Frits
When doing automated mixing, the level goes only downward. So if I want to have a swell on a track, I have to work backwards to determine what to set the track gain to so that when the automation hits 100% I get the swell I was after [also working out where to put the fader so that the start level is correct].
Is there a chance that either:
have the level automation start at 50%, giving a margin for up as well as down. ie 50%=0dB change
OR
since this would mean changes to all current projects which use level automation, have an option to set what %=0dB, with the default being 100% (as it currently is).
Or have I missed something somewhere? I’m currently in an internet cafe and have been pondering whilst wandering around the shops. Can the max/ min/ default be altered within the device spec?
Also, I have been in the habit of making most tracks have audio mixing, so I can adjust the levels all the way through the signal chain. Is there much of a processing leap in doing this, or is the only code added in the form of signal==signal*gain;? (I realise that it increases the graphical overhead too..)
Finally… If a track overloads, does Podium clip, or just pass the overloaded signal down the chain to be attenuated later? I’ve noticed that if (for example) I have a snare hit that goes red in it’s channel, the parent track goes less red, and the master channel stays green; and all of them have 0dB gain… [btw, I love the way you have done the VU colouring, it’s really intuitive]
There was something else, actually.. but I’ve forgotten what it was!! I write when I remember..
DSP
Traditionally most hardware (and software) mixer faders will allow you to boost the level of a track with up to +6, +10, +12 or +18 dB. This is necessary because on many mixers the faders are the only means to boost a signal on a track. I chose a different approach in Podium by implementing a gain setting for the track and let level automation be an attenuation of this gain setting.
I am pleased with how this split into a gain setting and level automation has turned out. You can adjust the gain setting on audio tracks and avoid creating a level track if you don’t want to automate the level. And for tracks with level automation, you can adjust the gain setting to ‘offset’ the entire level automation on the track.
So if you want to boost an audio track with +6dB at some point, you need to set the gain of the track to +6dB, and then start the level automation at -6dB. I’m going to improve the curve editor soon, so that it will be possible to see dB values for the curve points.
If it turns out to be a huge demand that level automation tracks should be offset with +6dB, one way of supporting this would be to define a new ‘+6dBFS Level’ parameter type in the Parameter properties dialog. Existing projects would then not be affected.
Also, I have been in the habit of making most tracks have audio mixing, so I can adjust the levels all the way through the signal chain. Is there much of a processing leap in doing this, or is the only code added in the form of signal==signal*gain;? (I realise that it increases the graphical overhead too..)
Every track that you audio enable will get an extra sub-mixer at that track. Meaning that there will be a buffer used for collecting the audio of subtracks, to calculate the meter values and to apply pan and level changes (if set differently from defaults). Tracks that have e.g. plugins assigned will already have the sub-mixer created, so audio enabling those tracks to get the gain setting will not add overhead to the audio processing.
Finally… If a track overloads, does Podium clip, or just pass the overloaded signal down the chain to be attenuated later? I’ve noticed that if (for example) I have a snare hit that goes red in it’s channel, the parent track goes less red, and the master channel stays green; and all of them have 0dB gain…
Podium does not apply clipping, until it has to output the floating point audio to fixed point buffers in audio interfaces or recording to fixed point sound files. Some plugins can handle that you put an overloaded signal through them, others will create digital garbage. So generally you should keep the meters out of the red zone before applying effects.
If your track gains are all set to 0dB the signal is routed unaltered up the track hierarchy, unless your audio source is mono and thus is applied the chosen panning law.
That was some tough questions considering you were just strolling around in town 😉
Frits
I always ask myself tough questions strolling around town!!!
The mono thing explains the signal not being red on the master channel!
I’m unresolved as to whether I’m for or against the gain/ automation theing. I must admit, I’ve gotten used to it; and I do understand how you arrived at your conclusion. We’ll see what others say!
ta for replying!
Duncan