vanblah

OK. This is something that we should clear up pretty quickly.

Since we're going to be using different systems it's going to be VERY important that we all use the same method for creating the files. WARNING: boring, but important, pedantic drivel coming up:

We can't rely on everybody's BPM to line up ... it only gets you in the neighborhood. Every computer and application calculates BPM differently; besides computers have terrible rhythm (and that's not just because they are robotic). Hopefully Martian will tell us the tempo of the song (ie., 105BPM). But my computer will calculate 105BPM differently than everyone elses. There are devices which sync up different hardware clocks, but that won't work in this situation.

So how do we make sure the parts line up? There are a couple of things we can do.

So let's say that the song is 4 minutes long but your part doesn't come in until 1:30 and only lasts a few bars; so you're out by 2:05. You don't want to have to listen to the entire 1:30 for every time you try to do your take so you set your start point to a few seconds before that and then perform the part and then you hit stop and you're done. The resulting file is obviously shorter than the whole song and if you send that file it's going to be next to impossible for the mix guys to line up.

So, instead you go through the steps above but EXPORT the file for the length of the song. This means you end up with a file that has 1:30 of silence at the beginning and 1:55 of silence at the end (I think my math is right). This should be no problem to line up because it should be the same length as the entire song. The mix guy drops it into his DAW and boom! it should sound perfect, right? Not so fast there. We're going to be using MP3s for the initial sessions.

Some apps (SoundForge being most notorious) insert a second of silence before an MP3 render ... I don't know why. Different MP3 encoders do different things to the file so we can't really rely on that in the beginning.

So, in addition to the full-length file mandate, Martian gives us a four-click count-in at the beginning of the guide track and the mix guys can use that as an audio/visual cue to line up the various files that will be pouring in. You just copy/paste the four-click count-in to the beginning of your part so that the two line up.

When we make the switch to .WAV for the final mix, we won't have the same problems. But we should still submit the four-click count-in as well as the entire length of the song for each part. If we could use Broadcast Waves it would be even simpler, but the stable version of Audacity doesn't support it.

But, Doug, you say, if I have over two minutes of silence in my part that means I'm going to be taking up valuable server space for nothing. Isn't this just a waste of resources? I'm glad you asked. There is a free LOSSLESS audio codec called FLAC (which stands for: Free Lossless Audio Codec and can be found at FLAC - Free Lossless Audio Codec) that strips out the silence for file transfer and then can re-insert it at the destination (UNFLAC). OK, that's an oversimplification, but you get the point. If you decide to use it I would be very, very grateful. It's not a dealbreaker though; FLAC can be intimidating in much the same way early MP3 encoders were. We only need to worry about FLAC when we switch to WAVs though.

I'm sure I've missed something, and probably raised more questions than answered, but we'll get there. It's not too difficult.

1. Render/Export/Save the ENTIRE length of the song regardless of whether your part lasts that long.
2. Copy/Paste/Lineup the four-click count-in for each part.

BTW, for those of you who don't have FTP clients (or don't want to) I have set up a rudimentary Web based PHP/FTP page. I'll PM the link to everyone shortly. I'm not sure how well it will work, but it works for me so far. I haven't tried to transfer a full size WAV though.