question about audio latency between racks

[woodslanding]

I know this was discussed here before, but I couldn't track it down in a search.

Just to be clear... If I would like to use a rack as a 'master' rack, and have a bunch of other racks that have 'master' for their output, will this then not introduce additional latency?

Right now all my racks just go straight out to my output device, but it's a little clumsy. That said, I'm not up for anything that will increase latency, so I'll keep it how I have it if that's best.

thanks as always!
-eric

[fractalist]

it shouldn't add any latency in this case

to get sure, you can test it yourself :
output your racks BOTH in the master rack AND in the output device directly, and insert a plugin (or module if there is one for that) between the master rack and the output device that inverts polarity / reverse the phase.
if you play your sounds and nothing comes out as a result of the addition of these two routes, then they are cancelling out each other = no latency has been added !

[woodslanding]

hah, slick! I'll check it out.... I don't know if there is a module for it though. I assume a plugin for phase would not introduce latency--that would defeat the purpose!

Thanks!
-e

[gurulogic]

Last i checked there was an invert phasefunction in HH3, but pretty sure sending audio from one rack to another causes one bloc​ latency regardless. (Except perhaps when using single core cpu?) As far as I know, more or less any multicore aware host will introduce a processing bloc latency when routing one track to another., so no different in Usine really. Also, if I recall correctly, devices input / output each introduce a bloc of latency when patched into a rack so actually less latency to patch in for example a devices input patch and use a bus to connect to an output patch, bypassing the racks altogether.

[23fx23]

phase invert can be done just by multiplying audio stream by -1 (in fact audio is an array of bloc size)