Why is sound still like this?

[Marc Wielage]

i'm all in for a multichannel transmitter and receiver made by red , that sends from an external device to the camera where it get's recorded multitrack on the camera aswell. and additional a wireless listening feed that enables crew to monitor

If we're recording 10 tracks, including multiple actors wearing wires, how do we monitor that?

How do you mix the show, constantly changing levels, bringing one mic up and other mics down, often on specific word cues or moments in the scene?

How do you identify each track with the name of the character? How big do you think a 10-track receiver will be? I've dealt with many cinematographers who balk at putting a 9-ounce mono receiver on the camera, and it runs on its own small battery pack for 12 hours. How long will a 10-track wireless receiver run?

And let's not even get into the problem of RF interference. You may not be aware that sometimes, we run into situations where wireless frequencies work for part of the day, then we suddenly have to change frequencies because of local interference. A dedicated sound mixer can do this in seconds, plus he or she can glance at the wireless receivers and instantly see the problem and come up with way to solve it.

Wireless transmitters are very involved and problematic, and they don't work automatically. There's a lot of troubleshooting that goes on -- none of them are "turn them on and just work" devices, because there's so much electrical interference in the real world, particularly in exteriors.

My point is that sound does not have to be difficult, provided you hire good people. At the least, if you're doing a no-budget project, ask for advice. Me personally, I honestly think the cinematographer and camera crew have the most physically demanding job on most sets, and I would rather they not have to work with watching audio levels, naming files, worrying about RF, checking batteries, getting the mics in the right places, and monitoring audio signals. I know if I'm responsible for that, it's going to be done right, and it's one less thing for the DP to have to be concerned about.

 

[Christoffer Glans]

Like Marc, I regularly work with both audio and picture, though mostly in field capture of both. In addition, I have a couple decades of experience producing projects that have entertained and returned results to their investors.
The OP misses two main points:
1) protocols are developed for valid reasons to acheive a desired result in the delivered project
2) planning is essential to all budget projects, and is better than free, it conserves the resources of all projects.

Now on to my latest rant.
Good production is not random, spontaneous, or magic.
It is not the result of accidental inspiration from amateurs.
Neither is good architecture, good writing, good theater, good music, good manufacturing, or good poetry, amongst many others.
All human creative endeavor benefits from practice, attention, and discipline.
When you couple these with vision and inspiration, you get great results.
When all you have is vision and inspiration, without hard work, you get one person's fantasizing and wishful thinking.

This forum is an outgrowth of the production of some very useful tools for image capture that can deliver amazing results. Due to the price point, there are a lot more high end image capture tools available to amateur film makers. Many of those amateurs will never seek training in their craft, and at best will read the operating instructions for this tool.

In the last 25 years, hammers and portable power tools have also gotten a lot cheaper and a lot better. At the same time, broad based training in the use of those types of tools has diminished.
Thus a much higher percentage of the population has those tools. However, the average owner of those tools is a terrible carpenter or handyman, as they have no protocol training in how why and when to use which tool.

Pinpointing amateurism among other speakers on this forum without knowing their knowledge is very condescending and with all due respect you entirely misunderstand what I was trying to aim for. It's also really strange that you call accidental inspiration an amateuristic feature. Have you ever been in real situations with a camera, trying to capture a moment that is very hard or even impossible to capture in a confined and staged set/location?
This kind of reasoning is the perfect example of a narrow minded view based on an experience with a technology that people are used to, has been around for a long time and because of it, an unseen enemy of a better fluid workflow.

I was aiming for a NEW technology that enables you to get high quality sound merged with the video files, not just for reference. Having the sound technician recording to their own device is still valid, but enabling sound and video to be synced directly on set while shooting, at a quality and fidelity enough for final sound design is a huge time saver.
Not all films have large budgets and it does not have to be an amateur film to have a low budget. Independent films struggle all the time to keep the costs down. Having a large assisstant edititors-crew working in post will cost money, adding the time for jam sync and waiting on the clapper is also time consuming, which could make a difference when the light is setting, when actors miss their mark due to disturbance etc.

For someone who speaks that a workflow that differs from perfectly financed productions is amateuristic and what "real productions is and is not", you seem to lack a lot of real experience on directing, acting or post production workflow. Because you only have one framework to go to: "productions should be like this", well, reality is a different beast, with all the errors that can happen, all the financial problems and the struggle that independent filmmakers go through.

I call for some innovation to reduce totally unnecessary areas in production/post production-workflows that is draining both time and money, but you seem to miss that mark.

If we're recording 10 tracks, including multiple actors wearing wires, how do we monitor that?

How do you mix the show, constantly changing levels, bringing one mic up and other mics down, often on specific word cues or moments in the scene?

What about HD sound as someone was pointing out (or what was it called)? It enable you to record at a better volume range.

How do you identify each track with the name of the character? How big do you think a 10-track receiver will be? I've dealt with many cinemaatographers who balk at putting a 9-ounce mono receiver on the camera, and it runs on its own small battery pack for 12 hours. How long will a 10-track wireless receiver run?

Does sound have to have 10 connectors on the camera? If you invent a new camera with a different input for sound, a new reciever that connects digitally instead of analogue?

And let's not even get into the problem of RF interference. You may not be aware that sometimes, we run into situations where wireless frequencies work for part of the day, then we suddenly have to change frequencies because of local interference. A dedicated sound mixer can do this in seconds, plus he or she can glance at the wireless receivers and instantly see the problem and come up with way to solve it.

Wireless transmitters are very involved and problematic, and they don't work automatically. There's a lot of troubleshooting that goes on -- none of them are "turn them on and just work" devices, because there's so much electrical interference in the real world, particularly in exteriors.

Why do we have to send sound through radio transmitters, isn't there any better, more reliable and digital way of sending sound as a wireless transfer? Forget about frequencies and radio transmitters, it's the tech right now, not what can be done.

