Moving tracks around

Moving stuff again?

Weeel, went back into my doc’s stuff around this, cant seem to find anything and it is not intuitive at all. I simply want to move a buncha tracks (mixdowns) onto the main track and then hit normalize to CD.

How do you do this? I opened up the main track, then opened up the other mixed down tracks: thought I could then drag and drop the other tracks to the top one, then normalize that conglemerated track to a CD?

I can not seem to “grab” a lower track and drag it to the end of another track.

What am I missing here? (tried cntrl-shft, paste, etc).

Thanks,

Scott

In order to move stuff around, a firm grasp of some basic concepts are necessary:

A Track is an entity beginning at time zero and extending for as long as the song goes, i.e. eternity if you want… - a track may thus not be moved around in time.

A Part is a ‘pointer’ or ‘handle’ to a wave file (in case of a part on an audio track) or a bunch of midi events (in case of a midi track). A part has a few unique properties, among them its starting time in the track. This is what distinguishes wave files from parts - wave files may be moved around in time at will.

To assemble a final CD wave file from a bunch of finished tacks in n-Track (which I understand is what you want to do) you may go about it in the following way:

1) Import the wave file for the first track. Note track number and the starting point - the track number should be 1 and the starting point should be set to time 0:00.000. Now, assuming the first track has a duration of 2:43.000 and you don’t want any ‘rill’ or empty space between tracks…

2) Import the wave file for the second track. When importing, adjust the starting point to be 2:43.000 and the track number to be 1 - n-Track guesses it to be 2 initially. Now you have two consecutive tracks in track 1.

3) Continue like this for the whole CD, adjusting the timings to start where their predecessor ends.

4) Render the whole compilation to one big wave file.

If you want crossfades on the CD, this is easy to do as well. Just adjust the staring point on the track to crossfade into to a couple of seconds before the previous track ends. Beware of the levels - you may need to lower the fadeout of the previous track a bit to avoid overs.

However, in spite of all this I would strongly suggest you compile your CD within a CD burner program instead. If you use the above method, not only would you get a huge wave file to keep track of on the hard drive, you would also get a finished CD without any markings for the tracks!! You would, in other words, be unable to select ‘Track 4’ (or whatever) on the CD when you played it back on an ordinary CD player. A CD burner program (personally, I prefer Nero Burning ROM) will insert these track separators automatically in a way the CD player can understand.

I know this was lengthy, but this is not easy to comprehend at first sight…

regards, Nils

I am a little confused as to what you are trying to do. There is not a “main track” in my version of N-tracks (3.3). All the tracks are individual and none is more important than any other. They are normally combined into a single wave file durring the mixdown process. A “mixdown” usually refers to that file, not what you hear when you play the song (the “mix” is the whole combination of parts, effects and such but the mixdown itself is the file).

To create the mixdown you open the mixdown dialog (blue CD symbol on taskbar). If you intend to normalize after you mixdown all the individual files I would choose “more options” and mixdown to 32-bit. Then you would open a new .sng file and insert the mixdown you have just created, click on the track to select it then normalize it. This will “destructively” modify your mixdown to set the peak level to 0dB (or wherever you tell it to normalize to).

This can then be “mixed down” to a 16-bit track for burning to CD.

I hope this clarifies things a bit,
Jim

Sorry, it’s been so long since I used N-track I’ve forgotten a lot.
My terminology was corrected very well by Nils, understand what he was getting at.

Jimbob, I’m talking about mixed-down tracks, selecting one as the “master” so-to-speak, then adding all the other mixed down tracks one by one to the end of the other one. One large “track” to be burned to CD.

I have a buncha separate mixed down final tracks of my son’s band that I wanted to put on CD. I thought it might be a good idea to add them to a sequential track, then normalize that track so that the volumes of each separate track would be similiar, and then burn that continuous track to CD. My idea was to avoid having to up/down the volume button when playing back this mix by normalize. As I said, it’s been a very long time since I’ve fooled with this stuff, so maybe I’m totally doing it the wrong way!!

Any ideas as to the final part of this burning to CD would be appreciated. I used to know how to do this stuff, but it’s been a long time not playing with it.

Thanks
Chiller

I do not normalize to get the levels equal. Normalization is a crude tool and typically only sets the highest value to full-scale. This has nothing to do with the percieved loudness of the track.

What I do instead is create a new .sng file and put each song on a different track but placed in sequence so that they do not overlap. I use the individual channel controls to set the relative levels then the use crossed-arrow cursors to adjust the spacing between songs until it sounds right (a quiet tune following a loud tune needs more time before it starts to allow the listener to adjust). I normally try to get the levels of the vocals to sound “right” and similar (different for a ballad and a belter) but you can use any criteria you want. I just set it to play and click at the appropriate spots of the timeline for quick A/B comparisons. For setting the intervals I just focus on the last measure or two and listen to see what sounds “right” for the start of the next tune. There usually isn’t any need to listen to the whole thing through for any of these adjustments, just find a key passage in each song.

Once I have it like I want it I scan the whole .sng using either the scan function in the normalization dialog or just play it through and check the peak levels in the playback meters. I use the master volume to set the highest peak just below 0dB.

