I am self-recording /mixing an album of my music that, once the mixes are complete, I want to have professionally mastered. I have never utilized a pro mastering engineer before, but I've been reading-up on mastering (including reading Bob Katz's great "Mastering Audio" book, reading things in this and TOMB forums, and reading things on John Scrip's site).

I am working in Digital Performer 5 at 24bit / 96k, and mixing in the box. And, the album will need to be mastered for digital release (e.g., CD, not vinyl).

Most of the songs on the album have been written and are being mixed for "gapless" playback, from track to track, aka crossfading between tracks. So, for example, a musical phrase at the end of one song continues into the next song.

Given this, what are good ways to prepare my mixes for the mastering engineer?

Should I create a single contiguous mix of all of the songs, produce a single stereo file, and provide a document listing the track start / end times? Or, might it be better to deliver a separate stereo file for each song?

I understand that mastering engineers like extra time at the start and end of tracks so that they can handle the fades and ensure that the track cues line up with the CD data frames, etc. But, I also need the track start / end times to match the timing of the music, which spans the track-breaks.

So, I can see how it would be advantageous for a mastering engineer to work with separate tracks, but, if this were the better approach, what might I then need to do to make clear to the mastering engineer how the tracks should fit together musically?

Any comments?

I would send both the individual files and the continuous file to the ME. Include a continuous file that contains the fades and track markers exactly as you want them as a demo for the ME (keep this in 24bit and he might end up using it). Include very detailed notes about where you want the crossfades and track markers. However, since you are wanting precise placement of the fades/markers, you should try to arrange to attend the mastering session so you can help direct that particular component.

On heads & tails -

If you're careful (and there's no reason you shouldn't be when this type of accuracy is available) cutting the heads at 10 frames is plenty.

Some M.E.'s (myself included) will place start points several (maybe 10) frames out from the start of the audio anyway. There are still an awful lot of players - CD, MP3, etc., that will ramp up volume or cut off the first 50ms or so. Having that 10-frame buffer at the front, you're still getting the feeling of "instant on" when you hit the "play" button, but you won't chop the head off the track.

It can get a little trickier on crossfades - Case-by-case. I still shoot for 10 frames, but if there's an obvious chunk of the previous audio, I might slip down to 6 or 7.

On the tails, there isn't anything wrong with cutting at or just beyond the "out point" of the file.

And I'm with drbam - Send individual files and a reference file - Should be easy to mimic.

Thanks - very helpful to have both of your input.

I have a couple questions about the 6-10 frame gaps:

Assuming this definition in Wikipedia is correct, each frame translates to six stereo samples at 16-bit.

So, 6-10 frames / 36-60 samples at 44.1k is less than 2ms?

I am guessing that, although I am working in 24bit / 96k, the gap time is the same duration, e.g., 60 samples at 96k gets converted to 60 samples at 44.1k for CD-A?

In any case, I don't think I have to worry about this with my mixes--just good to know.

Because some of albums I listen to have songs that run together (I've been listening to Bowie's "Hunky Dory" lately), and these albums were originally mastered for vinyl, I am curious about the "gapless" track breaks on LPs--do they also introduce a very short <5ms gap, or is there just something done with the vinyl so you can see the cue?

iTunes and one of my CD players mess up these recordings, and add space between the tracks. Grrr....! But, I've heard this is fixed in the new iTunes...

quote:

But, I've heard this is fixed in the new iTunes...

As I understand it, the new iTunes allows you to link songs together as you're importing them, so that they end up as one long song. I think there's still a gap between individual songs though. If that's the case (please correct me if I'm wrong), it's useless.

I'm curious about this frame business too.

I haven't updated my iTunes to v.7 yet, but this Apple support article says that gapless playback is on by default, and there is also a new checkbox, "part of a gapless album", that can be applied to individual tracks.

Sounds good. But, some of the other descriptions I've read suggest that maybe iTunes is creating new files to accomplish this. It seems more complicated than other players (and the older versions of iTunes) which have the same duration gap between all tracks.