My point is that sound does not have to be difficult, provided you hire good people. At the last, if you're doing a no-budget project, ask for advice. Me personally, I honestly think the cinematographer and camera crew have the most physically demanding job on most sets, and I would rather they not have to work with watching audio levels, naming files, worrying about RF, checking batteries, getting the mics in the right places, and monitoring audio signals. I know if I'm responsible for that, it's going to be done right, and it's one less thing for the DP to have to be concerned about.

I worked as a camera crew and I know how demanding it is. I also know how demanding it is to keep the current technology on track with filming schedule and it does not always work, it's broken as it is. If a system is invented that works flawlessly between the camera and sound equipment, then those worrying things you pinpoint would not exist. Instead of having one technology for sound and one technology for the camera, merge them, bring something new that works back to back.
There's a number of new technologies for transferring files or streams, why doesn't anyone think about new ways to work instead of being stuck in what is around right now?

Stop being so stuck with what you have, think about what you want, what you need and what might be a solution to it. Can't stand this whining about "how it won't work", make it work, make it real, invent... well, I hope someone is reading and think, well making that might make me rich :smiley:

 

[Marc Wielage]

What about HD sound as someone was pointing out (or what was it called)? It enable you to record at a better volume range.

We use digital transmitters now which have a 120dB dynamic range. How much range do you need? 120dB will blow your ears out completely and make you permanently deaf in 10 seconds or less.

Does sound have to have 10 connectors on the camera? If you invent a new camera with a different input for sound, a new reciever that connects digitally instead of analogue?

Digital or analog, it's big, sucks power, and has weight. My point was: the camera crew typically bitches when I want to place a 9 ounce stereo wireless audio receiver on the camera. A 10-track version would be five times bigger and heavier, and I don't want to think about the power consumption. And again: a human being still has to monitor it and compensate for level changes and RF problems.

Why do we have to send sound through radio transmitters, isn't there any better, more reliable and digital way of sending sound as a wireless transfer? Forget about frequencies and radio transmitters, it's the tech right now, not what can be done.

How well are they doing with wireless video monitoring? I see that screw up on lots of shoots every day. Audio is rarely a problem unless they get physically too far away (like over 300 feet). Come up with a perfect video village monitoring system that's $995, has zero latency, no dropouts, and works 1000 yards away, and you'll make a hundred million dollars.

I worked as a camera crew and I know how demanding it is. I also know how demanding it is to keep the current technology on track with filming schedule and it does not always work, it's broken as it is. If a system is invented that works flawlessly between the camera and sound equipment, then those worrying things you pinpoint would not exist. Instead of having one technology for sound and one technology for the camera, merge them, bring something new that works back to back.

I think your question avoids the observation that sound and picture are entirely different disciplines that require different philosophies, all towards the same goal. One might ask, "why can't we have the actors write their own dialogue and direct their own scene?" They could... but wouldn't it be better to have specialists good at one thing?

Again, you avoid my observation that sound is a constantly changing situation on set, particularly when you have multiple actors entering and leaving the scene, people raising and lowering their voices, issues to solve with mic placement and lighting, the trade-offs between wireless lavs and boom mics (and the need to sometimes use both). This is a vastly complicated area that you seem to believe can be solved with a few simple discoveries or technological changes, and it's not that easy.

I'm reminded of a conversation I had some years back with a very bright guy who had graduated from an LA film school and had a few years production experience. He asked me: "hey, what if we just shoot all our scenes in 4K from one angle, and then we just blow up all the shots to create our 2-shots, medium shots, and close-ups?" I shook my head and said, "no, you have to actually get it right on the set, move the camera in, and get all those shots by hand." He said, "why does it have to be this way? My way would save so much time!" I pointed out that if you blew up the 4K shot to get alternate takes, you'd be magnifying the grain, every cut would have different sharpness, and the perspective, depth of field, and angle of view wouldn't necessarily work, since you can't physically slide the image in 3D space.

He was looking for an easy shortcut to overcome what has always been very standard traditions in filmmaking: wide shot, move the camera, medium shot, move the camera, 2-shot, move the camera, close-ups. But there are no easy shortcuts. If there were, people much smarter than you and me would've done it ten years ago.

There are breakthroughs that are happening in sound -- Dante is a good example, where they can put up to 64 digital audio channels on one inexpensive CAT-5 ethernet cable -- and more and more transmitters and receivers are going all digital. Recorders are getting more and more tracks, and while 10 years ago you could get by with just 2 tracks on the set, now, I think 8 channels has become pretty much standard for scripted dramas and comedies. But there's an art to what goes on that technology can only supplant so far. There are no automatic buttons for that, just as there's no automatic button for editing or telling you where to put the lights, or what lens to use.

And the other problem with using lots and lots of sound channels on set -- whether you record them in camera or not -- is that it still requires lots of organization and skill in sound editing and mixing later on. That process is still complex and requires a lot of skill. I have great respect for today's sound editors, who often have to take very marginal dialogue, clean it up, and find a way to make it work under difficult conditions of time and money. That's another big technological change: new plug-ins like iZotope 3 allow salvaging less-than-optimal sound and clean it up without the need for ADR. Unfortunately, this creates in some producers' minds the impression that you can shoot in bad locations without regard to background noise, which is still not true.

BTW, speaking of sound syncing, I was reminded of this editorial written a couple of years ago by Michael Cioni of Light Iron Digital where he casually mentions that he has a method for syncing up sound in less than 8 seconds per take:

http://michaelcioni.tumblr.com/post/38993994656/the-dit-dilemma

That's faster than I can do it, but if Michael says his system works, I believe him. (With me, :20 seconds is closer to normal, but I'm slow and clumsy.) But sound syncing is still not rocket science, and Michael seems to believe recording sound separately is necessary. If anybody invented a new revolutionary system to do it more efficiently using another method, I suspect Michael would be involved in some way.

 

[Christoffer Glans]

We use digital transmitters now which have a 120dB dynamic range. How much range do you need? 120dB will blow your ears out completely and make you permanently deaf in 10 seconds or less.


