GregManning Posted May 21, 2021 Share Posted May 21, 2021 Hi, I just tried to null two identical mixes with full plugin compensation. One was done in real-time, the other an off-line bounce. They didn’t null when one got phase reversed. The off-line bounce seems to be delayed by about 3 clicks. Why? I’m using UAD Apollo x8. Quote Link to comment Share on other sites More sharing options...
fisherking Posted May 21, 2021 Share Posted May 21, 2021 is one mix, in some way, better than the other? otherwise, it's (probably) just the process itself, and nothing to worry about. in all these years, i've only done 2 or 3 real-time bounces (and only because a sustain pedal event only 'triggered' in real-time, and one odd synth did it's 'morphing' thing better in real time). if it sounds fine, it is fine... Quote Link to comment Share on other sites More sharing options...
GregManning Posted May 21, 2021 Author Share Posted May 21, 2021 The sound is the same, but when a client gets a file to play to it should start at the exact same time whether real-time or offline bounced. To my surprise it matters time wise how you bounce it. Not sure I express myself clearly... Quote Link to comment Share on other sites More sharing options...
fisherking Posted May 21, 2021 Share Posted May 21, 2021 i don't get it. if you send them a real-time bounced track, that's their reference point. if you send an offline bounced track, that's the reference point. in what situation is a comparison necessary? EDIT: seriously, has anyone anywhere ever said "i can't use this mix, it's 3 clicks off from where i expected it to be"... Quote Link to comment Share on other sites More sharing options...
des99 Posted May 22, 2021 Share Posted May 22, 2021 Strange. If you do a quick test in an new project, and don’t use UAD plugins, do you still get this delay between the different methods of bouncing? Quote Link to comment Share on other sites More sharing options...
Ninecows Posted May 24, 2021 Share Posted May 24, 2021 Not sure where the delay thing comes in, but I watched a video from present day production (can’t find it now), where they mentioned that many plugins have a slight randomness in there response to make them sound more analog. Just a small amount. Not enough to be able to hear it on and A/B listening, but enough for them not to pass a null test Quote Link to comment Share on other sites More sharing options...
triplets Posted May 24, 2021 Share Posted May 24, 2021 Not enough to be able to hear it on and A/B listening, but enough for them not to pass a null test But do you think the listener is gonna do a null test? Of course, not. Nobody can hear that and nobody cares anyway. Quote Link to comment Share on other sites More sharing options...
Ninecows Posted May 24, 2021 Share Posted May 24, 2021 No the listener is not. But I was trying to explain why the OP might not get 0 out when doing the null test Quote Link to comment Share on other sites More sharing options...
Ninecows Posted May 24, 2021 Share Posted May 24, 2021 But is it the plug-in compensation doing it here? Quote Link to comment Share on other sites More sharing options...
analogika Posted May 24, 2021 Share Posted May 24, 2021 i don't get it. if you send them a real-time bounced track, that's their reference point. if you send an offline bounced track, that's the reference point. in what situation is a comparison necessary? EDIT: seriously, has anyone anywhere ever said "i can't use this mix, it's 3 clicks off from where i expected it to be"... If I bounce a new mix of track fixed under a video, expecting to just replace it in the video editor, it kind of makes a difference whether the start point is identical or not. Same if I bounce for any synced-up production playback. I’d have had no idea what was going on. Quote Link to comment Share on other sites More sharing options...
fisherking Posted May 24, 2021 Share Posted May 24, 2021 i have never had a problem bouncing audio-for-film; i do offline bounces, label the smpte start point... and all is well (and if a director or editor wants to move the music slightly, then will do so). all good! Quote Link to comment Share on other sites More sharing options...
Solution gacki Posted May 24, 2021 Solution Share Posted May 24, 2021 There certainly is some "weirdness" in real time playback; not sure if that also translates to real time bounces. It feels to me as if the internal MIDI engine needs some "settling time". I actually had prepared an example song and wanted to ask about this (in the sense of "how phase stable is Logic"); however I wanted to test this further. Short description of the things I have observed: I created an Ultrabeat region with four Rim clicks, then I did a "bounce in place" of that region and phase reversed the audio track. In theory those two tracks (Ultrabeat and Audio) should null. And they actually do this but only if there is enough "pre-roll" before them. Placing them at measure 2 and starting from the beginning results in dead quiet. Starting from anywhere between measure 1 and measure 2 gives varying comb filter effects; so does moving the regions gradually towards the start. This was tested on two different computers with different interfaces and buffer settings (Logic 10.5.1). I'm still in the "requires more research" phase here but it appears to me that real time playback can have some small glitches. Not sure if that translates to real time bounces as well and how much offline bounce differ from this. Quote Link to comment Share on other sites More sharing options...
GregManning Posted August 4, 2021 Author Share Posted August 4, 2021 Thanks for your observation! Details matter. At least to me. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.