Hi all - haven't posted here in quite some time. We're having a bit of an issue with a recording that maybe I can get some ideas about...
Basically the situation is this: recording a demo/ep with a trio of mine, none of us seemed to have time to get together all at the same time, so we've been doing the passing files back and forth thing, and doing our takes on our own DAWs. Actually, the other two involved pretty much did all they're takes on the singer's system, so it's really just me that did bass tracks on my Logic5/AardvarkQ10 PC.
Believe it or not, but everything sounded great up until nearly the end - we used a rehearsal take as a bed track so everyone did in a sense play together. I wound up laying down the bass (upright double bass, with a healthy amount of rockabilly style slapping) last, over top of the rest of the tracks - I felt I'd be able to "glue" the other instruments together just fine that way.
Anyway, finished my tracking against a sub mix (one stereo track) of banjo, guitar and vocals. Mixed it down quickly on my system, just to get a sense of a) how my bass sounds were (first time Micing a double bass for slap style) and b) just to be able to listen to the whole shebang in the car and judge the performance better that way.
Everything sounded killer.
Now, in order to get the bass tracks (two mic tracks, dry - and one sub mixed with some light fx) into a format with a consistent start point (and also because there was some comping), I soloed and bounced each track from the smpte 1hr point, and put them on a disk and passed the whole thing along for the singer/songwriter to mix on his Cubase system.
Sorry for the long post, but we're getting to the meat of it here: So he sent me the test mixes, and there are major timing issues with the bass tracks. Listening to the mix I did on my system, I must say I nailed the feel I was going for on all the songs. From listening to the Cubase mixes, I sound like a drunk. At best. Kidding aside, the bass takes are really behind the other tracks, timing wise...
He's tried moving them around a bit, and pushing them about 24 ms forward seems to help, but it's still not quite there. So that tells me that it's not just a case of me hearing flaws in my playing - they're genuinely off, and quite a bit so.
So I'm looking for brainstorm ideas on how to fix this, other than just push the regions around until it "sounds right". I'm clueless as to why this would happen (although I have some ideas). I do know that Logic has some input correction and plugin delay compensation functions, but as far as I can tell, these are disabled at the moment. I'm not sure how fast buddy boy's system is, but again we're not dealing with tracking on his system so really any latency issues his interface (an Mbox, I think) or PC has shouldn't come into play, right? Anything else we could try? Help!
Comments
Thanks for the reply. Sorry, I should have clarified in the OP
Thanks for the reply.
Sorry, I should have clarified in the OP about the smpte start point. I had the singer dump down his tracks from a single point ("0" bar on his system) , and placed that file at "0" on my system, which is exactly where I started bouncing my tracks from, and then he in turn would import my tracks back to the start point that he bounced his from.... We're not sync'd to the same clock, but in my mind this should have kept everything synchronized.
It looks like we're going to have to just nudge the regions until it's right. I spent a lot of time on my takes, playing around with playing right in the pocket, ahead, behind etc., so I'd really prefer to know why importing/exporting like we did gave us the (less than stellar) result we got.
If you were to record into a single session into any quality DAW
If you were to record into a single session into any quality DAW, the DAW itself would generate the offset required to sync everything up. This offset is based upon the processing and sample buffer value. Also, single session is dealing with a single self generated "time code" rather that multiple machine generations. You have multiple computers generating multiple sessions-none of them with an actual time code generator. As you export and reimport into DAW's with different buffer rates you have to nudge different amounts. The nudge will not necessarily equal the buffer setting of the interface at all so don't get trapped with that idea either. Also, different DAW's will implement things in slightly different ways which can affect things as well.
Now, that's just the technical aspect. Most musicians have a real tough time playing metrically perfect and this is shown to be true whether it's a pop musician or a classical symphony musician. This becomes apparent when recording against a click track or a rough mix. People rarely can repeat something twice.
Long story short, you just have to spend the time nudging the tracks. Many times you can right click on the track and be more exact with your offset too. This is what a pro studio does when they work tracks they haven't done themselves.
You just need to keep adjusting the timing until it lines up. T
You just need to keep adjusting the timing until it lines up. This is just an obstacle resulting from everyone tracking individually on three (or at least two) different DAWs. Don't worry about the number of milliseconds you adjust. Just move it until its correct. The time label is plain irrelevant as is the smpte from your original. Your time code isn't necessarily your buddy's time code unless it is received from the same clock.