Digital or analog, it's big, sucks power, and has weight. My point was: the camera crew typically bitches when I want to place a 9 ounce stereo wireless audio receiver on the camera. A 10-track version would be five times bigger and heavier, and I don't want to think about the power consumption. And again: a human being still has to monitor it and compensate for level changes and RF problems.



How well are they doing with wireless video monitoring? I see that screw up on lots of shoots every day. Audio is rarely a problem unless they get physically too far away (like over 300 feet). Come up with a perfect video village monitoring system that's $995, has zero latency, no dropouts, and works 1000 yards away, and you'll make a hundred million dollars.


I think your question avoids the observation that sound and picture are entirely different disciplines that require different philosophies, all towards the same goal. One might ask, "why can't we have the actors write their own dialogue and direct their own scene?" They could... but wouldn't it be better to have specialists good at one thing?

Again, you avoid my observation that sound is a constantly changing situation on set, particularly when you have multiple actors entering and leaving the scene, people raising and lowering their voices, issues to solve with mic placement and lighting, the trade-offs between wireless lavs and boom mics (and the need to sometimes use both). This is a vastly complicated area that you seem to believe can be solved with a few simple discoveries or technological changes, and it's not that easy.

I'm reminded of a conversation I had some years back with a very bright guy who had graduated from an LA film school and had a few years production experience. He asked me: "hey, what if we just shoot all our scenes in 4K from one angle, and then we just blow up all the shots to create our 2-shots, medium shots, and close-ups?" I shook my head and said, "no, you have to actually get it right on the set, move the camera in, and get all those shots by hand." He said, "why does it have to be this way? My way would save so much time?" I pointed out that if you blew up the 4K shot to get alternate takes, you'd be magnifying the grain, every cut would have different sharpness, and the perspective, depth of field, and angle of view wouldn't necessarily work, since you can't physically slide the image in 3D space.

He was looking for an easy shortcut to overcome what has always been very standard traditions in filmmaking: wide shot, move the camera, medium shot, move the camera, 2-shot, move the camera, close-ups. But there are no easy shortcuts. If there were, people much smarter than you and me would've done it ten years ago.

There are breakthroughs that are happening in sound -- Dante is a good example, where they can put up to 64 digital audio channels on one inexpensive CAT-5 ethernet cable -- and more and more transmitters and receivers are going all digital. Recorders are getting more and more tracks, and while 10 years ago you could get by with just 2 tracks on the set, now, I think 8 channels has become pretty much standard for scripted dramas and comedies. But there's an art to what goes on that technology can only supplant so far. There are no automatic buttons for that, just as there's no automatic button for editing or telling you where to put the lights, or what lens to use.

And the other problem with using lots and lots of sound channels on set -- whether you record them in camera or not -- is that it still requires lots of organization and skill in sound editing and mixing later on. That process is still complex and requires a lot of skill. I have great respect for today's sound editors, who often have to take very marginal dialogue, clean it up, and find a way to make it work under difficult conditions of time and money. That's another big technological change: new plug-ins like iZotope 3 allow salvaging less-than-optimal sound and clean it up without the need for ADR. Unfortunately, this creates in some producers' minds the impression that you can shoot in bad locations without regard to background noise, which is still not true.

BTW, speaking of sound syncing, I was reminded of this editorial written a couple of years ago by Michael Cioni of Light Iron Digital where he casually mentions that he has a method for syncing up sound in less than 8 seconds per take:

http://michaelcioni.tumblr.com/post/38993994656/the-dit-dilemma

That's faster than I can do it, but if Michael says his system works, I believe him. (With me, :20 seconds is closer to normal, but I'm slow and clumsy.) But sound syncing is still not rocket science, and Michael seems to believe recording sound separately is necessary. If anybody invented a new revolutionary system to do it more efficiently using another method, I suspect Michael would be involved in some way.

You are still missing my point, first off, I'm not talking about any technology that exists today, otherwise I wouldn't talk about it. Of course current tech is bulky and heavy, of course current tech is dropping through wireless transfer etc.
But the example you brought up about the 4K image is not even close to what I'm talking about, because what you miss is that I'm not talking about the process of working with sound. I'm not degrading the importance and workflow of the sound technician on set. What I'm talking about is how sound is stored and how merging a sound and camera system would create a much smoother workflow both on set and in post production. I'm trying to think out of the box, outside of the parameters that you seem to hold so dear. Preaching experience in your way only underlines what I see as the problem, the lack of big picture viewpoint and innovative thinking. I'm talking about merging a camera with sound equipment, all in one, not separate. To enable the final high quality sound to be stored in sync with the video in one file. I'm also not talking about all the sound, just the sound that's supposed to sync with the filmed footage. If the sound technician needs to get additional sound there's still that option.

I have worked with so many projects where the current technology creates a huge gap between video and sound in a way that costs a lot of money and time in post production. I have also seen and been in the situation myself of directing and not being able to see the previous take on set together with the sound, having a camera and sound crew so far detached from each other that it's ridiculous and a wonder that collaboration is even possible. Not because they are unprofessional, but because a lot of the technology I've come across is faulty, breaks down or is not enough to process what's going on on set.

My main point is, current technology is in the way of creativity, spontaneity and time. Just because something has been done in a certain way for a certain time doesn't make it written in stone and it doesn't make it the "truth". Suggesting that you can zoom in a 4K image to get many shots in one is not the same as asking for innovation in bridging the camera and sound equipment to improve storage, viewing, editing, sound design and most importantly time and creativity.

I'm sitting here waiting for my assistant editor to finish going through and sync days of shooting. It's a truly unnecessary process if the technology had worked, but it doesn't, it never does, because a production is sometimes a battlefield and anyone who think that a production, set or location is a "perfectly controlled" place I would say haven't been through a lot of them.

 

[Les Hillis]

I think as long as we're still using lenses and microphones to record images and sound we're always going to be stuck with the laborious processes that go with them.

Until there are other ways of capturing photons and sound waves, I expect there will only be incremental increases in the efficiency of those physical processes. And at the moment, I don't think there's any real or pressing demand from within the sound/image capturing industries/practitioners, or from the consumers of their work, for finding alternatives to the current methods and results.

