New audio file while recording? (vs Optimize files later)

[roderickm]

Is there a way, during live recording, to ask Logic to close the current audio files and open new ones? Logic does this automatically when it decides, but I wonder if it's possible to request it during live recording.

 

Why? What possible use case might there be?

 

Each week, I record church services that alternate from 30 inputs of musicians to one or two speaking mics, and there's no time to stop recording, disarm unused tracks, and start recording again. So I just let it roll for the entire service, and come back later to slice up some regions and optimize files to save all the wasted disk space. This takes a bit of time that could be avoided if I could tell Logic, while recording, to create an audio file boundary where I know I'll need it later. Then I can simply delete the entire unused audio file rather than wait for Optimize Files to do its magic.

 

This may be a feature request, but I wanted to ask more experienced Logicians if it already exists before assuming anything. Thanks!

[fuzzfilth]

It doesn't exist and it will not be implemented, as it would need some drastical changes in basic structures to accomodate one very special purpose.

 

What do you do with all the recordings ? Archive them as is ? Who would listen to them then ? Or mix them down and publish ? Then you can erase the entire recording after you're done mixing. Either way, there's no point in implementing the feature.

 

Christian

[ski]

You should never arm/disarm tracks during a live performance of any kind. Just let the thing run in record and edit later.

 

As far as production value is concerned... I imagine that you'd want to kill the leakage from mics when they're not being used. In that case you can use two features:

 

a) strip silence

 

...or...

 

b) noise gates

 

Strip silence is probably the best bet if you want to treat individual tracks while minimizing the amount of manual editing (and it would result in a shitload of work anyway, as you'd still have to manually tweak each edit. So a better alternative is to bus all of your instrumentalists tracks to an Aux and put a noise gate on the Aux. Put another noise gate on each of the talking mics. Now you have just three noise gates to deal with instead of 32.

 

Next...

 

You can create markers on the fly and name them while Logic is recording: Set up a key command for Create Marker (your options are with and without rounding; suggest the latter) and use that to mark sections of the service in real time. You can name it right after you create it, even if Logic is recording.

 

In post-production you can move the playhead to those markers, select all regions, and use the Split at Playhead function to slice them all en masse. Move to the next marker and do the same. With this method you have to pay attention during the entire service to enter these markers-as-future-edit-points, so don't drift off to sleep during the sermon

 

The noise gate method doesn't save you disk space. The strip silence + manual editing method saves you disk space but requires a lot of editing time all the same. So ask yourself... are you doing this gig to save yourself time or to save disk space?

[roderickm]

Logic already initiates this action automatically, so I'm not asking to do something impossible. I'm asking that automatic audio file rotation also be user-triggerable. Why do you say it would require "drastical changes" to rotate the audio files on request during recording?

 

I do archive the session prior to mixdown, so there's a benefit to trimming the dead space. Also, copying the session for other Logic users is much easier at the smaller size. A number of the musicians also use Logic and have asked for copies of the files for review and personal rehearsal.

 

Before splitting the regions and running Optimize Files, the audio data totals about 20GB per session. Afterward it's usually 10-11GB. This operation requires about 40 minutes:

• Split regions by playhead
  Delete dead space in arrange window
  Select unused in Audio Bin and delete
  Select all files, Optimize Files and wait...

If I could mark the boundaries of audio files during recording, then deleting the unused audio files would take maybe 3-5 minutes, no need for Optimize Files.

 

I can accept "it doesn't exist," but I bristle at your comment, "it will not... there's no point..." Is this your sole opinion, or are you a product manager for Apple's Logic? And again, I'd be interested to know why you think this would require drastic changes.

[roderickm]

Hi ski, thanks for the tip. Creating markers on the fly already helps me quickly locate the places to Split Regions at Playhead, so the editing isn't difficult or time consuming. It's waiting for the Optimize Files to chew through 30-40 files that takes some time.

 

Disk space is cheap, but bigger sessions means we must rotate storage more frequently, the sessions take twice as long to copy for anyone, and they're generally more cumbersome. This isn't a huge pain, but this little easter egg of a feature would help me save a significant amount of time.

[fuzzfilth]

Well, the obvious hint lies in the fundamental difference between Punch On The Fly On and Punch On The Fly Off.

