|
|||||||||
|
Thread Tools | Search this Thread |
August 4th, 2008, 10:21 PM | #1 |
Regular Crew
Join Date: Dec 2003
Posts: 53
|
Audio Work Flow
I recently purchased a portable field recorder (Fostex FR-2le) and ran it through some tests. The sound quality is at least 2-3 times better then recording on to the DV tape. I had no trouble syncing in Audio Premiere either. It took a couple of seconds per clip and having the audio also from the caputred video helps speeds up the process big time.
Now for my question, how should I go about the actually recording of my video. Should I call or mark each take with the clapper and stop the video and audio recording between each take. Or just leave the video and audio rolling between each take? Which one will make for a better work flow? Any other tips or tricks for working with video and audio that are separately recorded? |
August 5th, 2008, 04:01 AM | #2 | |
Trustee
Join Date: Dec 2007
Location: Glasgow, Scotland
Posts: 1,546
|
Quote:
You may also like to check how long the audio will stay in sync with the video on single long takes. |
|
August 5th, 2008, 04:26 AM | #3 | |
Inner Circle
Join Date: Mar 2005
Location: Hamilton, Ontario, Canada
Posts: 5,742
|
Quote:
You should be identifying each take to make it easier in post production. Also sync is likely to drift signifigantly over the duration of a complete tape. Not to mention the waste of tape and recording space if you let it roll between takes. Besides, the "Roll it! ... Speed! ... Marker! ..." start up is so cool! LOL
__________________
Good news, Cousins! This week's chocolate ration is 15 grams! |
|
August 5th, 2008, 07:16 AM | #4 |
Major Player
Join Date: Jun 2008
Location: Dayton, TN (USA)
Posts: 219
|
Definitely stop between takes. Sync drifts over long takes is a royal pain to deal with. Trust me, you don't want to mess with it...
|
August 5th, 2008, 12:01 PM | #5 | |
Trustee
Join Date: Jan 2004
Location: Scottsdale, AZ 85260
Posts: 1,538
|
Quote:
Since posts here are being read all over the world by people trying to gain useful knowledge of the complexities of sound recording, I want to gently ask you to clarify your post. You claim "sound quality at least 2-3 times better" - a claim that will get a lot of people very excited. Yet, in my experience, that's virtually impossible to achieve by any reliable method. What, other than your subjective perceptions, might be "2-3 times better?" Signal to noise ratio? Frequency response? Convenience? Bass response? Mid-range response? I'm not trying to be harsh here, just to note that everyone on this board is trying to achieve something difficult - making recordings of unseen audio vibrations. To help achieve that - it's important to separate myth from fact, hyperbole from achievable results - and to employ language in describing this stuff that's as honest and accurate as possible. Also audio is terribly dependent and terribly complex. In my experience, that giant leap in results you felt you achieved might not have much at all to do with the MEDIUM you recorded to. And in fact, the inhereant audio response characteristics of DV tape are no where NEAR objectively THREE TIMES WORSE than the recording capabilities of the Fostex. Differences there might be. But in a somewhat narrow range of technical specs. So could other things be contributing? When you hear the difference, are you hearing the actual signal? Are you hearing the reproduction chain? Are you hearing the somewhat woeful playback pre-amps on the camera? Are you maximizing the recording capabilities of the direct to camera audio in the same way you're maximizing it when you run record to the Fostex? Just some areas to consider. I always applaud anyone's quest for better sound. But I also don't want the general population here to think that running out and laying down the coin for a Fostex (or any other piece of hardware) can actually instantly provide THREE TIMES the useful frequency recording accuracy - since even if that were possible, we'd be recording stuff that only dogs and coyote's could usefully hear. Again, don't mean to be picking on you. Just trying to keep the language used in this VERY technically hard task as accurate as possible so people have a chance to learn properly. Good luck in your tests. |
|
August 6th, 2008, 01:52 AM | #6 |
Major Player
Join Date: Oct 2007
Location: London, UK
Posts: 795
|
When recording double sound with my Fostex and XH A1 I sync the sound up in the NLE using the wave forms.
I have used a clapper board in the past but I don't see the point. As long as I've got sound from the onboard mic it's quick and easy to sync up in the NLE. I record long takes too because it means less syncing up. I have never found the tape and sound recording to go out of sync, even after half an hour. But I keep on reading comments in this forum that suggest recording short takes so it doesn't go out of sync. Am I just lucky?