There are several ways to proceed but the easiest is just to do partial mixdowns, selecting each song in turn (and deselecting the “mixdown entire song option” in the mixdown dialog). Noise performance will be best if you started with 32-bit mixdowns. To maintain the intervals you have set you can either make note of the interval between songs and input it into the CD burner program or make your selection include some silence before and after the song. Putting makers where you want the track IDs makes this easier. You then use gaps of zero between the songs in the burner software.

The advantage of the second approach is that if the listener rips the CD to a library on his computer, you can insure that there is some silence in the intro to maintain your prefered interval when played as an album and provide a sonic buffer between that song and another in random play. I have one song on a CD that starts almost inaudibly the builds over time. If I rip it to my computer the intervals are lost and the user can’t “settle-down” and start listening more closely, destroying the effect. The opposite may be worse, a quiet song ending with a sudden loud one can ruin your ability to savor the quiet one. Give the listener a little time to adjust. Explicitly inserting silence in the .wav itself makes a “portable” interval.

Jim

Jimbob, if you know a better way of getting different mixed down tracks at a peak level to CD, then tell me. Hand changes are what we are NOT looking for.

your point is mixing (All over again?), my point is final mastering to CD.

Big difference.

Let’s get on the same darn track (hoo hoo!)

Scott

<!–QuoteBegin>

Quote
I thought it might be a good idea to add them to a sequential track, then normalize that track so that the volumes of each separate track would be similiar,

You’re missing Jimbob’s point. If you normalise a single track made up of a number of songs, you’ll still have those volume differences in proportion to each other - ie, the loudest will still be so, and the softest will still be soft.

You can import each song to a seperate track in the proper place and adjust each tracks volume so that they all have the same apparent loudness (muting before and after each song on it’s respective track), or you can put them all on the one track and use envelopes to get the same effect. You’d then stick the mastering effects on the master insert either way, be it Ozone, Endorphin, or Waves to compress/eq/limit/reverb. That way your mastering effects are going to be working with the same relative levels - and you won’t have to worry about some songs getting hit with a limiter too hard because it’s so much louder than the others. Putting them all on one track is probably going to be less cpu intensive.

Willy.

Scott,

I was under the impression you wanted good results rather than just OK results. Doing things well always takes effort.

One point I didn’t make clear was that this process is done with mixdowns of each song. All you are doing is adjusting the relative level of the preliminary mixdowns of the songs by ear rather than depending on an automated tool which does not actually have a clue as to what the loudness is.

If you don’t care what the relative loudness of each track is just go ahead and use normalization. If your songs are all similar in character you may be able to get away with it.

The process I described is intended to get a whole CD to “hold-together” with controlled transitions between songs. It is particularly useful for live music where you are trying to give the impression of a live performance which typically builds in energy level as it proceeds. With this method you can place an accapella song in the middle of a metal set and have it “fit”. You can’t do that with automation. Meters have nothing to do with music. They are tools that provide valuable guidance to help avoid distortion but they cannot replace human judgement.

I did not describe the process of maximizing the level with overall compression or limiting which can be better applied once you know what you want the relative levels should be. Once you have established the relative levels you can go “peak-hunting” and selectively reduce the peak levels wherever they approach full-scale so that you can raise the overall level of all the songs. I usually spend about 3-12 hours on this process but that is nothing compared to the amount of time I spend on the mix itself (and I mostly record live-to-disk without overdubs). One of the advantages of being an amateur is that you can afford to take the time to do things well and can sometimes substitute care for expensive automation. I also generally prefer to use “manual compression” (using envelopes) to automated compression during the mastering process but I have heard good results with people who do use automated tools. On my mixes overall compression seems to take some of the “life” out of the mix but I may not have learned how to use it properly.

Jim

Got to agree with the use of volume envelopes for compression type results. Especially if you are liking the sound of your tracks, and don’t feel like getting into another stuation to have to “come back” from!

Ah yes: the surgical strike technique on roque peaks! Unfortunately, a very time-consuming task.

First, chiller, it sounds like you’ve mixed more than one song in a single song file. I would find that hard to manage! I use a separate song file for each song. When the mix is done for each song, I render that song to a 32-bit wave file (to avoid losing any significant bits). I then open up a new song for mastering and import each of the 32-bit mixdown files, each one on its own track.

First I usually focus on each track individually (using “solo”) and do typical mastering things like boosting the volume a bit using realtively mild compression (e.g., 4dB boost, which I find I can do using PeakCompressor, which generally improves detail rather than squashing things. This compressor also includes a limiter, which helps with the rogue peaks and isn’t so time consuming.) Often it helps to normalize each 32-bit track before starting this process: it puts everything on a more or less equal basis, with a truly insignificant loss in fidelity. However, if I’ve mixed ideally, this shouldn’t be necessary. For this first step I use track plugins.

Then I work on the song order by dragging wave files (“parts”) left and right, leaving them on their tracks for simplicity. This way I can hear how one song sounds following another, and make sure that the relative volumes of the tracks in succession are what I want.

