For the longest time I always thought the MIDI latency delay compensation was never an issue but I had the feeling since I made the switch to Cubase 6 that something sounded wrong. I tried doing a phase flop test from my instrument track vs an audio track, here’s what I did:
Create an instrument track with Massive (I tried out Twin 2, Battery as well. FYI I have one 1 oscillator with the Reset via gate turned on so the oscillator is at phase 0 each time the note is triggered).
Put a single note with the instrument track.
Set my loop locators and mixdown my instrument track without any plug-ins.
Imported the audio track and then phase flopped one of the tracks
To my surprise when I looped both the instrument and audio track the phase was random every time. This SHOULD produce complete silence not be all random and comb filtering.
Does anyone know of a solution to this? I search the forum but couldn’t find anything at all. This is very disappointing knowing that while I produce my mixes are completely being comb filtered.
This is very disappointing knowing that while I produce my mixes are completely being comb filtered.
Your tracks will not be comb filtered/phasing in normal use because you only have a single track of the same instrument sound…there’s nothing for it to phase against.
Maybe if you were augmenting real drums with a sample you could notice some change in the hits.
Vsti’s never seems to null with their audio export.
Actually while trying this I also found that real time export is way louder than non realtime. I couldn’t get these to anywhere near null with each other either. EDIT…Ignore this…User error!!
Yes this is true I won’t get comb filtering unless I have the exact same audio track. However it really annoying for drums and sample aligning. A friend of mine tried it in pro tools and gets the same thing except his was consistently off by like 15 samples or something so he could compensate for it, where in Cubase its totally random
To my surprise when I looped both the instrument and audio track the phase was random every time. This SHOULD produce complete silence not be all random and comb filtering.
I would expect this. Synths (instrument tracks), whatever they are are not audio tracks. There will be some system delays as the signal goes thru the instrument, computer and soundcard etc and somewhat less from the audio.
Now if you copied that audio track that you generated and that was out on phase inversion then I’d be a little surprised.
If someone mentions that P. Tools does not do it then it must be explained that that system has it’s own hardware and everything will be controlled, and maybe corrected by that. Cubase relies on the clocks on external soundcards which probably contributes to unavoidable artifacts. This is maybe one reason why P.Tools costs considerably more. And on a trawl around I’m not totally convinced P.Tools is completely free of noticeable artifacts.
We have to learn the product limitations every day.
Also it helps others to see your system specs so they get a clue as to why a particular event is happening. Plus the sample rates you are working at as well.
