Moving Selection Lengths - and - Recording in Sync

[Newb1]

Hello, I had a few questions to ask - first, thanks to Audacity I've been able to begin learning how to work with audio production, which, because of the loudness war (and just an overall interest in analyzing things, waveforms in music's case), I've had an interest in pursuing. Overall even if I'm not in recording process, I'm enjoying the ability to still analyze music and audio - so I wanted to say thanks for keeping it freeware (I wouldn't be able to afford anything else) and it's a great program!

I am pretty inexperienced with it, so I don't know too much (though I'm willing to learn, so please don't dumb down vocabulary for me - but at the same time don't get too complicated lol), but I had two questions:

1.) How does one move a selection length? When you select a part of a track, how do you move this: not moving the data that is selected, just the length. For example, I select 5.476... seconds of audio in a track - how do I move the selection of just the 5.476... seconds, and not the audio/data that was selected in itself? I'm not sure how to explain this...

I have four separate tracks to a normal song (vocals, drums, guitar, bass), and when I obtained the tracks, this time they were not all the same length - for example, vocals come in the song later than the other instruments, and the vocal track doesn't have silence before it, just goes right into the singing. The reason I said "5.476" instead of 5 or 5.5 earlier is because I wanted to express that I need the selection to be (pretty much) perfectly in sync. I added silence before the singing on the vocals track, so it is currently lengthier than the guitar track (example). How much silence I need to delete, is the differences in time when the two tracks end. The snap guide helps me select this difference, but how do I move this selection of time (just time) over to the added silence so I can delete the correct amount?

2.) Recording in Sync. My computer is not the fastest, nor does it have all the free hard drive space in the world (fairly low nowadays). I can usually record one track from scratch decently enough, but when I record a second track, it goes really slow and choppy - so the recording reflects this, pretty bad. (That, and it's difficult to play along the other tracks because the inconsistent speed) I have all the minimal settings that would still keep a good recording all set - 16-bit recording, Mono, no Software Playthrough - Overdub is on, but without that I can't play in rhythm to the other tracks. I assume the reason behind poor recording here is because of a lower CPU speed (the computer's 3~ years old, it's a laptop, does not have a great amount of free space left and it's Windows vista - which I heard was also a possibly negative factor in CPU/speed performance.)

There's the latency settings, which right now are at: Audio buffer at 100, Latency correction at -100. I don't understand the mechanics behind this, but I read how the recording was smoother (and full speed, consistent) when these were up. When I moved them up to 512, -512 (respectively) or 1024, -1024 (again respectively), it recorded much better, the backing tracks/other tracks were regular speed. There were still a few encountered problems however.

a.) Occasionally I would notice the recording would mess up - for a few seconds the recording sounds scatterred and the notes aren't where they were supposed to be. These "mess-ups" lasted no longer than a few seconds each until resuming back to normal, but it's still a problem - it's not really fair when you finally nail the guitar's part to the song correctly and parts of it are unusable!

b.) This sort of goes back to the original question #1: In this case, I think it does this because of the Latency settings: When the recording's finished, the recently recorded track plays later than the other tracks - it requires one to delete some space in the recently recorded track before it's instrument plays (so it would be silence being deleted, like in #1). Is there a way to have Audacity record another track and it already be placed in sync with the other tracks the same way it was recorded?

I suppose I could ask in a different way, although now it's not entirely an Audacity-based question anymore: How could one force the Audacity program to use up more power - In the Task Manager, Firefox uses about ten times as much memory, so I figure if I end Firefox I have more "room" for Audacity to use memory-wise. How would I increase this? If I'm even asking something plausible? lol


I apologize for the long questions and thanks in advance!