Seems to me any real change would have to happen elsewhere, for other reasons, and then be adapted for use by the media, if it were to happen at all. Thinking along the lines of some kind of scientific/technological discovery. Which is actually in line with how things have developed so far. And who's to say whether such an advanced form of communication would involve capturing sounds and images anyway...

I do think though that sound capturing and processing hasn't been as refined over time as image capturing and processing has been. That could be due to the fundamental difference between the nature of light and sound, with one being easier to manipulate than the other, but it does make me wonder.

Anyway. Stuff sure does seem primitive sometimes. Is probably better to focus on using these sticks in the sand to convey some kind of eternal or timeless truth than wondering about what was, will or could be different about them, in the past, now or in the future.

Not that there's anything wrong with that, lol.

 

[Marc Wielage]

I have worked with so many projects where the current technology creates a huge gap between video and sound in a way that costs a lot of money and time in post production.

It does not have to cost a lot of time and money in post if you hire the right people and know what you're doing.

I'm sitting here waiting for my assistant editor to finish going through and sync days of shooting. It's a truly unnecessary process if the technology had worked, but it doesn't, it never does, because a production is sometimes a battlefield and anyone who think that a production, set or location is a "perfectly controlled" place I would say haven't been through a lot of them.

Why didn't they start syncing from the first day the material was shot? It doesn't take more than an hour or so to sync up a day's worth of dailies. I've worked on productions that have literally shot 5 hours of material a day, and I don't think the syncing cost us more than 2 hours, tops. It took far longer to make the decisions on color timing, then rendering all the dailies for viewing and dailies, plus archiving backups. A good post supervisor would have told you, "don't let dailies get too far behind production. Send the material back to editorial so that somebody can be checking it, backing it all up, syncing it up, and making sure all the footage is good."

You seem hung up on the whole question of sound syncing when in truth, this is a very easy assistant editor function that should be done by somebody making $10 an hour (maybe $24 an hour if they're in the union). That costs $50 a day if they're slow and take 2 hours. Even on low budget productions, $50 a day is not gonna bankrupt anybody. And again, you still need a good assistant editor just to make backups, help maintain the editing system, keep track of trims, get renders done, drop in effects and new material, and anything else necessary to help the editor work. A good assistant will not cost you money... they'll save you money.

You also seem hung up on the idea of recording the sound inside the camera. Where the sound is recorded is not the difficult, time-consuming part of location sound recording; the hard part is wiring up the actors, deciding what mic to use, getting the mic in position, deciding which track needs to be faded up at any given moment, compensating for actors who talk too softly or too loudly, reacting to precise word cues in the script, making sure the boom isn't casting a shadow or causing camera position problems, worrying about noisy locations and how to solve them, solving issues with RF interference, coordinating with the gaffer and electrician to minimize noise problems from lights and generators, dealing with keeping all the equipment powered all day long (including wireless mics on the actors), knowing how to listen, and on and on and on. The recorder is something I don't even worry about for 5% of my day; it's the everything else that's the biggest headache for the sound mixer on set. And the director counts on us to just handle the problem so it doesn't become an issue for the rest of the crew.

Having also worn my dailies hat in post for decades, I can tell you that dealing with sound there is looked upon as trivial and simple to fix most of the time. If anything, it's far easier today to sync up sound than it was in the non-timecode days of actual film and mag tracks. That was hard. It's a miracle so many hundreds of thousands of hours of films and TV shows were done using those primitive methods. But in fairness to those of us working today, very few mixers in the 1960s, 1970s, or 1980s were expected to deliver more than a simple mono analog soundtrack with a reasonable mix on it. Now, the expectations are much higher, as is the expense and complexity. But it's all doable, given enough skill and experience (and the right equipment).

I have also seen and been in the situation myself of directing and not being able to see the previous take on set together with the sound, having a camera and sound crew so far detached from each other that it's ridiculous and a wonder that collaboration is even possible. Not because they are unprofessional, but because a lot of the technology I've come across is faulty, breaks down or is not enough to process what's going on on set.

Then I'd again say your real problem is with communication and people, not with technology. If you're that far away from the sound crew, get walkies and a good AD, or just move the sound department closer to camera. If you're running into delays with post, hire a good post supervisor. That's what they do: find ways to avoid spending lots of time and money in post, and get the work done quickly and efficiently. I know of great post sups who are actually there every day on the set, collaborating with the director and warning them, "that will cost X dollars to fix here, vs. Y dollars to fix in post. You're better off fixing this now." Or the opposite. The key is hiring people who are more knowledgable than you are -- people you can trust to solve a problem without giving you any headaches or expense.

And I again say that a simple wireless feed from the sound department to the camera is not costly, and will allow you to hear every word, every whisper being uttered in front of the camera. That's regardless of where the sound department is physically located. If your sound department is good, they're not going to have equipment that's faulty or breaks down. I generally bring along a truck-load of backups, not because I believe my gear is going to break, but just on the 1 chance out of 1000 it will. This is all part of what you pay for with an experienced sound crew.

 

[Gavin Greenwalt]

Why can't you just use an FM broadcast into a flock of discreet button mic sized receivers plugged into various XLR/3.5mm plugs? You could then sync off of the Radio Track. No wires. Should be super cheap for receivers.

 

[Blair S. Paulsen]

My idea is similar to Gavin's. Have a sync beacon/master clock signal that all devices can jam to. If low power FM is simple and reliable sign me up.

I wrote up notes about a year ago for an iPad based Scripty operated metadata/timecode "master" system. The idea was to push scene/take info into the metadata fields/naming conventions of all capture devices including parity reference for sync. (Yes, I know signal dropouts on real sets happen, the system would need to be designed so that it polled the master source repeatedly and only locked up when multiple pings aligned.)

Some of the issues Christopher brings up do appear to be procedural breakdowns as Marc and others have noted. That said, I agree with him that the current state of affairs could use some modernization and that will need to start with new paradigms.

