Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: powermonkey on September 27, 2008, 10:59:27 AM
-
I pulled a couple of tapes at David Ford at the Union Chapel last night, one aud and one board.
Both are 24bit, but the board is at 24/48 and the aud 24/96.
I reckon with a bit of t.l.c I can matrix the two, and my attempt with one track thus far sounds great to me.... one question, though:
If I mix a 48k and a 96k source, should I mix the two to 48 or 96k? I want to do this one properly, as it's far and away (by MILES) the best tape I've ever made.
Thanks for your help, guys, it's always muchly appreciated.
:)
-
You would need to convert the 24/96 source to 24/48 before mixing them.
What type of audio editing software do you have?
I prefer Wavelab for mixing 2 sources. I used to chop one source into tracks and individual sync them up (to minimize the amount of drift between the timings of the recordings). I just recently learned that streching the time of the shorter source to match the longer source (or vise-versa) works much better and takes less time.
-
Ah, cool. Cheers for that.
I'm using Audacity on my little MacBook, and as I'm usually skint I think I'm stuck with it for a while. Seems to do the trick, though.
:)
The way I've done matrixes before is to keep the audience recording intact, track out the board recording, and synch each track one by one. It's a bit of a pain, but I've got pleasing results. This time I want it to be as close to perfect as I can get it, so any advice to make it better would be fantastic.
Hehe... the sound guy I was talking to before the show was asking about my rig, and when I told him I would be recording at 24/48 and 24/96 he just said 'You might as well set both to 44.1, surely? It's going to end up as a CD anyway, right?'... perhaps I ought to have done that!
:roll:
-
Ah, cool. Cheers for that.
I'm using Audacity on my little MacBook, and as I'm usually skint I think I'm stuck with it for a while. Seems to do the trick, though.
:)
The way I've done matrixes before is to keep the audience recording intact, track out the board recording, and synch each track one by one. It's a bit of a pain, but I've got pleasing results. This time I want it to be as close to perfect as I can get it, so any advice to make it better would be fantastic.
Hehe... the sound guy I was talking to before the show was asking about my rig, and when I told him I would be recording at 24/48 and 24/96 he just said 'You might as well set both to 44.1, surely? It's going to end up as a CD anyway, right?'... perhaps I ought to have done that!
:roll:
If your final project is going to end up on CD, it might do you well to record at 44.1 so there is no re-sampling involved. I suppose if you are not worried about that (I rather dislike re-sampling) it is not a big deal. If you do want to record at higher sample rates but are going to CD at the end, 88.2 could work very well. High rate, and easy decimation to 44.1
-
It never hurts to start higher and go lower later on IMO. I would convert the 24/96 source to match the other first. Then I'd matrix the two and leave as-is and/or convert to 16/44 for CD. There's really no need to chop the second source into segments. Since they are digital sources the difference will be constant. You can use different methods to figure out the drift and adjust accordingly. I haven't used Wavelab but Mike mentioned it above and I've heard good things about it. I know Goldwave has a time warp feature to adjust and leave the pitch intact. I use Sony Vegas and line it up visually. By zooming in extremely close I'm able to line up recordings perfectly.
-
id actually upsample the other one to 96 before mixing, but the difference would be subtle at best, if even noticeable.
more importantly is to resample one of them to match the other one EXACTLY. use some good defined pointsa at the beginnign and end of the recording to synch.
you may end up resampling your 48K to 96002 Hz to ge tit to match, some iteration is involved
-
It never hurts to start higher and go lower later on IMO. I would convert the 24/96 source to match the other first. Then I'd matrix the two and leave as-is and/or convert to 16/44 for CD. There's really no need to chop the second source into segments. Since they are digital sources the difference will be constant. You can use different methods to figure out the drift and adjust accordingly. I haven't used Wavelab but Mike mentioned it above and I've heard good things about it. I know Goldwave has a time warp feature to adjust and leave the pitch intact. I use Sony Vegas and line it up visually. By zooming in extremely close I'm able to line up recordings perfectly.
I am trying to do the same on a recording - one board and one aud, both recorded at 24/48. I can line them up in Audacity well but by the end of the show it is drifting. how do you figure out how to stretch and what is the secret? i also have audition.
-
It never hurts to start higher and go lower later on IMO. I would convert the 24/96 source to match the other first. Then I'd matrix the two and leave as-is and/or convert to 16/44 for CD. There's really no need to chop the second source into segments. Since they are digital sources the difference will be constant. You can use different methods to figure out the drift and adjust accordingly. I haven't used Wavelab but Mike mentioned it above and I've heard good things about it. I know Goldwave has a time warp feature to adjust and leave the pitch intact. I use Sony Vegas and line it up visually. By zooming in extremely close I'm able to line up recordings perfectly.
I am trying to do the same on a recording - one board and one aud, both recorded at 24/48. I can line them up in Audacity well but by the end of the show it is drifting. how do you figure out how to stretch and what is the secret? i also have audition.
I use trial and error, but that takes a long time with stretching because of the length of the show generally because it has to process it, put it into RAM, then show you the new file while maintaining a backup in case you want to revert back to the original. If someone knows how to do this more scientifically (sp) please post.