Jump to content

16bit vs 24 bit recording


Michael P Clark

Recommended Posts

I started this in another thread and figured it deserves it's own topic. On a film job I am use to recording in 24/48. But for video jobs in which the cameras are recording 16 bit, mainly the HVX200, I've been running 16/48. I've been told that when they start their session, in Avid, or FCP they create a 16/48 session, because the camera has that in the files. My thought, and maybe I'm wrong, was to keep the audio the same as the session ultimately created (16/48) so that there is no conversion of my audio if they need to use the 744t audio. Down converting from 24b is better than upconverting to 24b, and since it is not the sample rate being converted, I guess there wouldn't be any sync issue!

Based on what I've read here on this board, am I to assume there is no issue with the bit rate conversion like there is with the sample rate? I hope I am not rehashing a topic already covered in previous posts, but I don't remember reading it anywhere else.

Thanks

Link to comment
Share on other sites

There are conversion issues potentially with bit rate, 16 bit or 24 bit, but your reasoning for sticking with the 16 bit of the camera audio and the picture editorial system (Avid or FCP) seems to indicate some confusion as to whether you are shooting true double system, picture and sound, ande whether the audio post work will be done in the picture editing suite with the Avid or the FCP system. I fully understand that these delineations are becoming increasingly blurred for all jobs, with the exception of traditional double system FILM and SOUND, but you need to clarify how you expect the audio on the 744T is likely to be used. If it is really only a safety backup recording to cover any possible problems with the camera audio recording, your choice of 16-bit makes sense. If, however, you believe that the tracks from the 744T are to play a substantial role in the soundtrack and the sound will be completed with ProTools, for example, choosing 16-bits would be a mistake. The substantial difference in sound quality from 16-bit to 24-bit should not be ignored. If, however, the project is never going to leave the 16-bit arena, your choice of 16-bit on the 744T is appropriate.

Regards,  Jeff Wexler

Link to comment
Share on other sites

When I first got into digital non-linear recording, I started out always recording at 24bit -- it only made sense to me to get the highest quality sound I could, but after speaking to many editors (at facilities like SoundOne in NY) they were saying that they prefer the 16bit files because it saved them so much time (smaller files, less information to upload and process) and they argued that the difference between the two in-so-far as a dialog track is concerned would be negligible at best (especially since everything winds up being converted down to 16bit anyways.)

It really crushed me to think that, though we've increased our technology in the digital realm in an attemt to get as close as possible to the real thing, we've lowered our standards in a sense to save a little time.  I argued, but eventually acquiesced as I was outnumbered and couldn't really seem to sufficiently argue my point.  I was assured that with orchestral recordings for instance, it's still definitely worth getting the extra bits of info, but dialog tracks should be recorded at 16 bit -- especially with the advent of multitracking (again, the editor's arguement is that there is just too much information -- what used to take a few hours now takes all freakin' night.)

I can kinda see the reasoning behind that way of thinking, but isn't it ultimately lowering the standard?  Who places the bar nowadays?

Should we as the Production Sound Recordists simply put our (feet) down and refuse to record at anything less than the highest quality we're able?

Link to comment
Share on other sites

Should we as the Production Sound Recordists simply put our (feet) down and refuse to record at anything less than the highest quality we're able?

The answer to the above is YES...  we should record everything at no less than 24-bit rate. It is not that we make any pronouncements about "refusing" to record at 16-bit, it really just takes a simple statement that 24-bit/48K is the current industry-wide standard for high quality production sound recordings and that is what we're going to do. If the picture department or even sound editorial requests something less because they have a budget/time/storage/system problem, that's their problem. The equivalent, in the old analog days, would be a request from a production that you use a modified Teac recorder, or worse, one of those sync enabled Pro-Walkman cassette recorders rather than your industry standard Nagra because this could save them some money. I actually did have a producer on a low budget feature come to me and ask if I could record at 3 3/4 ips on my Nagra because he had calculated that it would save the production approx. $526.00 in tape stock purchases. I declined the offer to work on that production.

As for the "problem" of multi-track recording causing too many headaches for the people in post, starting with the complaint that it is SO MUCH data to sort through and store, this is an issue that should take up with the producers, directors, camera people, location managers and UPM's. I certainly would much rather record ONE good track on my Deva IV (8 tracks) which would certainly silence any of the above complaints, but if the production style of shooting and the scene requires 6 wireless, 2 booms and a plant or two, I am going to use all those tracks and ultimately the production, including sound editorial, will benefit.

