Home of Professional Cinematography since 1996

style="margin-bottom: 0"> 

class="style5" Clockits

>Published : 15th May 2005

class="style8">>>each job was quite definite that he had been instructed by >>Panavision (or, through his own experience) NOT to use the HD tri->>level sync provided by the Ambient Clockit timecode/sync generators >>that velcro to each camera.

>Will someone please tell me? I am dying to know : How did the MYTH that attaching an cigarette pack sized external sync/time code generator ("Clockit") to an HD camera somehow helps things along?

>I've been hearing about these clockit boxes for about a year , and when I finally saw one last month, I nearly burst out laughing, (nearly causing a fistfight with my audio man colleague, who thought they perform some sort of magic.) Yes. It is a sync/time code generator, but, without a hard wire to lock sync and time code to all the cameras, it doesn't do anything.

>So, there is no upside to putting a free running external sync generator on an HD camera. Zippo. Zero, Nada, NoDo, Nyet, Bupkis. But there is a down side : If the sync cable goes intermittent, or the battery runs low? Yes Virginia, there will be "Green Flashes" (Loss of sync event) and other such nasties.

>I'm a video engineer who has been doing remote television for nearly 25 years. I do hundreds of shows a year. I've done shows with up to 34 cameras at a time. I've done hundreds of shows with free-roving, un-cabled cameras that require synchronous time code. (Concerts, The L'eggs Mini Marathon,2 Boston Marathons, 9 NY Marathons, 2 Atlanta Olympic Marathons, etc, etc) We've never used external TC/Sync Gens. I don't know why anyone would.

>I've been researching this endlessly. The best I could come up with is that when the film industry started adapting HD cameras, there were and still are some lip-sync issues. Somehow the myth/superstition got out there that attaching an external TC/Sync Gen would solve the lip-sync issues, when in fact the origin of the lip-sync issues lies elsewhere.

>LET THE FLAMES BEGIN!!!! I'M READY!!!

>Lew Comenetz - Video Engineer/DIT


class="style8">>>I'm a video engineer who has been doing remote television for >>nearly 25 years. I do hundreds of shows a year. I've done shows with >>up to 34 cameras at a time. I've done hundreds of shows with free->>roving, un-cabled cameras that require synchronous time code.

>So, are you saying that in 25 years you've never seen sync problems on a multi-camera shoot?

class="style8">>>when the film industry started adapting HD cameras, there were and >>still are some lip-sync issues. Somehow the myth/superstition got >>out there that attaching an external TC/Sync Gen would solve the lip->>sync issues, when in fact the origin of the lip-sync issues lies >>elsewhere.

>What is your supposition as to where the sync issue lies?

>John McDaniel
Audio Post Facility Owner
Sonic Arts Digital Audio Services, Inc.
Cincinnati, OH USA


class="style8">>>So, are you saying that in 25 years you've never seen sync problems >>on a multi-camera shoot?

>No. Though its rarer than you'd think. And when it happens its because of external generators running individual cameras. You don't run external sync unless you MUST, as in a live show.

>I'm saying in 25 years of live TV and roving Betacam coverage of concerts and such, we either hard wire sync (we have to for live) or we don't run it at all. It accomplishes almost nothing to attach a free-running sync generator (no hard wire) to a roving Betacam or HD Cam.

>Clear?

>Lew Comenetz - Video Engineer


>John McDaniel

class="style8">>>What is your supposition as to where the sync issue lies?

>I'm not sure if there even is a sync issue. I've never seen it. I'm trying to fathom how what I regard as a truly BIZARRE trade practice evolved (use of non-hard wired sync gens on free roving cameras).

>I mean, its like some pagan ritual to keep the devil away or something! Can anyone on this list tell me where this "Clockit" thing started? Was one of you there that day the guy whipped one out of his kit and said: "Here, try one of THESE! I smoke 'em 'He' smokes 'em!" [Woody Allen movie takeoff on a cigarette ad: The priest says : "He [God] smokes 'em" as he points towards heaven.]

>More seriously : I think lip sync issues might have evolved either from shooting double system sound, and/or on an external tape deck. Apparently if you don't record embedded digital audio in the HD Cam picture, there is a variable delay in the video with respect to audio due to digital processing delay within the camera with respect to externally recorded audio. This has caused some problems. I thought they fixed it but maybe not.