Cheers - #19

 

[Marc Wielage]

Why can't you just use an FM broadcast into a flock of discreet button mic sized receivers plugged into various XLR/3.5mm plugs? You could then sync off of the Radio Track. No wires. Should be super cheap for receivers.

FM is fraught with interference and distance problems, plus there are legal issues on some bands. The FCC continues to take away more bandwidth and sell it to phone companies and internet providers; it wasn't so long ago when they got rid of UHF Blocks 27 and 28, commonly used for wireless mics (right around UHF channels 60-80). FM band transmitters around 72MHz have been used for IFB transmitters like the old Comteks, but newer models shifted that up to 216MHz for better sound quality.

When I work on sets, I typically send a wireless feed via a Lectro SRb receiver to the camera (Red, Alexa, or otherwise), and I do what I can to make it as minimally invasive as possible for the camera op and DP. One trick is, this is one more audio meter that I can't see, and a recorder that I can't hear, so there's always a risk that somebody will bump a knob or unplug a cable and kill the feed. Don't forget that no sound can be recorded at speeds other than 23.98 in a 23.98 project, but we can keep recording on a separate sound recorder, so that sound can be usable for the final mix.

My main question to Christopher would be whether the Red he was using for his project had a timecode jam box attached to it. The moment you do that, syncing problems typically vanish and the camera timecode and sound timecode are identical all day long. Once the two are identical, you can sync up each shot in about 1 second. If the timecodes are different... that makes it a lot tougher.

 

[James McLellan]

OK so as long as no-one threatens to fire me I will elaborate on how I see things could change. As far as I know what I am going to describe does not exist.

Imagine a waveform slate. A digital slate that, when clapped sends a UNIQUE wireless waveform signature to any/all recorder and camera scratch tracks with tiny receivers attached. That is, it is capable of generating upwards of a million unique waveform signatures with a unique header identifying the exact slate. Software simply searches out the signatures and syncs them up. Using this you could probably have nle's set up so that when you drag a clip into the timeline it automatically pulls the matching audio clip in sync with it if it is in the project. No timecode.

This is a brief explanation. Perhaps it is a terrible idea. It's how I see things evolving.

 

[Troy Bakewell]

I think that James is on to something here.

 

[Christoffer Glans]

It does not have to cost a lot of time and money in post if you hire the right people and know what you're doing.
Why didn't they start syncing from the first day the material was shot?

Because we filmed the entire feature at a location far from any post production house or equipment needed to transcode r3d files fast. There's that and that the production budget wasn't able to cover the cost of hiring someone.

You seem hung up on the whole question of sound syncing when in truth, this is a very easy assistant editor function that should be done by somebody making $10 an hour (maybe $24 an hour if they're in the union). That costs $50 a day if they're slow and take 2 hours. Even on low budget productions, $50 a day is not gonna bankrupt anybody. And again, you still need a good assistant editor just to make backups, help maintain the editing system, keep track of trims, get renders done, drop in effects and new material, and anything else necessary to help the editor work. A good assistant will not cost you money... they'll save you money.

What money? your money? The entire budget went into production costs, to get the right people on location, the logistics because almost everyone wasn't from the place where we shot the film etc. We had a DIT on location who managed all the material and backups.
The shooting hours was 07:00 A.M to 10:00 PM almost every day for five weeks. The DIT would later be the assistant editor but there aren't any more hours left during the days of shooting.
The bottom line is that you have this "picture" in your head about how a film is done and more importantly, you seem to think that the film industry and film making process is the same all over the world. This is an independent low budget film with high ambition where we max out what is needed compared to what is not.

During this process it's easy to see how the whole process could be made easier, but you advocate against innovation with your reasoning. I do not know where you are coming from with this kind of arguing? And I don't know why you keep dictating how "we should have done it", when you know nothing about the circumstances of the shoot.

You also seem hung up on the idea of recording the sound inside the camera. Where the sound is recorded is not the difficult, time-consuming part of location sound recording; the hard part is wiring up the actors, deciding what mic to use, getting the mic in position, deciding which track needs to be faded up at any given moment, compensating for actors who talk too softly or too loudly, reacting to precise word cues in the script, making sure the boom isn't casting a shadow or causing camera position problems, worrying about noisy locations and how to solve them, solving issues with RF interference, coordinating with the gaffer and electrician to minimize noise problems from lights and generators, dealing with keeping all the equipment powered all day long (including wireless mics on the actors), knowing how to listen, and on and on and on. The recorder is something I don't even worry about for 5% of my day; it's the everything else that's the biggest headache for the sound mixer on set. And the director counts on us to just handle the problem so it doesn't become an issue for the rest of the crew.

What are you aiming for here? Have I in any way pointed out that the process for a sound technician is not all of that? Really, seriously, listen, I'm talking about the technology, nothing else, I'm talking about improving technology to reduce the amount of unnecessary work on set and in post production. Jam syncing and clapping is unnecessary, really, it is. The sound technician usually names their files according to scene, slate and take. There's nothing more then technology that prevents that data to be stored as metadata in the video file. And jam syncing is also unneccessary because it isn't needed if storing sound merged with the video is done in camera, only technology prevents it. Stop stupifying this whole thing by turning it into a black and white analysis of what a sound technician does and does not, I'm talking about improving the bridge between sound and video through technology, what the sound technician is doing on set has nothing to do with this, or actually it should reduce his headaches. Read again.

Having also worn my dailies hat in post for decades, I can tell you that dealing with sound there is looked upon as trivial and simple to fix most of the time. If anything, it's far easier today to sync up sound than it was in the non-timecode days of actual film and mag tracks. That was hard. It's a miracle so many hundreds of thousands of hours of films and TV shows were done using those primitive methods. But in fairness to those of us working today, very few mixers in the 1960s, 1970s, or 1980s were expected to deliver more than a simple mono analog soundtrack with a reasonable mix on it. Now, the expectations are much higher, as is the expense and complexity. But it's all doable, given enough skill and experience (and the right equipment).

This has nothing to do with what I'm talking about.