It reminds me of an article I read in the Editors Guild magazine, an interview with two veteran picture editors, guys who worked with Billy Wilder, Frank Capra, Hitchcock, where they said in the old days they would get about 6000 feet of dailies and there was rarely any question about what they needed to do, they would break the film down and get right into cutting the scene. Nowdays, they said they can get up to 30,000 feet of footage from multiple camera shooting and they don't know what to do with ANY of it. The days of shooting one camera, a properly designed and composed shot, with the appropriate single track of sound, followed by another intelligent and well thought out setup that will cut with the first, I am afraid are long gone for most of us. The loss of that sort of moviemaking is what everyone should be complaining about, not the fact that 24-bit multi-track files are too large.

Regards,  Jeff Wexler

Link to comment
Share on other sites

Well said Jeff!

Unless it is a Film shoot, the 744t is mearly a backup. They always want to use the camera audio as main audio. I have requested, countless times, that we treat some of these shoots(2 hvx200's) as double system. But they insist that they don't have the time to sync the audio, later. So they want to pay for 2 Zaxcoms, and use the audio on camera only. "Even if there was a problem, Sound Editorial would have to sync the multitrack recordings, on their time."

It sounds like I am correct to be in 16 bit with these shoots. Here's the senerio, my idea now is that since we are always shooting both cameras for every set up, that when I have 3-4 tracks rolling, that I send the A camera 1 & 2, and B camera 3 & 4.  I know that could be a challenge for them to resist the temptation to stop rolling in the middle of a piece. But if we agree in pre-pro that this is how we will operate, then it falls onto the camera dept not to mess it up. What I find frustrating is that they want to streamline post, but sound wants iso's. But they don't want to spend time syncing in post. I keep telling them that you can't have it both ways, unless we go this way. So this was my solution. Multitrack to Cameras! Wish me luck!

Link to comment
Share on other sites

Once again, thank you Jeff for stating it so clearly and eloquently -- I wish I could have said it like that.  It's like when you know you're right about something, but can't seem to explain to others why you're right.  You put it in a way that will be really hard for future 'oponents' of 24bit recording (and multitracking) to ignore.  Thank you! (you won't mind if I quote you, right?)  : )

Link to comment
Share on other sites

The comment about knowing the workflow of your job is very important.  If, like most smaller video-based shoots, the audio that will be used in the final show is going to pass thru the FCP or Avid (and be OMFed out to audio from there or finished in the picture system) and the audio from your recorder is there to backup and supplement the camera sound, then using 16 bit is a good way to go.  I haven't been happy with how these systems truncate 24 bit files into 16 bit, and if the editor is actually paying attention to bit-width of the audio files doing this may result in some confusion and unnecessary phone calls.  If the show is of a scale where they will take the trouble to REPLACE the camera audio (16/48) with your recorder audio (24/48) then the increase in audio fidelity is dramatic, mostly due to how much better a recorder your SD box is than ANY video camera at any bit-width.  (I do this heartbreaker on every video job:  while rolling, switch back and forth between the return from the video camera and the return from your audio recorder.....)  The issue is really one of time and committment in post production--nothing to do with production sound.  Conforming original audio to an EDL is complex time consuming work, and if there are any problems at all then there is a great temptation to chuck all the double system sound and go with the camera audio.  (This has happened on several indie features mixed by my friends and I, despite early good intentions.)  The latest version of ProTools (7.3.1) has some interesting new abilities to do this kind of work, but everyone has to be in concert about how metadata is done, there are some issues involved if the dailies-mix is being recorded on a separate machine from the isos (sharing only TC) and some other wrinkles that are being hashed out on the DUC as we speak. 

Philip Perkins

Link to comment
Share on other sites

Yeah Phillip -- that's exactly what happens:  they use the camera sound instead of the (SD702T) files I provide at the end of each day.  And you're right about monitoring the returns from the cameras too -- good lord!  I really don't know what the big deal is about these camera systems (Vipers & 900's especially -- on board cooling fans that have no interrupt circuit, tape head cleaning mechanisms that cycle on and off at random intervals...) all that and the sound still sucks on 'em.  You'd think if a camera cost $90k it would have some decent sound.

I make it a point to try and establish an early dialog with Post, but it seems like more and more it's about what I need to do to make their lives easier, and less about getting the best sound possible.  I know the whole industry is in a state of flux -- and once everything comes out in the wash, there will probably be some pretty remarkeable solutions out there from entities like Digidesign, Sound Devices, Gallery, etc. but until then I think we have a bumpy road ahead.  I can't say how invaluable it is to me to have a forum to which I can turn -- where many of the greatest professionals our industry has ever seen are readily available and eager to offer their support.  Thanks again guys.