>Lew Comenetz - Video Engineer/Crumudgeon


>Does this issue come down to shooting double or single system sound?

>Film style productions using HD cameras still shoot double system sound and expect a reference (though the old fashion way works fine - clapper - if the editor is paying attention) for the audio recordist. Event television is nearly always recorded with single system sound.

>The issue is fascinating so a further explanation of what problems are or are not be solved here would be helpful.

>Robert Goodman
Author/Photographer
Philadelphia, PA


class="style8">>>...no upside to putting a free running external sync generator on an >>HD camera. Zippo... there is a down side: If the sync cable goes >>intermittent, or the battery runs low? ...there will be "Green Flashes" >>(Loss of sync event) and other such nasties. LET THE FLAMES BEGIN!

>No reason to flame, just add a couple of things...

>1) The "green flash" issues I experienced on a pre-/3 Sony f900 (well documented here on CML) were clearly due to INTERMITTENT sync, or sync INTERRUPTION. Signal was fine with sync, sans sync. It was during the transition that the camera hiccupped as it re-clocked itself and gave what the post engineers told me was a "concealment green flash". Note also that most of the flashes gave no error lights on the f500 decks which thought green flashes were artistic intent.

>2) Our problem had nothing to do with a Clockit/Denecke or any type of sync box, but with a hardwire sync signal into the camera where we had a bad cable/bnc connector right at the camera. Very frustrating as the cable looked mint, but a wiggle while recording showed it was the culprit.

>3) One has to wonder how much the manufacturers were able to address this issue with software/hardware updates in the most current camera models.

>4) I have found that on many shoots tri-level sync is not necessary - time code is enough. Every time I check with post and they have zero issues. Live events humming through a switcher is of course a different issue.

>Mark Doering-Powell
LA based DP


class="style8">>>Does this issue come down to shooting double or single system >>sound?

>Definitely.... I think if you read through the archives there's all kinds of quite detailed descriptions of the problems that led to the Clockit solution.

>I remember doing an experiment a couple of years back where I jammed timecode to my two Betacams from an Ambient slate, then came back ten hours later to check the sync... The two Betas were within a frame of each other, but were almost a full second off of the Ambient. This seemed to confirm reports from the field that cameras and sound recorders would "drift apart" which is the problem people were experiencing.

>The consensus seems to be that the TC generators in sound world – the Ambients etc. - are more consistent with each other and accurate over the long term, than those in the cameras. Hence the Lockit boxes.

>George Hupka
Director/DP
Downstream Pictures
Saskatoon, Canada


> Lew Comenetz wrote :

class="style8">>>magic.) Yes. It is a sync/time code generator, but, without a hard wire to >>lock sync and time code to all the cameras, it doesn't do anything.

>No flames here, just to enhance your understanding of the Clockit and Denecke sync and T/C boxes. The point of these boxes is that you can jam as many of them as you want to a single master sync-T/C source.

>The Clockit's have extremely accurate crystal oscillators in them and also have the ability to have individual units software "trimmed" in the field so their oscillator frequencies are extremely closely matched. (Someone else can add the actual numbers, I don't know them).

>The point is that if you put a Clockit box on each camera, you have approximated a wireless "genlock" between the cameras. One Clockit box will accomplish nothing, but one on each camera locked to the sound department (if you are running double system) and you have matching T/C for everyone.

>No fisticuffs are necessary, we're just making movies here...

>John Gilman
Zoundz Audio


>John Goodman writes :

class="style8">>>Does this issue come down to shooting double or single system >>sound?....[etc.]

>Good point Robert. Glad you brought it up.

>My intuition is double system sound is the underlying issue. My theory is that double system sound recordists (or someone) mis-attributed what is actually digital video processing delay to time code drift, and that is how this all started.

>After that it was like "Give 'im an Enema!!!" (You know the old joke: An actor collapses on stage. Some guy in the audience keeps screaming: "Give 'im an Enema!" "Give 'im an Enema!!!" "Give 'im an Enema!!!" The stage manager says: "How will that help?".

>[Punch Line:] "It can't hurt!"

>So it is with this: The lip sync is off so: "Give 'im a sync generator!" "Give 'im a sync generator!"

>"How will that help." "Well, it can't hurt!"