[billw58]

Before I tell you how to delete that extra silence you added at the start of the vocal track, let me observe that you're doing things the hard way. You can simply use the Time Shift tool http://manual.audacityteam.org/index.ph ... #timeshift to slide the vocal track to exactly the right position. There's some examples here http://manual.audacityteam.org/index.ph ... _and_Clips . If you cut out all the silence at the start of the vocal track, the put a label at the spot in the guitar track where you want the vocal to come in, you can slide (time shift) the vocal track so the start of the clip snaps to the label line.

You can make precise selections (including start time, end time, and length) using the Selection Toolbar http://manual.audacityteam.org/index.ph ... on_Toolbar.

You can cause a track to start at a specific point on the timeline by first clicking in the track to select it, then using the Selection Toolbar to move the cursor to the time you want, then doing Tracks > Align Tracks > Align with Cursor.

Latency and Audio to Buffer do two different things. As you've noted, setting a higher Audio to Buffer amount reduces CPU load and prevents choppy audio and inconsistent speeds. The latency correction is a different thing. When you play a note on the guitar it takes the computer some finite length of time to process the audio and write it to the disk. This delay, relative to the track you are listening to, is dependent on two things: the audio to buffer setting (which you can control) and your computer's sound hardware and operating system software (which you can't control). The upshot is that if you set latency correction to 0, the track you record will be late relative to the track you are listening to. The Latency Correction setting attempts to correct for this after you stop recording by pushing the recorded track back in the timeline. A tutorial on how to set the latency correction for you computer is here: http://manual.audacityteam.org/index.ph ... tency_Test

One of the Windows elves will need to chime in here with strategies to give Audacity access to more resources on your computer. Exiting all other programs while using Audacity can't hurt, though.

Finally, have a look at the manual: http://manual.audacityteam.org/index.ph ... =Main_Page There are no tutorials on multi-track recording and mixing (yet - if you are successful in doing what you want, perhaps you could write them!). There is one on mixing a narration with background music http://manual.audacityteam.org/index.ph ... ound_Music that talks about basic editing, time-shifting and mixing.

Good luck. Let us know how things work out and don't hesitate to ask more questions.

-- Bill

[waxcylinder]

You really should to try to free up some more disk space. Get yourself an external USB disk (or better still get two so you can have backup copies too) and archive off onto the external disk all those files you use or need infrequently - actually can be frequent files too as you can work with files on the USB disk. Just don't try and use the USB disk as your target disk for Audacity work - Audacity needs the higher speed access that your onboard disk should provide.

And once you've cleared the disk a bit, give it a good defrag (this is something you should be doing fairly regularly antway).

WC

[Newb1]

Thanks for the replies! The part about using the Selection Toolbar was pretty obvious, I don't know why I never thought about it. Thanks.

I'm thinking about getting an external drive or external disk of some sort, but I'm on a low budget.

As far as latency to buffer goes, I'm having no problem testing for the latency correction, except for how I'm trying to find an accurate value for it when the audio to buffer is over 1,000. Even then the recording can get kind of choppy, which I'm sure it due to a number of things with my computer - lack of space, not de-fragmented recently, etc... I'm working to fix some of these but I don't have the ability (or money) to improve all those aspects. Is there anyone who knows how to force Audacity to use more memory so it can handle recording much more smoothly? Like I said earlier my web browser uses at least two times as many memory as Audacity so when I close my browser I figure I could force Audacity to use more power and it would be able to handle things better and my computer would still be fine?

[kozikowski]

Let me try that first one. You can Generate > Silence whenever you want and you can make it whatever length you want and insert it wherever in the timeline you want.

You can cheat and Copy your work or clip and then Paste it into the new location. Then, while it's still selected, Control-L will turn the selection into silence -- effectively copying the time but not the show.

Koz

[kozikowski]

You never said the words: "Yes, I defragment my hard drive regularly." Audacity will not work into a fragmented hard drive.

The fuzzy rule is not to go over 90% full for any kind of theatrical production -- audio or video. Your 50GB hard drive should never get fuller than about 45GB. Beyond that, the machine may start to become unstable, slow, or sloppy.

Koz