How to nudge a vocal take backwards a bit

[lutherbaker]

I have my recording buffer set to 512 and that causes just a bit of recording delay.

After recording a vocalist, I'd like to shift that audio back just a bit. How would I do that? If possible, I'd prefer not to trim anything - I just want to move it back a shade.

-Luther

[Tritonemusic]

DP7 User Guide, pg. 535.

[lutherbaker]

In my DP7 book, 535 covers REPEAT, MERGE ...

Ah, maybe you mean 537 in this version ... SHIFT.

That works - no way to do this in a scrubby, dragging, trial and error sort of way?

[Shooshie]

SHIFT is your friend. Command-L. Use it often. It remembers your last shift, so if you just need to keep shifting by the same amount, it's as simple as typing Command-L, Return.

Shoosh

[Tritonemusic]

In my DP7 book, 535 covers REPEAT, MERGE ...

Ah, maybe you mean 537 in this version ... SHIFT.

It's definitely pg. 535. Pg. 537 covers SPLIT and TRIM in my book. I guess there are different versions.

That works - no way to do this in a scrubby, dragging, trial and error sort of way?

Check the Index under "Soundbites, dragging..."

[amplidood]

Also check out pg. 307, all about Nudging. I have mine set to 1ms at all times when working on vocals.

[dix]

Weird, my DP7 manual agrees with 666. The Merge info is on 533. Is there more than one version of the DP7 manual? …anyway, i believe you do own a manual so:

You can make the Edit Grid whatever you want. Click on the G in the circle next to the grid-value note (up on the right) and Snap Info inspector will show up - if you need non-standard increments. That inspector also has the Nudge info (you can then assign a key-command to Nudge. that's how I normally experiment with slipping tracks). You can change the grid to display time rather than note-value in the Setup>Time Formats pref.

[HCMarkus]

Nudging using the arrow keys is super handy, but if you have your computer and DP set up properly, audio should be recorded without any "recording delay" regardless of the buffer size. There are instructions in the manual (maybe "Getting Started" guide) about how to make sure everything, audio and MIDI, is slotting into the proper timespace. It is a one-time setup thing. The OP may want to explore this.

The exception to the above is when recording VIs... although your MIDI performance should be recorded as played, if using a higher buffer, there is a tendency for the performer to rush as he or she attemtps to play in a manner that sounds in time. As such, always use as small a buffer size as possible when recording with VIs or record using a MIDI module and assign the MIDI output to a VI only after the performance has been layed down.

[dix]

…oh right! What HCMarkus said. DP compensates for buffer latency. There is no lag on recorded audio.

[rickorick]

Hi
You can drag either snap to grid or hold down the command key to over ride the snapping, I do this to get
things close and then nudge from there.
Good Luck
Rick

[lutherbaker]

@amplidood that's funny, "Nudge" is on 309 in my manual. Just got the software 2 weeks ago so I must have an extra page in there somewhere.

Great feedback everyone. Thanks. I didn't realize that audio could be treated so similarly to MIDI. My grid was setup so large that I just thought I couldn't drag the waveform.

One thing I can't do (which I can do in Live) - I can't move the waveform to the left (past zero). That may sound odd - but since my recording is delayed, I need to move the track into the negative so that my late noises show up on time ... I really don't want to hear anything before time 0:00, but I need to shift left nonetheless.

> if you have your computer and DP set up properly, audio should be recorded without any "recording delay"

I actually thought this was the case. I will go back and review the getting started guide. In my test case, I have 4 simple, external MIDI tracks and and another track dedicated to MOTU's BPM providing the beat. The 4 MIDI tracks are going external to a MOTIF XS6.

I monitor externally so BPM gets piped out of my FF800 into channels 1 and 2 of an external mixer while the Motif gets piped into channels 3 and 4 of that mixer and everything plays perfectly together.

The mixer has a B mix - so for this recording, I simply routed the Motif channels back into the FF800 and recorded them. I obviously can't monitor them via the FF because of the monitoring latency - but I was surprised when the actual recording - which sounds perfect when played live, was delayed.

