Audacity on "older" PCs-some fine-tuning suggestions(SOLVED)

I’m not sure that Audacity is actually aware of that.
Audacity probes the sound system to see what rates the sound system says that it supports. If you select a rate that the system says that it does not support, then Audacity will request the data in one of the supported formats and resample it on the fly.
If the sound system does not actually support one of the rates that it says that it does, then there’s a problem, but not one that Audacity can do anything about.
Setting the sample rate in Audacity to the same as Jack ensures that this problem does not occur.

My system, when using Jack, claims to support the following recording rates:

Device ID: 8
Device name: default
Host name: JACK Audio Connection Kit
Input channels: 32
Output channels: 32
Low Input Latency: 0.011610
Low Output Latency: 0.011610
High Input Latency: 0.046440
High Output Latency: 0.046440
Supported Rates:
    8000
    9600
    11025
    12000
    15000
    16000
    22050
    24000
    32000
    44100
    48000
    88200
    96000
    192000

I’ve not tested all of these as I generally stick to 44100 Hz or 48 kHz.

Thanks for the help, Steve. I don’t know exactly why Audacity didn’t “see” Jack’s setting, but at least it’s good to know to check it.

I’m now using Audacity on an upgrade from Ubuntu Studio to a system called KXStudio. Audacity works like a charm here, and I’m having a blast!

Question: If I want to do an append record, is there a way to set Audacity so that it gives me a few countdown beats? Because I can’t hit shift+record and then start playing the keyboard with both hands, without a delay.

What would be real cool is a function where you play back your track, and it automatically begins recording when it reaches a certain point. A type of “punch-in” record.

Brian

No, Audacity will start recording immediately.
Even if Audacity did have that feature it would be near impossible to get a really “clean” start to the recording.
I find the easiest thing is to add a label at the end of the track (END key, then Ctrl + B, then ENTER) before I start Append Record. It is then very easy to go back and delete the lead-in silence and get a really good join.

Audacity does have a kind of “punch-in” record, but it is not quite how some people expect.

The traditional punch-in record method was developed on and for tape based recorders where the number of available tracks was extremely limited (originally only one mono or stereo track). It was a form of “crash editing” that rarely gave clean joins, so was usually accompanied by razor-blade and sticky tape editing. As multi-track tape recorders became more common and more sophisticated, punch-in recording became more refined. The enabling/disabling of record/playback/erase heads was carefully timed so that the recording would just overlap the previously recorded material.

Early digital recorders were often closely modeled on the familiar tape based machines, so it was a natural progression to include “punch-in” record. However a intrinsic problem still remained - If you get the punch-in wrong, you have overwritten part of the “good” recording. To work around this problem, various methods were developed - one of the most popular on hard disk recorders being the idea of “virtual tracks”. Some software recorders use a similar approach by creating track “layers”. This can work reasonably well, but it can also get very complicated, especially if you want to edit the ends of the layers to neaten up the joins. To aid editing of multiple takes, it is often possible to move “layers” to new tracks, so that the “drop-in” is on a new track below the original track. Precise editing of the ends can then be performed to achieve a seamless join.

Can you see where this is going?

If “punching” into the original track was developed because of the limited number of tracks (not an issue for software recorders) and the new recording is being moved to a new track so that the ends cam be edited precisely, then what’s the point of punching into the original track? Why not “punch in” to a new track? This is the approach that Audacity takes.

To make a “punch-in” recording onto a new track:

  • Silence the part that you want to “record over”. Either use the Envelope tool, Split Delete, or Ctrl+L (Silence audio)
  • Place the cursor before the punch-in point - allow a good amount of “pre-roll” (count in).
  • Start recording.

The new recording will be on a new track, and if you have the latency correction set up correctly it will be perfectly lined up with the first track ready for editing to create perfect, seamless joins.

Yes a “crash edit” may be a bit quicker, but highly unlikely to give a really good result.

I see your point, Steve. Just start a new track and start playing when the right point is reached; then append the one to the other later on. It makes perfect sense. I guess I was too “locked in” to my Yamaha Motif’s way of recording, when in reality, software like Audacity has really made things not simple, but super-simple! :smiley:

I’ll re-read your tips tomorrow when I get a chance to do some recording. It should be a piece of cake, really…

Thanks!

brian

PS: I’m really liking how Audacity works in KXStudio compared to Ubuntu Studio. As I’m not a programmer, I don’t know what is different in the “background” (ie dependencies, upgrades, etc) but I haven’t had any glitches or hitches with Audacity at all. I’d recommend KXStudio to anyone wanting to use Audacity and the plethera of audio production tools available in Linux so well just “out-of-the-box.” Really well done, imho…

I tried KXStudio a long time ago - it looked nice but it was horribly unstable, so it’s good to hear that it’s working well now. I’ll have to give it another go some time.