[Separate audio/video recording] Avoiding drift?

Submitted by Winfried on Fri, 03/30/2012 - 18:09

Hello

Ultimately, I decided to go for the [="http://Tascam.com/product/dr-40 Tascam DR-40[/]="http://Tascam.com/p… Tascam DR-40[/] and will buy an affordable camcorder so I can record lectures and interviews for non-professional use.

I have no experience with audio/video editing, and read the following:

Any two digital devices that are not physically synchronized with a cable transmitting time-code and/or word-clock are invariably going to drift. [...] Ultimately, if you’re not running a physical sync connection (and are recorders that can read and/or write it) you’re going to need to do some manual re-syncing during a project. Usually, that means making some cuts to the audio and nudging the audio forward or backward. There are software solutions for this too – such as VocAlign that can automatically sync a separate recording to scratch audio printed to the video. ([[url=http://="http://transom.org/…"]source[/]="http://transom.org/…"]source[/])

If this issue is likely to occur when recording one- or two-hour segments, what can I to do solve it? I read about an application called http://www.synchroa… VocALign[/]="http://www.synchroa… VocALign[/]: Is this is a good alternative?

Thank you.

Comments

Well you might be able to utilize that software, it really won't be necessary to bother with that nonsense. All video, from the VHS & Betamax days our own machines that while they are not utilizing actual timecode are in fact still locked via Crystal clock synchronization. That is to say, their speed is so highly accurate, once you get non-interconnected cameras transferred into your video editing timelines and once you synchronize them, even within a 1.5 hour-long production, you may only experience 1-2 frames of drift. And that's easily corrected by simply finding where the drift becomes noticeable, backing up a minute or two, splitting the track and re-synchronizing it. It's easy, it's fast, it's consistent. I have been doing this for quite a few years even before the days of digital video recorders. No big whoop. All of your audio will be at 16 bit, 48 kHz though it could also be 24-bit all the way up to 96 kHz. 16 bit, 48 kHz is a very well established audio for video format so you must make certain you are working in the 16 bit, 48 kHz which is what 99% of camcorders are recording. Besides, in order to use that other software, you would need to be running Avid, ProTools also. So spend more money if you want to but it's highly unnecessary.

Timecode is more important when you are utilizing edit decision lists from an off-line editor.
Mx. Remy Ann David

Thanks for the clarification.

So that quote above only applies to extreme cases and I should be OK.

I have a 1h30 segment that I recorded yesterday as my first attempt at recording audio and video independently, so I'll soon have an opportunity to check for any drift.

Is there a good, free or affordable video editor you would recommend for Windows?

I played a bit with VirtualDub and AviDemux, but find them a bit so-so (VD only works with AVI; AviDemux is cross-platform so a bit slow to start).

Sony makes a wonderful product called Vegas. That program originally started off as a multitrack audio program. It was later improved and augmented to include video. So it's a multitrack audio and video program. The full-blown version costs upwards of $600 US. But here's a great part. They make a much more affordable version that is scaled down called Vegas Studio (or something to that effect) that can be had for a little over $100 US. It can absolutely do what you need done. It only lacks some of the more complex features designed for more complex productions. It's like any other LE or SE version of a multitrack audio program so it's still very capable. And that's what I recommend for you because it's a professional product which offers much finer control over what you need to do instead of those clunky free consumer oriented conversion programs you are attempting to utilize. Unfortunately it's not a program you will find bundled with anybody's hardware so you will have to purchase it.

Pinnacle also offers some decent consumer oriented video editing programs that can accommodate an additional audio track or two but I find it to be far too limited for the kind of productions in which you and I do. Of course there is also Adobe Premier Pro & Adobe Premier Elements. The Premier Elements program is much more capable than Pinnacle's Studio and is also quite affordable in that $100 price range in the US. I used to use that program for everything I did before I switched to Vegas. Adobe's program converted all of my 16-bit audio to 32-bit which I did not want it to do. There was no way to disable that feature and that's why I stopped using it. Vegas doesn't care what you do or what you use. It keeps what you've selected intact the way you select it. Things are only converted if you tell it to convert things. I'd rather have programs that work the way I want to work not the way they want me to use it. So screw Adobe. They told me it doesn't hurt your audio but I don't care. I want it the way I want it.