Then I'd again say your real problem is with communication and people, not with technology

Again, you completely misunderstand what I'm talking about and now you are lecturing about what my problem is? Well, your problem is that you do not understand a word of what I'm saying and you do not know a thing about the production I'm involved in, so get off your high horse. How you do things is completely irrelevant to what I tried to bring up. This is all about improving the technology and workflow. And the one who is really bad with communication is yourself but implying such a thing when it has nothing to do with what this thread is about, so please get out of this thread, you have contributed enough, or rather nothing at all.


NOW TO THOSE WHO ACTUALLY STAY ON TOPIC

 

[Christoffer Glans]

Blair S. Paulsen
My idea is similar to Gavin's. Have a sync beacon/master clock signal that all devices can jam to. If low power FM is simple and reliable sign me up.

I wrote up notes about a year ago for an iPad based Scripty operated metadata/timecode "master" system. The idea was to push scene/take info into the metadata fields/naming conventions of all capture devices including parity reference for sync. (Yes, I know signal dropouts on real sets happen, the system would need to be designed so that it polled the master source repeatedly and only locked up when multiple pings aligned.)

Some of the issues Christopher brings up do appear to be procedural breakdowns as Marc and others have noted. That said, I agree with him that the current state of affairs could use some modernization and that will need to start with new paradigms.

That's something like what I'm after, wireless transfer of metadata. What I also meant about storing sound in camera is to have the feed sent back to the sound technician so that the sound monitored is through the cameras interface, in that way they would know if there was any problem, if it sounds bad etc. But transferring all metadata through all system might just fix a lot by itself. If you program an auto-arrange software which analyze the metadata of all files involved with a shoot and sync them up, it's not just sound that could be synced, the screenplay could have marked pages and lines that could link to footage, much like the phrase find feature in Avid, but both earlier and not just tied to the editorial process. Would save a lot of page turning

My main question to Christopher would be whether the Red he was using for his project had a timecode jam box attached to it. The moment you do that, syncing problems typically vanish and the camera timecode and sound timecode are identical all day long. Once the two are identical, you can sync up each shot in about 1 second. If the timecodes are different... that makes it a lot tougher.

We had a jambox attached, but the big problem was the Red One which drifted all the time. We used 2-3 cameras during the entire shoot. Even though they were in sync, the process of organizing the shots in post and syncing them is a bit of a headache due to how we shot. We moved between the cameras more fluidly, taking the red one as the main camera, switching that to the smaller scarlet and epic when doing car scenes, switching back, having the main camera for second unit and third unit and so on. So you couldn't just take everything and auto sync it, they need to be categorized before syncing and organized and checked meticulously all the time, it's really time consuming. This shoot was a battlefield against time, that's why I thought about all these processes and how they could improve these type of productions. If you have unlimited production funds then it's not a problem. As for jam sync, if there is the slightest risk of drifting, it will ruin the easiness of syncing in Avid, it might sometime become more hard to pinpoint where errors lie when it works for half of the shots then if it hadn't worked at all.

magine a waveform slate. A digital slate that, when clapped sends a UNIQUE wireless waveform signature to any/all recorder and camera scratch tracks with tiny receivers attached. That is, it is capable of generating upwards of a million unique waveform signatures with a unique header identifying the exact slate. Software simply searches out the signatures and syncs them up. Using this you could probably have nle's set up so that when you drag a clip into the timeline it automatically pulls the matching audio clip in sync with it if it is in the project. No timecode.

This is a brief explanation. Perhaps it is a terrible idea. It's how I see things evolving.

Interesting

If low power FM is simple and reliable sign me up.

I'm thinking about something else. Look at bluetooth and how it's become more and more reliable over the years, protocols and range is getting better etc. There are transferring technologies that are around the corner which might erase the risk of quality loss through wireless transfer. I mean, there's an easy technology for wireless power but no one is really funding technological applications for it.

 

[Steve Sherrick]

Christopher, I get what you are after but I also have to call foul a little bit because there are solutions already in place that make this process almost instant. The problem is money. No one wants to pay for the systems that can automate this stuff.

nextLAB, Cortex Dailies, Lightiron Lilypad, etc all have the ability to sync up sound and picture instantaneously and take the metadata from the BWF files and create editorial files with that metadata in place so your scene take info is all there ready to go. They also have the ability to make sync adjustments if you need to.

Things can always get better. No offense to RED and to Dragon but do we believe this is the last sensor we will ever see? Of course not. But feeling entitled to something better than what exists is a lot different than actively pursuing solutions. I have seen many people in this industry identify a problem and then implement solutions. Grips and electric folks are notorious for this. If something can be made better they create tools to make it better. I'm sure you would rather not have to put in the work to come up with a solution on your own, because who has time for that, but sitting back and saying you need this process to be improved needs to be met with realistic expectations. If "they" did come out with something that improves this workflow how much are you willing to pay for it, or will you be willing to pay for it?

I don't question your desire to have better solutions. What I have trouble with is the recent attitude that I see permeating throughout the industry that people want all of the best tools for as little money as possible and then complain when productions don't have time, money, etc to do the job right. If I took the time to list every single improvement that I think is needed in production/post-production workflow the list would be incredibly long. But it's shorter than the list from a few years ago. Things do improve, but despite the seemingly lightning fast pace of technological innovation, the reality is this industry still moves slow in some sectors. So unless you are willing to make the innovations yourself, there is no guarantee when things will get done. Look at what Dan Kanes has done with Paralinx. He took the initiative to develop something he felt was missing in the production world and to improve the cost/performance ratio.

I know it is frustrating to hear, but Marc is correct about production values. With the right people and right tools, things tend to go fairly smooth (there are always hiccups but the idea is to minimize the frequency of them). Having no budget, no time is fine but that doesn't mean you automatically are entitled to the best tools and quality. When I work low budget, I do the best with what I have and when improvements come along that are affordable, I take advantage of them. But I can't demand others make my life easier unless I am willing to actively help make that happen or encourage those who can make it happen.

