Studio Setup

From YSTV Documentation Wiki

Vision/Audio Path

SDI patches 91-94 are routed from the studio into cameras 1-4 on the ATEM Television Studio 4K in the control room.

The wire for camera 4 in the loom is dodgy and needs replacing, for now it's hard-patched. Marks.polakovs (talk) 11:21, 16 November 2022 (UTC)

SDI 95 is patched to one of the ATEM's non-program outputs, for use with camera control and tally (there's a SDI splitter on the studio side that runs it into the cameras, to avoid needing to run four more SDIs from the studio).

The ATEM's program out is patched into SDI 0 on edit2's DeckLink for broadcast (usually through OBS).

Audio comes in through the XLR patches into the X32 Compact, and its program out is wired into the ATEM's XLR inputs.

VT Playout

VTs are played using CasparCG on edit2. Caspar is configured with two layers: layer 1 outputs onto SDI 1 on the DeckLink, which is patched into "camera" 8 on the ATEM. Layer 1 also outputs audio onto edit2's default audio device (usually the headphone interface, which is patched into the X32 Compact). Layer 2, meanwhile, outputs over NDI - this is added in OBS as an overlay on all sources, usually for pre-rendered lower thirds and such.

With this setup there's a bit of delay on VT audio (very small - <100ms - but noticable) - my theory is that this is simply software audio latency (and also runs the risk of getting Windows system sounds unless we remember to mute them). This is, to quote Rhys, not ideal. Possible alternatives:
  • USB direct into the X32 - still has latency issues
  • Audio onto SDI direct into ATEM (set to audio-follows-video on Cam8) - means we can't mix VT audio on the X32, which is useful for levelling
  • Audio onto SDI, into a de-embedder, into X32 - spenny money (the Decimator can do it, but it's a bit overkill and a waste of a decimator)
Further tinkering is required. Marks.polakovs (talk) 11:21, 16 November 2022 (UTC)
Could also be that Caspar is configured to use too high a latency. Worth experimenting, but that path may end up futile - per the Caspar developers, system audio is intended for testing, and is "not in sync to the decklink". Marks.polakovs (talk) 12:06, 16 November 2022 (UTC)
More or less settled on using the Decimator MD-DUCC as an audio de-embedder, unless we decide at some point to get an actual de-embedder (which I'm not fully convinced is a valuable use of money). Marks.polakovs (talk) 14:37, 6 December 2022 (UTC)

Caspar is controlled using a Stream Deck running BitFocus Companion. As a nice bonus, this can be set up to trigger the VT, transition the ATEM, fade up the X32, and trigger OnTime, all with one button press.

TV Graphics

The TV in the studio (usually on Tom Lee) is connected (via a Micro Converter) into SDI patch 71Unverified or incomplete information, which is connected into the AUX output on the ATEM. This is usually set to show Media Player 2, but can be changed (press the AUX button on the top-right, under MENU, then select the output you want on the program row).

OnTime

OnTime is the stage timer system we use to manage time for live shows with a specific running order. It is a piece of software that usually runs on edit1.

The timer view for OnTime is displayed on the iMac on the control room monitor wall and the Mac Mini rigged in the studio lighting rig. This allows crew members to see the timings for the current production.

When the Macs are switched on, they should automatically display the timer view. See the OnTime page for more information on how this works and how it can be changed.