Finally I focus on the whole collection and apply (on the master channel) any effects I want to apply to the whole collection. (Frankly, so far I do very little of this, mostly due to inexperience and using the “less is better” approach.) But this would probably be a good way to use a harmonic balancer plugin, to help give all the tracks the right tone for the collection. You could also do a little more compression here – or a lot more, for some genres, though I’m not a fan of the over-compressed sound. Another good plugin to use for all tracks equally might be a tape saturation emulator or something like that.

Here’s the point that Jimbob made, although he added some of his own personal preferences that are not typical in the industry (he does mixing and mastering together, more or less, and most people don’t). There is no magical “master” button to take a collection of songs and make a good sounding CD. Normalize doesn’t do that: it only makes it so that the highest peak doesn’t exceed the specified limit (typically 0 dB). Peak levels have very little to do with perceived volume. A better meter for that purpose is an RMS meter, like the “Inspector” VST plugin. Furthermore, while a plugin might be able to adjust tracks to have the same RMS volume, that’s usually not quite what we want: we want the volume to sound “right”, where parts that should be louder ARE louder, etc. This is a matter of judgement, so no tool automatically does the job – at least, not yet.

Mastering is the process of preparing a mix for rendering to a medium. For more info on that, see this n-Track Wiki page: Mastering 101. Multi-band compressors are one of the big tools used for mastering, but I’ve never been able to achieve good results with one yet.

Back to your original question. I don’t know why you’re having trouble dragging tracks, unless they’re “locked” (which probably isn’t the case, because you have to do that manually). Can you slide them left and right? Is the cursor a plus sign with arrowheads when you try to drag? (It should be, that means you’re using the drag tool.)

Finally, you don’t want to normalize all the cuts together. When you’re done mastering, each cut should have peaks that come near zero, even though the volumes of different cuts may vary dramatically. When they don’t you’re just wasting bits. So, rather than mixing each song individually and ignoring peak levels (more or less), and then taking the group and normalizing them, we master each song individually to make it use the digital format the best it can (peaking near zero), while comparing it to other cuts on the CD (to balance the tone and levels).

In professional audio production, mixing and mastering are nearly always done by different people. One reason is that the tools and approach are rather different, and another reason is that it really helps to get a fresh set of ears on it. However, another common practice lends credance to Jimbob’s method (to master and mix at the same time): rather than providing a just a mix file to the mastering house, they often provide “mix-minus & seps”, meaning, a mix without bass, drums, and vocals, and submixes of those items, to give the mastering house more flexibility and control. But those pros spend countless hours and thousands of dollars on mastering. As a home recordist I’m not going to bother with that much work!

PS: I’m still using n-Track V3, and I’ve never used n-Track’s ability to burn CDs, so I don’t know whether my method works well with that feature – you might still have to collapse the files into one track as the last step. Which was what your question was in the first place!

Oh – I thought of something: n-Track won’t let you drop a wave part if it overlaps anything. Also, it has a habit of sometimes bumping the track to the right while you’re dragging it (to avoid overlapping a part you’re dragging it over). These two things can sometimes make it difficult to drag parts up and down between tracks. You can rearrange tracks by dragging the track header (where the track record button is) up and down. Sometimes this helps.

You should also be able to do nondestructive copy/paste but said it doesn’t work and I can’t imagine why not. Just drag to select, copy, drag where you want the file to go, and paste.

Jimbob, about compression sucking the life out: Try using much milder compression and a little more peak limiting. I agree that surgical envelopes are the best sounding technique, but there just aren’t enough weeks in a day. I used to do that but find that I get very pleasing results using PeakCompressor’s limiter, along with a little compression.

For example, ratio of 3:0 or 2:0, with threshold set around -10dB, limiter set at -4dB and gain set to 4dB. As a quick little trick to set the limiter level & gain, I stop playback, turn on “Auto”, and adjust the limter. The gain level tracks that adjustment. When I have them where I want them, I turn off “Auto”.

For the kind of music I’ve been recording, this works very well. That amount of boost is enough to get my tracks in roughly the same volume range as commercial tracks in the genre (Americana, e.g., Sugar Hill Records: Jerry Douglas, Sam Bush, Nickel Creek, Alison Kraus), and without any undesirable side effects that I can hear.

Interesting about the peak comp. You just want the ‘squash’ that you want, not more. I’ll try that!

I have a question about that “harmonic balancing” plug-in: Is it realistic to try and do this, get similar results as, say, ‘Har-Bal’ or Ozone, without laying out cash for a specialized plug?

Would a spectral display on a suitable EQ panel get you something close? Or the ‘Spectral-View’ in Cool Edit?

I haven’t used 'em yet, I’m pretty primitive at this point.

I hear good reports; that’s about all I can say, not having used it.

It might be as simple as getting the spectrum for the entire song, comparing that with the spectrum for the whole “target” song, and compensating accordingly. And I suppose you could do that “by eye” (if the wave editor has a spectral display for the whole song – and I think CE96 does, if you select the whole song) comparing the two and using a good graphic EQ. A bit tedious, I suppose. I can’t say whether that’s all HarBal does.

I’m forgetting what Ozone is. I know I know it, but don’t use it and can’t remember.

Ozone - mastering suite plugin.