To sum up the current situation, there are tools in place that can sync audio and video with metadata available right now. Yes, they rely on timecode but they have other ways of syncing as well. With competent crew and the right equipment, jam sync is not that hard. Take that out of the mix and you make the post process harder. The last few movies I have worked on, sync has been pretty easy, and goes very fast. I will gladly welcome any solutions that make life easier. I'm not married to any one way of doing things. So to your point, I don't think the idea of improvements in this area is a bad thing but I do know that what we have now is workable when done right. But it can be expensive. By the way, using Resolve and Avid you can accomplish some relatively fast dailies workflows as well but you have to know how to do that. It's not complicated just involves the right steps.

I also respect that others work differently. I respect that we all work on projects that can't afford this sometimes. But does the problem lie in the technology or the affordability of the technology and people needed to implement it?

 

[Gavin Greenwalt]

No what's actually I think clever about my idea is forget transmitting meta data or sound over FM. That's fraught with interference. What I'm suggesting is far simpler and more resiliant... just stream random music or anything. Then load it up in something like Pluraleyes and boom. All synced.

If you want more meta data (slate data) then you can have an tablet or cell phone app that records 5 seconds of sound along with the slate data. Then again you can pluraleyes your slate meta data based on a sound clip.

If you wanted to get fancy you could transmit a stream of metadata in the sound file, but really any random dialog/music would work. It would be resilient to interference since hopefully all receivers would receive the same interference and even if they didn't, it would still be probably similar enough waveforms to sync.

 

[John DeBoer]

I think there are two points being made in this discussion, and they are very separate.

The two points: First one is James idea for an automatic sync system. There are several available in the market, all working with various degrees of success.

The other point is scratch tracks for editorial. Scratch tracks recorded to the camera leave a HUGE hole in the post process........if you use the scratch track for dailies, you don't know if the original sound recorded on the second (Main) system are actually in sync until you do the final audio conform. If the sound for dailies is synced everyday, and if there is a problem on set with TC sync boxes, the most you will have is one day of footage. Film set protocols are based upon redundancies.........is it worth the risk to the producer to risk a huge sound sync bill, or pay upfront for it to be checked on a daily basis? Most producers would not take the risk. I do know of one show that had over 200,000 in sync bills due to using a scratch track and not syncing dailies the old way.

This redundancy argument also works for dailies done on set.......if I was a producer, I would mitigate the risk by having someone assigned to QC all footage, rather than save the money.

John DeBoer
COO, SIM Digital, Bling Digital, PS Production Services.

 

[Nick Gardner]

This really all comes back to the fast, cheap, good -pick two principal. It's easy to do but it costs money.

Nick

 

[Marc Wielage]

Because we filmed the entire feature at a location far from any post production house or equipment needed to transcode r3d files fast. There's that and that the production budget wasn't able to cover the cost of hiring someone.

You couldn't ship it via FedEx? Heck, shipping a drive counter to counter is maybe $30-$40, right?

What money? your money? The entire budget went into production costs, to get the right people on location, the logistics because almost everyone wasn't from the place where we shot the film etc. We had a DIT on location who managed all the material and backups. The shooting hours was 07:00 A.M to 10:00 PM almost every day for five weeks.

Why didn't the D.I.T. sync up the sound? Why not hire a second person to help? Did you go into this not having any post-production budget at all?

As to long hours: read Haskell Wexler's excellent website 12 On 12 Off on the terrible dangers involved in pushing production people too hard. Brent's Law -- named for Brent Hershman, a low-paid camera assistant who died coming home after working several 17 hour days in a row on Pleasantville -- specifies that nobody should have to work more than 12 hours straight, and then they should have at least 12 hours of rest before the next call. I know that indie films (and studio films) sometimes go long, but then they need to push back the start of the next day to give everybody the chance to get proper rest. Otherwise, chaos can result. I've almost died myself a dozen times while driving home after 18-hour days in post and/or in production. There have been times I rolled into my driveway and was horrified to realize I had no memory of actually making the drive home. (True story.)

We have a saying in production and post: There is a great danger of Unrealistic Expectations. If you don't have a lot of money, then set limits on how much you can spend, how much time you can schedule, and what you can do with your available people and resources. Find a way to do so without taking advantage of people or expecting them to work for free.

In the case of waiting until the end of a 5-week shoot to start syncing, what if some or all the sound was bad? What if some of the media was bad? What if some of the lenses were bad? What if there were data integrity problems? What if there were flaws in the black balance? What if there were timecode issues? There are so many potential problems, I'm kind of stunned anybody would wait 5 weeks to start syncing up and looking closely at footage. But don't feel bad: I know of some disastrous 1990s film shoots where the crew waited weeks before developing the film... and only then discovered there were lens back-focus problems. Almost all their wide shots were bad. It was a very sad day when that was discovered. Luckily in that case, they did have insurance and they were able to reshoot, but it was not pleasant.

What are you aiming for here? Have I in any way pointed out that the process for a sound technician is not all of that? Really, seriously, listen, I'm talking about the technology, nothing else, I'm talking about improving technology to reduce the amount of unnecessary work on set and in post production. Jam syncing and clapping is unnecessary, really, it is.

It is necessary. One could argue that lighting is unnecessary. Just use available light, and then go in with VFX and light every actor and every part of the scene after the fact. It could be done. But to what end? Why not just do it right to begin with? Just because you can do it in post doesn't mean you should. Nothing can beat just getting it right on the set, with all the actors there, using the classic techniques that have been around for 100 years. And that goes for lighting, sound, and many other disciplines.

How you do things is completely irrelevant to what I tried to bring up. This is all about improving the technology and workflow.

Sometimes when I have discussions like this, I feel like I'm dealing with people really trying to get something out of nothing. It's not about technology at all: it's really about money. There is a necessary and realistic amount of time, money, and skill needed to pull off production and post. At some point, you gotta know when to hold 'em, and know when to fold 'em. Know what your limits are, and work within them.

Again, I've worked on some very, very low-budget projects as favors to friends of mine here in LA, and we managed to get by -- with actual timecode slates, with real syncing, with real color correction, and with humans doing all the work. Not that big a deal. Some shortcuts don't work.