Link to comment
Share on other sites

I have a question regarding film-to-tape transfers.  All standard digital video tapes (DVCPRO, Digibeta) are only capable of 16-bit and I have been worried in the past that if I record at 24-bits, there will be problems with sample rate conversions if the audio is transferred digitally.  Is that true, or is the audio generally transferred in the analog realm?

I know that it would be best to record at 24-bit audio and make them conform the audio before the dialogue, but I worry about editorial having problems in the immediate, while I'm on the production.  I also work on low-budget productions and have generally been told that they will not re-conform the audio unless there are substantial "problems."  Nevertheless, I almost always record audio at 24-bits, but I didn't on the last job because they couldn't tell me whether or not the film-to-tape audio sync transfer would be digital or analog.

Tim

Link to comment
Share on other sites

Guest Eric Lamontagne

I use the fact that my audio is always BETTER than onboard audio through multitrack or bit rates or the ability to directly monitor the recorder (cause it's in front of me) to sell production on syncing up my audio instead of cheaping out and just using the onboard audio. Better quality is a great arguement for the small expense. Also, the same argument for quality applies to dual system audio and thus maintaining our rentals!

The post supervisor should respond to logic of syncing up dual system and the producer should respond to the romance of better audio makes a better picture.

Eric Lamontagne

PS. Slightly off topic: Currently working my first ever HD project where no audio is being sent to the two cameras. Telecine and post audio both only recieve DVD-RAM multitrack polyphonic files containing iso's and single mono mix. No problems except occational lack of preroll on the HD camera. This is a breakthrough for Vancouver shows. Mix backup for my purposes only (ie. not handed in) on FR-2-TC.

Link to comment
Share on other sites

Guest Eric Lamontagne

Some of these prosumer HDV cameras use mpeg4 compression ONLY when in HDV mode! I certianly would not be recording my audio in mpeg4 format just to match!

.06cents

Eric Lamontagne

Link to comment
Share on other sites

Eric Lamontagne wrote:

"PS. Slightly off topic: Currently working my first ever HD project where no audio is being sent to the two cameras. Telecine and post audio both only recieve DVD-RAM multitrack polyphonic files containing iso's and single mono mix. No problems except occational lack of preroll on the HD camera. This is a breakthrough for Vancouver shows."

Eric,

Please explain further.  What are you referring to as the "breakthrough"?  NL recording without sending audio to the cameras has been going on for at least a few years now.  Have I missed something?

-Darren

Link to comment
Share on other sites

I use the fact that my audio is always BETTER than onboard audio through multitrack or bit rates or the ability to directly monitor the recorder (cause it's in front of me) to sell production on syncing up my audio instead of cheaping out and just using the onboard audio. Better quality is a great arguement for the small expense. Also, the same argument for quality applies to dual system audio and thus maintaining our rentals!

The post supervisor should respond to logic of syncing up dual system and the producer should respond to the romance of better audio makes a better picture.

Eric Lamontagne

PS. Slightly off topic: Currently working my first ever HD project where no audio is being sent to the two cameras. Telecine and post audio both only recieve DVD-RAM multitrack polyphonic files containing iso's and single mono mix. No problems except occational lack of preroll on the HD camera. This is a breakthrough for Vancouver shows. Mix backup for my purposes only (ie. not handed in) on FR-2-TC.

That IS the best arguement, but unfortunately it is not a small expense to post sync audio, and the larger the project the more this is true.  (Remember, they'd be going from NO post-sync expense--audio on camera tapes--to whatever the cost of the method in use is.)  There are considerable metadata issues involved, and the whole work flow must be tested before the job starts.  Producers who try to do a lot of post syncing w/o having the bugs worked out of the whole system will often be the ones who give up on this and vow not to do it again.  I've had some success w/ getting even low-end video jobs to post sync my audio, but I did A LOT of handholding before during and after the shoot.  Even with these successes, those producers still use the camera audio as their default mode.  Of course, probably the best way to get post syncing to happen is to enlist the camera dept in the idea of either less or no wires going to the camera....

Philip Perkins

Link to comment
Share on other sites

Eric Lamontagne wrote:

"PS. Slightly off topic: Currently working my first ever HD project where no audio is being sent to the two cameras. Telecine and post audio both only recieve DVD-RAM multitrack polyphonic files containing iso's and single mono mix. No problems except occational lack of preroll on the HD camera. This is a breakthrough for Vancouver shows."

Eric,

Please explain further.  What are you referring to as the "breakthrough"?  NL recording without sending audio to the cameras has been going on for at least a few years now.  Have I missed something?

