TriCaster VT[5] Hardware Audio

What's the best way to handle both Windows and VT audio as they're used?

Simply enable the on-board audio, and tell Windows to Only use default devices at the Sounds screen. That way, Windows will only use the sound card for its sounds, and the VT audio will take over in VT.

What are VT[4] audio offsets for?

All digital equipment that records audio as integer numbers has a maximum allowable level before which it just "hard clips" the audio. In contrast, analog equipment tends to allow you to push the level quite high and just starts to exhibit non-linear artifacts that get worse and worse until something actually catches fire (note that the guitar overdrive effect is actually this non linear distortion occurring, which has now become a desirable thing if you are a hard-rocker.)

Digital equipment has always placed the maximum level that it can record at a +0dB level, which means that in theory if you go above that you will get horrible clipping. Because this is somewhat undesirable, what has been done is that all signals that are recorded on digital equipment are typically dropped in level, normally by -12dB (or -20dB) so that a 0dB level going into the gear is now actually recorded at -12dB (or -20dB) instead. This then allows levels up to +12dB (or +20dB) to be recorded before there is distortion.

It is common terminology to call the "reference level" of the file as being the level at which "0dB" really is. So a "-12dB" reference level means that 12dB of dip was applied when recording the file, meaning that "real 0dB" is now at -12dB on the tape.

So, since when you recorded a file you applied 12dB of dip to the level, if you want it to play back and have the right volume, you must do the exact opposite and apply 12dB of gain.

Unfortunately, there is no one standard on whether -12dB or -20dB is applied, and in "generic" digital audio files there is just no way to know how much dip was applied to the file. This is why in VT we allow you to set the level for the file ... Basically, when you play it back you have to set what level the file is at, which "undoes" any dipping that was applied to a file when creating or recording it. Likewise, when you "render" you need to specify what the "dip" in the file is so that you can ensure that you have enough head-room to avoid clipping.

Some final guidelines ... the most common reference level for digital files is -12dB, which means they have been dropped in volume by 12dB to avoid clipping. This means on playback we boost it by 12dB and upon rendering we need to re-apply a 12dB dip. This applies to most consumer DV and most digital audio files.

Some professional decks use a -20dB reference level, so if you audio is way to soft (8dB to soft) then you will need to set your reference levels accordingly.

Finally, if you use floating point audio, the world is all good, since there is no clipping, there is also no need for reference levels.

My audio starts crackling and then dies. Is this the software or has my VT card died?

There is a way to determine if it is the hardware:

  1. Turn on an analog audio source so live (external) audio is passing through
  2. Exit VT
  3. Kill WinRTME

Then, answer these questions:

  • Do you still hear the audio passing through?
  • Is it distorted?
  • Is there ticking?

This leaves the VT card in an un-monitored hardware pass-through state. If you hear the audio passing through while in this state, and it later fizzles out then the problem is most likely hardware as there is no processing being done on the audio data at all.

If, however, killing VT/RTME will make any audio passing through come back ON, then it is most likely a software problem. where the internal state of the RTME loses track of the hardware's progress. There have been some device drivers (video cards, disk controllers, leaving a CDROM in PIO mode instead of DMA) which really wreak havoc on the computer's overall real-time performance.

What is the difference between floating point and 16-bit audio?

The audio file bit-depth refers to processing accuracy. You should not hear any difference in 16-bit and Floating point. Floating point audio just affords more head-room during recording. While some audio apps don’t yet recognize floating-point audio, 16-bit will be readable by any other application.