>Lew Comenetz - Video Engineer


>Mark Doering-Powell wrote :

class="style9">>>2) Our problem had nothing to do with a Clockit/Denecke or any >>type of sync box, but with a hardwire sync signal into the camera >>where we had a bad cable/bnc connector right at the camera.

>Happens all the time. I see it every day. Its the reason you don't want to externally lock any video device unless there is a compelling reason to do so. Which is why you do not want to piggyback any external time code or sync boxes on a free-roving camera.

>EVER.

>Lew Comenetz - Video Engineer


>The Kinetta camera has an extremely accurate crystal, and should hold sync after being jammed for at least 12 hours. Compatible with Ambient and Aaton stuff -- jam it and disconnect. Of course, you only need to do this for double system shooting.

>Jeff Kreines


>Mark Doering-Powell wrote:

class="style9">>>2) Our problem had nothing to do with a Clockit/Denecke or any >>type of sync box, but with a hardwire sync signal into the camera >>where we had a bad cable/bnc connector

>Lew Comenetz wrote:

class="style9">>>Which is why you do not want to piggyback any external time code >>or sync boxes on a free-roving camera. EVER.

>Living life in constant fear of bad connections is not the way to go. CRITICAL connections are made and depended upon every day (ask any location sound mixer!).

>One of the most painful experiences one can have in post is to sit around with a producer and talk endlessly about whether or not certain shots are in sync. Once the question is raised, it is never resolved to everyone's satisfaction (pun intended).

>There ARE reasons to use such devices.

>John McDaniel
Audio Post Facility Owner
Sonic Arts Digital Audio Services, Inc.
Cincinnati, OH USA


class="style8">>>...without a hard wire to lock sync and time code to all the cameras, it >>doesn't do anything.

class="style8">>>The point is that if you put a Clockit box on each camera, you have >>approximated a wireless "genlock" between the cameras.

>I recently worked as Video Op on a multi-camcorder shoot, using a mix of HDW-700 and F900 systems. I had not used the Denecke boxes (SB-T) before and the other engineer, who provided the equipment for the shoot, brought several along. He explained that he had had drift problems with Camcorders in the past and found the Denecke boxes to be more stable and, if you took the time to trim them, extremely consistent.

>Denecke claims a stability of +/- 1ppm for their generators. Ambient claims up to +/- 0.2ppm This is considerably better than the SMPTE 274M spec for sampling frequency: +/- 10 ppm. I was unable to find a spec for the F900 so I have to assume that it is at least within the SMPTE spec but may not be as stable as the external boxes can be. This can be important if you want to lock a camera with no cables for the duration of a long concert. You can come close by having a cable in an accessible location for the camera and jamming after each battery and/or tape change but it can cost you valuable time if it happens at the wrong time (it will.)

>Another issue is the relationship between 30 Hz and 24 Hz code (or 29/23) and the 4/5 frame cycle which links the numbers. At 24 fps, four frames pass in the time that it takes five 30 fps frames to occur so unless your 24 code starts at the right place in the 30 sequence, you can generate problems between off-line and on-line post in syncing tapes. While most of these items can be worked around, they can be avoided by taking care to keep TC and sync together and syncing from sources that know about the relationships. This is less of a problem with modern cameras and generators but old boxes, separate cameras and recorders, and multiple TC sources can screw you.

>The worst thing that you can do is to hook up a free-running TC
generator to a camera. In that case, the camera must either ignore the phase of the code with respect to video, or relock to the code when it is different from the camera code. That is definitely a recipe for the dreaded green flashes and dropped or added numbers in the sequence. Editors don't like glitches. Slow drift between cameras is much easier to handle. If the camera jams only when it stops, the problems area less obvious.

>And of course, whether these are problems or not depends on your approach to editing and the criticality of the sync. Some folks still use manual slates and ignore time code and get away with it. You can get away with a lot if you post your own material. Working with others is trickier.

>References :

Denecke SB-T Time Code + Video Sync Generator


http://www.denecke.com/prod03.htm#SBT


Info on some of the issues


http://www.denecke.com/24pinfo.htm

Ambient trilevel sync Lockit box


http://www.ambient.de/Trilevel_lockit_instWS.pdf

>SMPTE 274M

1920x1080 Scanning and Analog and Parallel Digital Interfaces for Multiple-Picture Rates(not available on-line)


http://www.smpte.org/smpte_store/standards/

>Cheers,

>Charles R. (C.R.) Caillouet, Jr.
Vision Unlimited/LA
HD production technical support since 1987
...searching for the right tool for the job...


