Gear / Technical Help > Post-Processing, Computer / Streaming / Internet Devices & Related Activity
Syncing R-44 SBD/AUD Matrix in Audacity (OSX) - have read sticky already
acidjack:
As noted, I have read the sticky on this already, but I think my question is slightly different.
I have an SBD/AUD matrix recorded on an R-44 with the mics back at the board. The board is maybe 100ft from the stage. As I understand it, there is about a 1millisecond difference for every 50ft one is from the stage.
I hear very little noticeable echo, but certainly looking at the waveforms, the peaks on the aud look visually a touch "later" than the SBD.
The sticky describes lining up two recordings based on samples, which will not work for me, since both have identical numbers of samples given that I used an R-44 that recorded both exactly simultaneously.
My questions are:
1. What is the best way to "line up" the recordings - is there any feature that I'm not thinking of, or do you literally just have to zoom way in and eyeball the peaks?
2. Assuming that the answer is to chop some time off of one of them, should I chop the SBD or the AUD? Another TSer who I trust said to chop 2msec off the SBD- but this seems counterintuitive to me as the AUD is the delayed signal. Shouldn't I chop a tiny amount off of the AUD?
3. I have loved using Audacity and it has always met my needs, but now that I will be doing more 4ch work, is there another program that makes this much easier? (I'd especially like to hear from people who have tried it with Audacity and another program).
The most difficult thing for me, and the other thing that the very helpful post in the sticky didn't quite address, is again, how you can measure the differences. I am frankly not that comfortable trusting my ears to determine whether it's off (especially in this case, where the delay is not very significant given room size). I also am not entirely clear on the best way to measure the distance between two points in two separate 2ch stereo recordings. I see how to select, say, a length of music in one 2ch recording and have it tell me the length, but I'm not sure again, because the number of samples is identical, that that is apposite to telling me how far "off" the two samples are.
Any help much appreciated. I know there are always a lot of questions about this kind of stuff.
Walstib62:
I don't have screenshots, but there is a time shift tool in the upper left. It looks like L/R facing arrows. Zoom in to get a good view, then select the aud track, then select the time shift tool. then just simply slide the entire waveform L/R to line the 2 up. You then do a slight cut of both at the front end, and you are good to go.
Todd R:
I don't know about the specifics of audacity, but in general:
Sound travels at about 1,000 ft/sec, so it takes about 1 ms to travel 1 ft. So if you are recording 100' from the stage/PA stacks, then your aud recording will be about 100ms delayed from the soundboard recording, not 2ms (ie, 1ms/foot, not 1ms/50'). The soundboard recording will be getting the information at the speed of electrons, on the order of the speed of light, and practically speaking, instantaneously. The aud recording gets the information of the band playing from the PA stacks, via soundwaves, and hence it is delayed by the time the speed of sound takes to cover the distance you are from the band/stacks.
Forgetting the time shift capabilities, as you mention, the easiest way is to just chop 100ms from one recording. And yes, you are right, you should chop 100ms from the aud recording.
Think about it this way: if you started the R44 exactly 500ms before the band started (and pretending there is no sound at all in the room except the band), the soundboard recording would have 500ms of dead air and then the band would start playing. The aud recording would have the 500ms of dead air from you hitting record a bit early and then another 100ms of dead air while it waited for the soundwaves from the stage/stacks to reach the mics. If you chop off 100ms from the aud copy, then both recordings would see the music start at 500ms into the recording.
If you chop 100ms from the board recording, then the board recording sees the music start at 400ms into the recording and the aud recording sees the music start 600ms into the recording, doubling your problem.
page:
--- Quote from: Walstib62 on January 25, 2010, 12:30:25 PM ---I don't have screenshots, but there is a time shift tool in the upper left. It looks like L/R facing arrows. Zoom in to get a good view, then select the aud track, then select the time shift tool. then just simply slide the entire waveform L/R to line the 2 up. You then do a slight cut of both at the front end, and you are good to go.
--- End quote ---
Yeah, the "line up on samples" bit is intended to be more of a "line up on the visual peak samples" and not a "line up on the start/stop samples".
Which ever one you chop I think is negligable, but I don't ever use the entire wav form (I'll start recording 2 or 3 minutes before the first song and leave a tail of maybe 30 seconds after the encore and any announcements I think are interesting). Render your wave form after adjusting the time difference as Walstib62 described and then chop stuff. As long as you started recording more then 5 seconds before something you are interested in keeping, you can just chop off a hand full of seconds at the beginning and end to resolve it.
I'd need more information to address your third question. I use audacity frequently, but I'm not quite sure what your trying to get at.
edit: expanded #2.
SmokinJoe:
Yes, the delay is more like 1ms per foot, and you chop it off the front of the AUD waveform to visually align.
Some people say "align the peaks". Others say find a quiet spot and then "align the beginning of a ramp up." I'm not sure there is a single right answer. I generally try both and then listen to the cymbals. If the cymbals sound like someone is rattling a coffee can with a few coins in it, keep trying.
I'm not sure if you are indoors or outdoors, but let me throw this out. When outdoors (like a music festival) at a distance, if there is any wind at all, it seriously screws with your ability to line up your wave forms. It has to do with that swirly phasing you hear when it's really windy, and it only effects the AUD tracks. You'll find one spot... here I need .100 to align it.... another spot I need .080 to align it. Everywhere you look it's something else. Which do I choose? I don't have a good answer for that one. The best answer I have is "pick a spot in the middle" and don't try a 50/50 SBD/AUD mix, because that maximizes the error and it will sound like crap. Go for something like 80/20 mix where the AUD is 6db lower than the SBD. This way the difference becomes less noticable. Another time I did one where the songs where 80/20 SBD/AUD, and then the applause between songs faded to 20/80 with more AUD. That was tedious, and I doublt I'll never do that again.
Navigation
[0] Message Index
[#] Next page
Go to full version