These are the types of conversations post houses and post supervisors will have with producers and filmmakers prior to the start of production, warning them of the many pitfalls they'll encounter if they don't follow certain steps. We generally tell them: "do what you want, but if you can't take care of X, Y, and Z, it's at your own peril." As I said before, I'm almost as unhappy as the client when they encounter terrible delays and expenses in post, particularly when they could have been avoided early on. When you ask the right questions and do the right thing, all of this can be solved -- not with some major technological breakthrough, but just with common sense and smart people. In many cases, having those two things is ultimately more important than a huge budget.

 

[Marc Wielage]

Christopher, I get what you are after but I also have to call foul a little bit because there are solutions already in place that make this process almost instant. The problem is money. No one wants to pay for the systems that can automate this stuff.

DOH! I swear, I didn't see Steve's message before I wrote mine. Great minds think alike. Thanks, Steve!

 

[Marc Wielage]

That's something like what I'm after, wireless transfer of metadata. What I also meant about storing sound in camera is to have the feed sent back to the sound technician so that the sound monitored is through the cameras interface, in that way they would know if there was any problem, if it sounds bad etc. But transferring all metadata through all system might just fix a lot by itself. If you program an auto-arrange software which analyze the metadata of all files involved with a shoot and sync them up, it's not just sound that could be synced, the screenplay could have marked pages and lines that could link to footage, much like the phrase find feature in Avid, but both earlier and not just tied to the editorial process.

The two problems I see with this are 1) it'd be much more complicated and expensive than what we have now, and 2) as I said much earlier in this thread we'd have to get all the major camera manufacturers, sound equipment manufacturers, and related manufacturers (editing software, DIT software, script supervisor software, etc.) all aligned to one compatible standard. We don't got that yet. It's all manual. Heck, Peter Jackson just shot two $200 million movies in a row where they had to sync it all up essentially by hand, because none of the cameras had timecode -- all 48fps. Imagine the difficulty with that -- very time-consuming. And they use a custom database that exists outside the camera, outside the sound department, and outside the editing software. (I'm hoping Weta and Park Road will release these as commercial products someday.)

We had a jambox attached, but the big problem was the Red One which drifted all the time. We used 2-3 cameras during the entire shoot. Even though they were in sync, the process of organizing the shots in post and syncing them is a bit of a headache due to how we shot.

Ah! Now we get to the specifics. You needed to do a couple of things: 1) get one sync box per camera, like a Denecke SB-T, jam it to sound timecode, and make sure all the cameras are genlocked to external reference and locked to external timecode. Do that, and there will be zero drift (assuming a common sync frame rate). 2) do a workflow test to make sure everything will function correctly in post, prior to the start of production. My usual advice to production is to shoot one five-minute scene with a sync clap at the head and a sync clap at the tail. Pull the footage into an NLE with sound, and make sure everything is synced up precisely, to the frame. This can work -- I've seen it done every day.

We moved between the cameras more fluidly, taking the red one as the main camera, switching that to the smaller scarlet and epic when doing car scenes, switching back, having the main camera for second unit and third unit and so on. So you couldn't just take everything and auto sync it, they need to be categorized before syncing and organized and checked meticulously all the time, it's really time consuming.

A recent project I worked on shot on multiple Alexas, a Red Epic, a (non-timecode) BMC, and a Canon 7D. Somehow, we got all of that synced up. None of it drifted. It can be done -- it just takes a lot of work and skill.

This shoot was a battlefield against time, that's why I thought about all these processes and how they could improve these type of productions. If you have unlimited production funds then it's not a problem.

I'd propose that these were logistical problems not necessarily related to technology or money. Next time you're in LA, let's go to lunch and I'll be glad to give you free advice, no obligation.

As for jam sync, if there is the slightest risk of drifting, it will ruin the easiness of syncing in Avid, it might sometime become more hard to pinpoint where errors lie when it works for half of the shots then if it hadn't worked at all.

Once there's drift, all bets are off. That's a disaster. I know of multi-million-dollar shoots where this happened and wreaked havoc.

I'm thinking about something else. Look at bluetooth and how it's become more and more reliable over the years, protocols and range is getting better etc. There are transferring technologies that are around the corner which might erase the risk of quality loss through wireless transfer. I mean, there's an easy technology for wireless power but no one is really funding technological applications for it.

Yeah, I know some of Tesla's history, and I'm still not convinced wireless power ever really existed. It's a terrific legend, though, kinda like the water-fueled engine.

I don't think Bluetooth can work for timecode or for sound because of low power constraints, distance problems, dropped packets, and inherent delay. There are alternate systems that show promise -- Timecode Buddy is one of many systems (this one using WiFi) being tried out now -- but all cost more money than the timecode jammed boxes mentioned above. And they still don't pass metadata... yet.

The problem with ideas like this is while new workflows are being devised, data requirements are skyrocketing up and up and up every year. I can recall a time when we'd do a feature, and if we had more than 10TB of data, that was considered swamped. Our facility was groaning at the seams trying to handle that many files. Now, 10TB is trivial; there are people shooting that much in a day (like The Hobbit). And with every new camera that comes out, there's a new format, a new LUT, and new technical issues. It's a never-ending spiral of challenges. Sound is tinkertoys compare to these larger problems.

The key thing I believe in post is you have to keep it simple, stick with proven methods, document everything you do so that everybody is on the same page, employ people who understand what needs to be done, and you have to have enough people willing to work the hours needed to get the job done. No super technology, no bluetooth -- just hard work and smart people. It's doable, I've done it, and I've survived, with big projects, small, and inbetween. Not that big a deal. It helps to have very fast computers and a great pipeline, but that's a different discussion for another time.

John DeBoer's points above parallel mine: I would add that if you're running multiple tracks, it's a good idea to sync them up in dailies to aid in the OMF and sound conform on down the line. If you aren't sure, get some good advice from a post house, or talk to a good sound editor before you cut a single frame. This will save you time and money in the long run in preparing for the final mix.