Quote:
Quote:
I would say it is added post master fader and pre first insert. Signal is always truncated to 24 bits in inserts and aux inputs in Pro Tools TDM, so the only way to keep the 48 bits resolution of the TDM mixer is to dither before the signal being truncated in the input insert or aux. If it was post inserts there would be a 48 to 24 bits non-dithered truncation in the first insert of the master fader, and dither would be redundant since the last plug-in it is dithering its output to 24 bits and the wordlenght is the same between the last output insert and master output since there is no level change involved there.
|
I guess this scenario makes sense, but then that dither would come undone after the first plug-in is inserted on the master & would need to be re-done after each plug-in. Meaning all the hardwork to dither your 48bit signal to 24bits is useless if it truncates through the proceeding plug-ins. It still doesn't prevent the audio truncating to 24bits on individual channel plug-in inserts either. The fact that this is left up to the individual plug-in manufacturers is a problem. Basically the more plug-ins you have in your session the more truncation distortion is being added to your audio. For one or 2 plug-ins this is less of a problem but for bigger sessions this weak link will become much more audible. It can also be debated that even if every plug-in did dithered back to 24bits before output that the accumulative dither would be additive & become unacceptable at some point.
Really the best way to implement precision in a DAW is to run it at the highest resolution through the entire path. In the case of TDM this would be 48bit (with 56bit accumulator) all the way through & dither the final output to 24bits on the way out. While the HD mixer has come a long way from previous versions, it is still a far cry from perfect & if you could compare what a 64bit floating point DAW sounds like to HD you would understand why I am mentioning this.
Anyway food for thought...
Matt
|
The way audio is processed within plug-ins has to do with the plug-in architechture, some plug-ins dither their output and some plug-ins don't, I like to see TDM plug-ins like hardware inserts.
IMO additive dither from plug-ins is far from being a problem, 100 24 bits dither generations would be needed to increase the noise floor in 20 dB, and that would be 120 dB of dynamic range, no many converters can accomplish that and I donīt think it can be audible in the real world.
I agree with you respect running the highest resolution through the entire path would be better, specially for getting rid of all the points were signal can be clipped without being noticed if you are not careful with levels, such as aux tracks and some plug-ins that don't show when their inputs are clipping. Ironically it is not a problem at all in LE.
Cheers,
Daniel.