This Vegas never has any losers.
Mx. Remy Ann David

Thanks much for the infos + feedback. I don't know which consumer-grade camcorder I'll end up buying, but I was recommended to check Canon, Panasonic, and Sony.

Using two different pieces of equipment and doing some basic post-production work isn't as easy as just pointing a camcorder, but I find models that support mic input (+ headphone output) a bit too expensive for personal use. Maybe even a Flip on a tripod could be good enough to shoot lectures and such :-)

When doing multi-camera shoots it's always nice to have matching cameras. Sony's look way different from Canon's which look way different from Panasonic's. So they can become awkward in postproduction.

When shooting multi-camera shows, one should start the cameras and never stop them until the end of the show or a set. Otherwise, you'll be synchronizing tracks until next year.

I've been synchronizing cameras for 20 years now. I don't think I'll be stopping anytime soon?
Mx. Remy Ann David

Sorry about the misunderstanding: When I mentioned the three brands, I didn't mean that I would buy more than one camera, but rather that those brands were recommended to me. This is just to record lectures and interviews for my personal use, so a single camera is all I'll operate.

I guess the Flip-type camcorders like Sony's [="http://www.sony.fr/hub/appareil-photo-hd-mobile"]Bloggie[/]="http://www.sony.fr/…"]Bloggie[/] line of products aren't a good idea to make two-hour-long recordings.

What about SD vs. HD? Is HD really needed? I'm more concerned about lighting condition being too poor, and contrast being too important when I need to go from speaker to slides on the screen:

Winfried, post: 387454 wrote: I find models that support mic input (+ headphone output) a bit too expensive for personal use.

There are very affordable ways to use very good mics with very ordinary camcorders. [="http://beachtek.com/"]Beachtek[/]="http://beachtek.com/"]Beachtek[/], and [[url=http://="http://www.studio1p…"]Studio1[/]="http://www.studio1p…"]Studio1[/] are the first devices that come to mind - I'm sure there are others, but these are the two I'm most familiar with.

If I were recording lectures & interviews that didn't require more than 2 tracks of audio (and wanted to avoid the unnecessary hours of post-production) I would consider recording the audio straight to the camera. In my opinion, unless you need more than 2 audio inputs, a standalone audio recorder needlessly complicates (and lengthens) your job.

The problem is that the first camcorders that support a mic entry are about €800-1,000, so I figured I'd just get a compact sound digital recorder, an OK entry-level camcorder, and perform some basic post-prod editing on my PC afterwards.

Thanks for the tips on [="http://beachtek.com/"]Beachtek[/]="http://beachtek.com/"]Beachtek[/], and [[url=http://="http://www.studio1p…"]Studio1[/]="http://www.studio1p…"]Studio1[/]. I'll add those items to the list of microphones to check. I'll get a shotgun/cardioid next week.

I don't know the best places to shop for this sort of thing in France, but the Canon consumer cameras are the best equipped when it comes to mic inputs and headphone outputs. The Canon VIXIA HF M40 in particular looks like a good bit of kit for the price.

Although your local source would have the PAL versions, you can use this [URL="link removed[/URL] site for quick reference. The models shown all have mic inputs and many have headphone jacks.

I wouldn't trust just using the camcorder for recording the sound while recording picture. Recording the sound from the dedicated solid-state recorder as a Primary should also be performed. You never know when you are going to have some kind of glitch or drop out? And with lectures, you don't want to miss a word. Coming from a network television background, we always practice redundancy. So that's also where that secondary camera shot would be utilized and the soundtrack from your solid-state recorder.

Zooming in and zooming out is also Obnoxious to look at and wreaks of amateur hour. That's where that cutaway camera comes in. You switch to the cutaway camera when you are zooming your primary camera. Then you go back to your primary camera with the newly framed shot. You may also need it if there are any screen projections. Then you go to the cutaway camera while framing your main camera again for the screenshot, blackboard, whiteboard, computer displays. Panning and tilting a camera on a tripod without a "fluid head" is 10 times more of noxious looking. And that's why you want that secondary shot that's locked down generally. You might even want that shot to not include a view of their lips? That way, you can use the cutaway whenever necessary with or without synchronization. Since you have already indicated you're going to do some postproduction in your computer, this is the way to go. You want to be a professional and this is the way professionals do it. You don't want to use your main shot camera as a cutaway due to a video glitch since there will be no lip sync. And that secondary cutaway camera could be just about anything since you're only going to go to that shot just for a couple of seconds. You'll also be able to get away with a far more affordable main camera if you're not worrying about trying to get quality audio Into or,out of a common consumer camcorder or even a cell phone. That's a lot of cabling and a pain in the butt. Otherwise it could get so poopie looking & sounding, that you'll destroy your reputation as a professional. This can be a good source of income for you so don't screw it up.