__________________
http://www.gooderick.com |
August 6th, 2008, 02:38 AM | #7 | |
Trustee
Join Date: Dec 2007
Location: Glasgow, Scotland
Posts: 1,546
|
Quote:
|
|
August 6th, 2008, 04:04 AM | #8 |
Inner Circle
Join Date: Mar 2005
Location: Hamilton, Ontario, Canada
Posts: 5,742
|
You guys must either be incredibly lucky or the nature of your material is such that it has a greater tolerance for sync errors. You can be less precise in the sync in a long shot of a speaker at a podium than you can if you push in to an ECU where his face fills the frame or if you're shooting closeups of sticks hitting the drum heads and cymbals in a music shoot. You also can get away with less precision if the final propduct is for TV viewing versus for theatrical release.
__________________
Good news, Cousins! This week's chocolate ration is 15 grams! |
August 6th, 2008, 07:42 AM | #9 |
Major Player
Join Date: Jun 2008
Location: Dayton, TN (USA)
Posts: 219
|
Yeah, y'all are definitely lucky! I've had audio go as much as 1 full second out of sync over a 1 hour tape.
|
August 6th, 2008, 07:51 AM | #10 |
Major Player
Join Date: Apr 2006
Location: Cedar Rapids, IA
Posts: 563
|
Well, I'm one of the lucky ones, I guess. I record a lot of events, mostly comedy shows, using several Sony HDV cameras and an Alesis HD24 hard disc recorder. The most drift I've ever seen (with an 85 minute tape, no less) forced me to correct synch by two video frames (i.e., insert or delete a frame at two different places over the 85 minute clip). In other words, the drift was 1/15th of a second over an hour and a half. That was my worst case so far.
One second over an hour, as David mentioned, that would be a nasty postproduction challenge. Is it possible that part of the problem was caused by dropped frames or other problems? Clearly, modern timers in digital devices should be more accurate than that. If my watch had this kind of drift, it'd be off by almost three minutes every week. I guess I'll be holding on to my gear! :-) - Martin
__________________
Martin Pauly |
August 6th, 2008, 08:12 AM | #11 |
Major Player
Join Date: Jun 2008
Location: Dayton, TN (USA)
Posts: 219
|
Dang! What camera and tapes do you use? My problem was probably due to the fact that the specific instance I have in mind was a college graduation which was recorded on a GL2 and a VX2000, both of which are cameras that are frequently checked out by students for their personal projects and there is no policing of what tapes are used in the cameras (no consistency) and I doubt they've ever had head cleaning tapes run through them. No dropped frames though...
|
August 6th, 2008, 12:53 PM | #12 |
Major Player
Join Date: Apr 2006
Location: Cedar Rapids, IA
Posts: 563
|
A couple of Sony Z1s, and occasionally a Sony HC3 as a third camera. For tapes, I use Sony HDV tapes (DMV-63HD or DVM-85HD).
- Martin
__________________
Martin Pauly |
August 11th, 2008, 12:46 AM | #13 |
Regular Crew
Join Date: Dec 2003
Posts: 53
|
First up sorry for not responding in this thread since I posted here. I talked over my work flow with audio engineer the other day. It was an interesting conversation. With my limitations we both agreed that every take needs to be marked visually with a clapper and also logged then handed over to the person in charge of audio mixing.
The takes will be short but it's worth the repetition of the "call" to get everyone ready and do things professionally. As for the question about my jump in audio quality I upgraded my mic from Audio-Technica AT897 to Sennheiser MKH-416. Also the jump in recording quality to 96k/24bit allow me and the engineer to say the quality is quite improved. Im not an audio tech but I have ears and there is a difference. Sorry for the lack of language. I thank everyone for their input in this matter! |
August 11th, 2008, 01:09 AM | #14 |
Regular Crew
Join Date: Apr 2006
Location: Brooklyn, NY
Posts: 111
|
Don't tell a messageboard what format you're recording in! You'll get everybody all excited!
Seriously though, 96khz might be a bit much - 48 is pretty standard throughout the film world, at least for location sound. It also takes up half the room! |
August 11th, 2008, 08:30 AM | #15 |
Trustee
Join Date: Jul 2003
Location: Burlington
Posts: 1,976
|
When shooting a lot of short takes, I always found it easier to have the camera operator audibly call out the camera's timecode instead of using take numbers.
Take numbers can get confusing very quickly depending on your director and the material; with aborted starts, small shot-size adjustments of the same action, quick "do it again" without stopping the camera etc. Tail-slating of good takes with a quick comment can also be helpful when the pace of shooting is fast and your team is short-handed. |
| ||||||
|
|