-Darren

Perhaps he just means that shows shot in Vancouver haven't gone this way before.  It would be true here in SF too.

Philip Perkins

Link to comment
Share on other sites

Do you guys think double system will still be around in the next 10 years or so ?

I think traditional double system recording, where images reside in one place, audio in  another, will go away... probably sooner rather than later. I believe that there will be one centralized container for the data that is the images and the audio --- this will take the form of a mass storage unit that is possibly not even located close on the set (or, hopefully off the set somewhere, similar to where we like to have lighting ballasts put). The device that actually does the initial imaging, the "camera" will be very small, light weight and quiet, and will have nothing to do with the recording of sound. The devices we use to record audio will also be somewhat different but will in many ways remain the same --- microphones, mixers, wireless, etc. So, we will continue to have double "system" in terms of our jobs, our work and an understanding of workflow, and in a way it will be liberating that camera (imaging) crews will again be able to focus on images and the sound (audio) people will be able to concentrate on recording.

Regards,  Jeff Wexler

Link to comment
Share on other sites

Do you guys think double system will still be around in the next 10 years or so ?

This is a very timely topic for me. I am currently doing some pre prod testing for an upcoming series. We will be using two Vipers. The decks are going to be separated from the heads and two  three hundred foot fiber optic snakes will feed them. The producers do not want to use traditional double system. They instead want me to feed multi channels to the record decks. There is four tracks of audio. I can use three for audio and one for TC. I am told the only time they will go to the Deva tracks is if there is a problem from the video decks. Which means that the picture edit and the dialouge evaluation will be done from the video sound. I understand that they do not want to pay for post syncing but I do not understand why the picture editor could not just load my disk right into the avid with the dailies.

So to answer the question do I think double system will be around in 10 years?

I agree totally with what Jeff has to say. Maybe in the feature world it might be but in the TV world, no way. If in fact the FCC and the CRTC are mandating high definition the days of 16mm or 35mm TV shows are gone.

Anybody want to buy a Deva V

Link to comment
Share on other sites

The decks are going to be separated from the heads and two  three hundred foot fiber optic snakes will feed them. 

What if they're doing a steadicam shot or a cablecam, or some crazy complicated move? That formula goes out the window for them at that point and you will need double system. That's why there will always be a need for double system. I suppose you could velcro 4 wireless recievers, and a transmitter for return to the camera, but the steadicam op is gonna hate you.

Tom

Link to comment
Share on other sites

I just finished shooting a "hybrid" video job in which we had a commercial-style "video village' full of watchers and listeners (managed by a professional video assist guy w/ the usual gear used for this), a DIT with his scope+critical monitor+paintbox, and me w/ a one-man "heavybag" rig.  It was a tough chaotic job with seriously bad locations and weather issues, so the deal made between the techies was this: multisnake HDSI from cam to DIT, normal Betasnake between me and camera (I'm on the set w/ the cam since I was one-man band), everything else wireless (Comtek to video assist, wireless TC to me for the P2).  This kept us from the usual cable wrangling nightmare typical of HD shoots.  I did record double system (and think I will continue to)--it was very convenient to be able to go off w/ agency and script to record the zillions of wild lines they wanted, and I needed to be able to play them the audio from sync takes they wanted to match.  The camera audio backup thing is not to be taken lightly either--I have had whole scenes saved where there was an issue w/ the camera audio not audible during recording (and who has time to check thru whole-scene length playbacks of the master these days) or there were connection issues no one caught.  Having the right time stamp on the files makes them just that much easier to use (not to mention good notes and/or filenames w/ Sc/tk info in them).  I totally understand why producers would want to use the multiple tracks of a VTR as a multitrack--there is a huge streamlining of post to gain, even if the sound isn't as good as even a 16 bit DAT.  But doing this will further slow down production, as any move will involve some serious snake-handling.  Also--many "real" locations will preclude the use of a big Lucasfilm style DIT Tent full of air-conditioner sized VTRs that run on AC--there still will be shots and scenes that have to be done w/ lighter, smaller gear.

Philip Perkins

Link to comment
Share on other sites

We used the Viper cams on Home of the Brave and I can't see what the big deal is.  They're really noisy -- and it's an oscillating sound, so forget about trying to notch it out or clean it up in any way.

Great for action shots and such -- but getting clean sound for an intimate dialog scene -- forget it.

At least with the Genesis cams you could shut off the fans while it was rolling.  It makes me wonder -- how much consulting do these people do with sound professionals during their design process?  They need to do more.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...