The solid-state recorder can even make your life so much more easier. You could even simply drop it into their breast pocket and if it's small enough, even in a woman's brassiere (guide her as to where to place it. Don't do it yourself unless she wants you to. Then you may also have to take her out to dinner later?) You might also want to purchase an inexpensive lavalier/tie tack microphone to plug into the solid-state recorder. Then the solid-state recorder can simply be placed in any pocket. Make sure your solid-state recorder is set to ".wav" and not MP3. You also want to set to 48 kHz sample rate. 16-bit is perfectly fine, perfectly common and easy to utilize in most any program. Also, with spoken word, if you're going to do some postproduction in your computer, it would also be a good idea to include some limiting to smooth out the lecturers voice. And before long, people are going to ask you to start doing all sorts of other video work. So whether it's just a lecture or not, you want to put your best foot forward. Keep your cabling and set up to a minimum. You're at the back and there at the front which means you'll be taping down cabling all the way back to you. Why bother if you don't have to? So, let's review... crappy consumer camera in the front. Crappy consumer camera on the side. Simple solid-state recorder in their pocket eh voilà!

If possible, you may also want to place the camera into manual focus and/or manual exposure control. The focus can be confused quite easily by somebody in the front row getting up to move and stuff like that. Automatic exposure settings can wash out bright items such as a large sheet of paper they may be drawing a diagram upon making it unviewable. You can also set your automatic exposure to tell it to cheat the aperture down .5 to 1 full smaller aperture to keep the whites at 80% instead of 100% video. That's what I do. Because once you've had 100% video, nothing is recoverable from that. It's the visual equivalent to pure audio distortion which I think is an oxymoron? There is nothing pure about distortion when it's distorted. It's easier in software to increase the brightness if necessary and really not effective when you're trying to lower the brightness level because detail has already been lost. Also remember when shooting female lecturers, you might get arrested? The same holds true for shooting male lecturers. So just run the camera, please. No shooting.

Lecherous lectures
Mx. Remy Ann David

You've been given some fantastic advice by Mx. Remy. Specifically white-balance, disable most anything that's "auto", and avoid stopping the recording once started.

I'd add if you're looking to stretch your budget and buy an older tape-based DV camcorder, use a brand new high-quality tape every time. Most drop-outs occur at the beginning or end of the tape, so get it rolling at least 2 minutes before the event starts. If it's a new camera pick a brand of tape and stick with it, I believe you'll have fewer issues if you always use the same tape. Archive the tape, because 2 years later someone will ask if you still have a copy. It's not like the old days of expensive 3/4" video tape, miniDV is cheap - use it once and archive it - is my policy.

In your case for a situation as you've described to, 'record lectures and interviews for non-professional use', I'd have no qualms recording the audio & video to the camera if the proper interface was used and the levels monitored. If you want to record the audio separately, by all means, do both.

If at some point professional results are expected, you should be compensated accordingly. When you start semi-pro recordings you can worry about redundancy. I'm a fan of double or triple redundancy. When I record an event I'm usually doing 3 SD cameras mixed on-the-fly through a video mixer - with that mix going to 2 (sometimes 3) separate digital video recorders. Plus, each camera is recording to digital tape - so no matter where a glitch might occur, I have a back-up. Or if I decide later I should have taken a different/better shot, I can lift it from the camera and drop it in during post. For audio, the live on-the-fly stereo audio mix also gets recorded to those DVRs simultaneously, in addition to a standalone studio quality CD recorder, that mix is usually all I need. And when applicable, there is also a multi-track 24-track recording done from the audio mixer via firewire to a laptop, plus via Direct Outs to an Alesis HD24. (everybody, but the CD recorder, running at 48k so it can drop into the video without conversion).

