is not for sure yet, it might also be that FOSDEM will get part of the uplink bandwidth from the plug in the Debian devroom...

According to the FOSDEM website, the room will be AW1.121, which appears to have 5 more seats than the AW1.125

Don't worry, there is of course a solution for both cases:

No FOSDEM equipment is present: you can plug straight into the uplink port that is provided by the ULB in your room. It might be BNC, but Gerry will have hubs available that do BNC->RJ45, so no worries there.

   Settings (fixed):
     address 164.15.3.4
     netmask 255.255.255.0
     gateway 164.15.3.254

As a DNS server, use whichever you like, but make sure it answers to requests from your source IP.

FOSDEM equipment is present: I will make sure there is a port available for your wired uplink. Configuration will be provided by DHCP.

= Sound =

I (daven) am regrettably not available to come to this FOSDEM. Should you need assistance / troubleshooting, feel free to call.

I suggest that the sound for this event is setup as follows:

Inputs

Channels 1 - 3: Condenser microphones taped from ceiling over audience (left, right and center). Pan left and right mics to 10 o'clock and 2 o'clock positions on the pan dials. Reduce input gain for the center channel by 2 dB with respect to the other mics.

Channel 4: Spare condenser, have this on the desk at the front, you never know when it will come in handy

Channel 5/6: Headset receiver 1 (connect using 1/4" jack to input marked L; this monos the stereo channel). Also remember to set the receiver to +4dBu output.

Channel 7/8: Headset receiver 2 (same as above)

Outputs

Headphones (monitor what you're doing!)

Main output: Send to Twinpact (make sure correct input is selected on the Twinpact; select this via switches on the bottom of it)

FX send or Aux: Send (preferably via DI box) to camera.

Camera Input

The only camera with a sound input seems only to have a microphone input. This is suboptimal; because it both runs a possible risk of damage to the mixer (if there is plug-in power on the mic input); because it will be noisy as hell; and because the level expected is very very low, leading to a very easily clipped input.

I would highly suggest using the Twinpact input for a number of reasons, most relevant of which as follows:

If it is absolutely necessary to use the mic input on the camera, workarounds as follows should be used to make this work:

If a DI box is available

FX output from the mixer -> cable (1/4 inch mono jack to 1/4 inch mono jack, 5m long) -> DI box -> cable (3 pin female XLR to mono 1/8 inch mono minijack, 2m long) -> camera microphone input.

If a DI box is unavailable

FX output from the mixer -> cable (1/4 inch mono jack to 1/8 inch minijack, 5m long) -> camera microphone input.

You will be using the FX send bus (or Aux send bus) on the mixer. The reason for this is threefold:

a: The aux send bus on the mixer is sent post-fader, and allows additional attenuation for each source (invaluable when attempting to go from line level down to mic level)

b: You would rather damage the Aux send bus by incorrectly designed mic power than you would the main outputs

c: The main outputs are in stereo; if you route these to the Twinpact, you could still do the production in stereo; whereas the mic input will only be mono.

Mixing

Setting up initially using the FX bus is slightly different, but should be used to feed the camera. Operation is as folows:

Above each channel pan/balance knob on the mixer, you will the FX send for that channel. Start by turning all of these all the way down.

Make sure the main output is connected to the Twinpact, and cut to the Twinpact sound input on DVSwitch. Increase the master volume of the mixing desk to unity (0 dB), and fade up the volume of a condenser mic channel to unity as well.

Assuming the condenser mic is hanging over one of the rows of seats, get someone to stand underneath, and speak as if they were asking a question to the presenter. Adjust gain on that channel until there is an indication of signal on the VU meter on DVSwitch. You want this to be around -24dB on average, peaking at around -6 dB. Higher than this, and you risk clipping due to audience / handling noise, as you are not using a dynamic compressor.

Now to adjust level to the send for the camera input. Get the person standing under the mic to continue speaking; or to sing a single steady note if possible. While doing this, switch the DVSwitch sound source between the Twinpact and the recording camera; and SLOWLY increase the level of the FX send until both levels are roughly similar. If you are not using a DI box, you will not need very much level here at all. Once you have the level for that channel, apply the same level for the FX send to all the other channels. The FX send bus is post fader, so you do not have to worry about running the mix twice, and can operate as usual.

Once you have this set up, proceed to adjust gain for the other channels as usual.

Sound gear and cables required for this setup

Schematics

attachment:fosdemvideo.png

attachment:fosdemvideo.dia

ToDo