 

Off will give you the basis for your feature, but at the same time, and because of this, it will disrupt any audio stream in or out when you hit record, during which it negotiates with the OS about how many files are to be created where and of what size.

 

On will prevent this from happening as it handles this during preparation for recording, when you arm a track for record. Once you hit Play, Logic is already recording in the background, even without you actually taking notice. Thats why you can punch in and out seamlessly, and even repeatedly, without creating files left and right as you go on. The price is that you can't change the record status once you're rolling.

 

Christian

[ski]

Hi ski, thanks for the tip. Creating markers on the fly already helps me quickly locate the places to Split Regions at Playhead, so the editing isn't difficult or time consuming. It's waiting for the Optimize Files to chew through 30-40 files that takes some time.

 

Disk space is cheap, but bigger sessions means we must rotate storage more frequently, the sessions take twice as long to copy for anyone, and they're generally more cumbersome. This isn't a huge pain, but this little easter egg of a feature would help me save a significant amount of time.

 

You're welcome.

 

A few additional thoughts...

 

1) If you record at 44.1K, 16 bit you'll save not only disk space but time as well, not only in the copying/archiving of files but also in the optimization process.

 

2) Considering that disk space and post production time are important, let me ask, "what is the end product?" If it's to make CD's or MP3's then you really don't need 24-bit dynamic range and/or 24-bit file sizes. So anyway, what is your end product?

 

3) Are you working with a mixing board or are you taking all of those inputs directly into a very large array of audio interfaces? My thinking here is that if you're working with a board that you could save tracks (and thus disk space, copying time, editing time, and optimization time) by recording submixes of various instrumental groups (i.e., if you have a choir, submix S, A, T and B or even just a stereo pair).

 

4) I think it would be a cool thing (in theory) to be able to seamlessly take channels in and out of record just like we used to be able to do with tape machines. But the technology just isn't sophisticated enough in Logic to allow this to happen, as fuzzfilth described. So I'm thinking that if post-production time is such a big consideration, and you are indeed working with a mixing board, creating submixes or otherwise minimizing the number of inputs would help you there.

[roderickm]

1) You're right that 24 bit resolution isn't required -- I'll test with 24-bit recording disabled.

 

2) The workflow has two primary end products: a podcast of the pastor's message (lo-fi, 32kbps MP3) and mixes of the worship songs (published as 192 kbps MP3). The interest in smaller file sizes comes from others that want copies of the raw session, and the desire to archive them.

 

3) I'm sourcing from a Yamaha M7CL with a couple of Dante MY16-AUD cards, so the audio comes from direct outs via ethernet as 48k/24-bit streams via Dante's Virtual Soundcard software. I do take a mix straight from the desk, and that is usually good for the podcast source. Submixing drums or guitars leaves fewer options for mixdown, so I keep them on separate tracks.

 

4) Maybe I've been misunderstood -- I don't care to disarm some tracks while recording.

 

Logic normally creates sequential files in the audio bin numbered like Bass#02.aif, Bass#03.aif, Bass#04.aif. The point at which Logic rotates its disk output from one file to the next is automatic. If I could ask Logic to start new files for all channels at transition points in the event (when the band stops playing and the pastor begins speaking, for instance), then trimming the unnecessary data would be as simple as deleting entire audio files rather than manually splitting regions and waiting for Optimize Files.

 

Or to put it another way, if I could stop and restart recording without missing a moment of audio, keeping the same tracks armed... just a little blip of telling Logic to close the files and to continue recording on newly incremented ones, at a keypress rather than when it decides to automatically do so. For what it's worth, Boom recorder allows this by enabling the "rec button starts new take during recording" preference. The Boom Recorder manual says, "This is useful to manually split recordings during a live concert for example." Yes!

 

Thanks again for the tips and discussion. I'm going to test with lower resolution, but I can live happily with getting coffee during my next run of Optimize Files.

[cnse]

How about this? It's not a perfect solution, but it might get you closer to the desired result.

1. Create and open 2 projects simultaneously. Pastor (stereo input), and Music (30 inputs). Arm all tracks.

2. Start recording with the first (for ex. Pastor).

3. After message, stop Pastor.

4. Switch to Music, hit record.

5. Repeat as necessary.

 

The only thing you'd loose is the 2 seconds it would take to switch projects and re-initiate recording.

You could setup a key command for record toggle to make the process for efficient.