I'm a fan of double or triple redundancy. I'm a fan of double or triple redundancy. I'm a fan of double or triple redundancy. But then again, I'm getting paid, this event is only going to happen once, and it's my job to capture it all and deliver a glitch-free finished product.

To address one of your other questions that we have overlooked, I'm still shooting SD - mostly because I can't justify the expense of an HD video mixer. Nobody is complaining (yet), but it's only a matter of time. If I was a one-camera run and gun videographer I'd be using an HD camera. HD isn't just the future, it's here. I don't think Standard Definition will be viable much longer. SD 4:3 is gone, SD 16:9 will be next to go.

Good luck (or should I say, 'bonne chance'?)

I just installed the trial version of Sony's Movie Studio HD 11.0 build 42, but it complains that it doesn't support the following file that I recorded with the camcorder:

Video: MPEG-1/2, resolution 720x576, Frame rate 50

Sound: A52 (aka AC3)(a52)

What would be a good way to convert that file to something Movie Studio supports, before replacing the sound channel with the WAV file that I recorded with the Tascam digital sound recorder?

Thank you.

You know that's a good question. I've actually experienced the same thing with the Sony software. It can create AC 3 audio files but it doesn't want to read them. For Christ's sake! They invented this stuff! It's their format. They created it. You can create but they don't want to read it. Go figure? And on their equipment and software.

Interestingly enough, the other Sony software that creates DVDs, Sony DVD Architect has no problem reading the AC 3 file. So what I've done as a workaround is simply import the raw production into Sony DVD Architect. I have to create a DVD which I can then rip. And when I ripped the audio, I can read that as ".wav" or, ".mpg". Then I can import it into Sony Vegas Pro 9. I know, that's idiotic but that's the workaround. You indicated you are using version 11 which I would think at that point should be able to deal with AC 3 files? Obviously the little Japanese guys are not thinking clearly. Probably too much WASABI?

My favorite are UNI microphones and UNI sea urchin caviar sushi.
Mx. Remy Ann David

I personally wouldn't record audio directly to a camera unless it is a really high-end one. Small camcorders only utilise 3.5mm sockets and have lousy preamps. Larger professional models have XLR inputs and better preamps, but still not as good as those found in quality portable recorders. Record audio to the camera if you wish, but also record it to a portable recording device as well. Syncing is an issue that you will want to think carefully about "Before" you record anything. I wrote a tutorial about this stuff a while back. IT may give you some good pointers to go along with the great advice already on this thread. IF you want to read it, you can find it here :
[[url=http://[/URL]="http://rocksuresoun…"]Sound For Film and Video: The Importance of Getting Good Audio[/]="http://rocksuresoun…"]Sound For Film and Video: The Importance of Getting Good Audio[/]

dvdhawk, post: 387699 wrote: 50fps isn't the standard for DV PAL is it?

I don't know about the PAL thing. This is what MediaInfo says about the MOV file:
Format : MPEG-PS
Overall bit rate mode : Variable
Overall bit rate : 6 120 Kbps

Video
ID : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Bit rate mode : Variable
Width : 720 pixels
Height : 576 pixels
Display aspect ratio : 16:9
Frame rate : 25.000 fps
Standard : PAL

Audio
ID : 189 (0xBD)-128 (0x80)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Muxing mode : DVD-Video
Bit rate mode : Constant
Bit rate : 384 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Delay relative to video : -80ms

Do you have the AC3 codec in the right place?

I did install a few codec packages, but I'm not sure. How can I check (XP)?

rocksure, post: 387705 wrote: I wrote a tutorial about this stuff a while back. IT may give you some good pointers to go along with the great advice already on this thread. IF you want to read it, you can find it here :
[[url=http://[/URL]="http://rocksuresoun…"]Sound For Film and Video: The Importance of Getting Good Audio[/]="http://rocksuresoun…"]Sound For Film and Video: The Importance of Getting Good Audio[/]

Thanks for the link. The WAV file recorded by just putting the Tascam in front of a loudspeaker sounds very good. Ideally, I should connect it to the sound mixer, but it's usually off-limit to attendees.

The issue at this point is how to combine it with the MOV file.

Next time, when rolling the two appliances, I'll probably play a short clap or whistle that I can search for in the editing application so as to make it easier to combine the two files.