Basically, when I play it back later (internal BPM + audio) the audio recording of the MOTIF is late. If I turn the MIDI tracks back on and mute the audio, it all syncs up perfectly.

Maybe I am doing something wrong. I have intentionally set my Buffer to 512 (a) because BPM is a beast and DP keeps stopping and warning me when my Buffer is set much smaller and (b) because I want to make sure that for this test, I choose a buffer that does introduce noticeable delay so that when I get this fixed, I know it is working correctly. I can't always tell, with a small buffer, if it is recording correctly since the buffer starts to become negligible.

Hope that makes sense. I will check and see what other setting I might have wrong.

Thanks again for all the feedback,
-Luther

[lutherbaker]

Ok - so an update.

This works: if I create an audio track and wire the Motif signal directly to it, the audio lags while monitoring but once recorded, plays back perfectly in time.

But the following doesn't work .... for some context: let's say I just spent 4 hours working on a new idea and I'd like to bounce the current sequence to a stereo scratch track to listen to later without having to setup all my keyboards.

Since a single track can't record from more than one set of inputs, I create a bus (this is the problem). I then create a new audio track - my stereo scratch track - and assign "Bus 1-2" to its inputs.

For the drums, I send the outputs of the internal drums directly to this new bus (Bus 1-2).

For the external Motif, I create an aux channel and set the inputs to the Motif signal (Fireface channels 5-6) and the outputs to Bus 1-2. I plug my headphones directly in the FF and I can immediately hear the lag. The drums are dead on with the click track but the Motif is late. I chalk this up to monitoring latency and record anyway.

No dice. The final results sound just like I heard. The Motif is late.

For grins, I create another audio track and assign the Motif signal directly to it (channels 5-6). When monitoring - I hear that the signal on that track too is delayed (it is in perfect time with the monitored Motif sounds on the Stereo Scratch Track - both are delayed). BUT ... I re-record both the dedicate track and the stereo track and sure enough, on playback, the dedicated Motif track plays perfectly in time whereas the stereo scratch track plays the drums perfectly - but the Motif is dragging behind.

---

Ok - it seems that I only get this latency when I use a Bus. I wonder if sending the external audio signal from a Bus to a track means that the track records with latency since the audio track is 'reading' the Bus ... which, while recording, is late. Whoa.

Make sense?

If that is the case, what is the best way to create a quick stereo scratch track of several MIDI'd keyboards, a few scratch vocals, virtual drums and virtual bassline? I hope I don't need to track every keyboard individually just to then record a scratch track.

Thanks,
-Luther


----

EDIT: If I simply record ALL external devices directly to a stereo track ... that is actually good enough. I can play the virtual instruments along with that stereo track ... or mix everything to a stereo scratch track at that point.

[Shooshie]

One thing I can't do (which I can do in Live) - I can't move the waveform to the left (past zero).

In the Chunks Window, set the start time to zero) Then select all (including conductor track) and shift one bar later. Or, you can start by inserting a blank bar at 1/1/000, then set the start time to 0/1/000. You'll be starting with an empty bar at this point, but you'll be able to shift anything to the left of 1/1/000. You may find it necessary to start at -1 (negative one) for some things. I tell you this just so you realize that DP will actually do negative bars in its timeline -- as far back as you want.

There are other ways of getting there, including trimming your soundbites so you can start them where you want, but I very often start my sequences with a blank bar, and I think a lot of people do this.

Shooshie

[lutherbaker]

I guess the trick there is to loop the sequence correctly if it comes up as part of a chunk in a song and repeats. I basically need the sequence to START by playing measures 1 to 10 but when it loops back, it needs to loop to bar 2 and play bars 2 to 10 from then on out ...

Maybe that is possible with chunks.

It also makes sense that if audio starts right on the downbeat, I'd want a blank measure infront of the beginning anyway - just to catch everything.

At any rate, that indirectly throws the bars off .... no more dividing by 4 to have an idea that a transition might be coming up .... unless we add 4 bars to the intro. I guess there is likely no general consensus there.

Thanks Shooshie!