>John McDaniel wrote :

class="style8">>>Living life in constant fear of bad connections is not the way to go. >>CRITICAL connections are made and depended upon every >>day...most painful experiences one can have in post is...talk endlessly >>about whether or not certain shots are in sync.

>Wish it were that easy. Just to clarify why I mostly suggest "no sync" in my own work environments.

>I normally shoot with 2 cameras, maybe 3 on a busy day. Not a live event so nothings through a switcher.

>Usually Double-system : select takes synced to sound and cloned onto a "circled takes" HDCam or D5 used for post. The Camera Original HDCam vaulted for protection. TC slate and "clap stick" backup always used. Mixer's Master TC on an audio channel via Ambient or Denecke (or onto VITC when post or an engineer/DIT insists, but then you're supposed to have Tri-level sync as well ?).

>The only time we clock all the cameras together is, for example, if we're slaved to 24fps monitors - or something like that. Besides the TC which may not have the benefit of Tri-level sync, the smart-slate & "clap" can insure that the sound is synced with frame accuracy. Perhaps the same producer debating in post whether the sound is 1 frame to-or-fro would vastly prefer this method over "why the hell do we have green flashes in dailies and nobody knows what caused them". And really, we've had no problems syncing sound as long as everyone provides 8 sec of pre-roll.

>The other thing to note : while many critical connections are maintained and depended upon each day, few of them have the traits of Tri-level sync : you often don't know its a problem until the next day playing it back. There are no error lights, and the problem is intermittent and therefore difficult to trace. Even camera playback on OB LCD monitors masks the problem (much easier to see the flashes from an f500 deck on an HD CRT). All the Sound Mixer's critical connections can be confirmed on-set, with confidence monitoring and sampling what data's coming out of their sync boxes.

>I know it borders on the paranoid, but when the software driving the gear tries to be smarter than it ought to be, and there is no reasonably efficient way of "confidence monitoring" our footage, except with rec/review (needless to say we missed all of our green flashes that were mid-take) you're forced to take action when something like this occurs.

>Mark Doering-Powell
LA based DP


class="style8">>>Which is why you do not want to piggyback any external time code or >>sync boxes on a free-roving camera. EVER.

>Straying briefly off-track, I worked a rock concert recently where the DIT put external sync boxes on all the cameras, which were SDX900s. Apparently no two SDX900 internal clocks run at the same speed. Jam syncing would have resulted in camera time codes varying by several seconds by the end of the evening.

>The three fixed cameras were synced by cable; the handheld cameras all had external sync boxes. There were no problems in post.

>I now return you to your HD chat currently in progress.

>Art Adams, DP [film|hidef|video]
San Francisco Bay Area - "Silicon Valley"
http://www.artadams.net/
Local resources : http://www.artadams.net/local


>Art Adams wrote:

class="style8">>>Apparently no two SDX900 internal clocks run at the same speed.

>Panasonic needs to fix this.

>Jeff Kreines


>Art Adams wrote :

class="style8">>> DIT put external sync boxes on all the cameras, which were SDX900s.

>My experience in this:

>Recently used SDX900's with sync boxes jamming the cameras, and immediately unplugged. Cameras held that time fairly well - I think the Mixer figured out that we'd have to re-jam every 2-3 hours.

>Issues re : green hits were on Sony f900's, 2 years ago and pre/3 software.

>Mark Doering-Powell


>Art wrote:

class="style8">>>Apparently no two SDX900 internal clocks run at the same speed. Jam >>syncing would have resulted in camera time codes varying by several >>seconds by the end of the evening.

>I've never experienced this. Any chance of operator error? Perhaps Jan or Mr. Goodman could elaborate.

>Illya Friedman
Production Account Executive
Wexler Video
Burbank, CA


class="style8">>>Apparently no two SDX900 internal clocks run at the same speed. Jam >>syncing would have resulted in camera time codes varying by several >>seconds by the end of the evening." I've never experienced this.

>No operator error, but perhaps operator misunderstanding of theory vs. reality, and over concerned-ness of things being "Right" when they don't really have to be. I've had editors tell me that as long as time code is in the same universe, they can deal with it. They even have a term for it: They say "We can slip frame the edits".

>BUT:

>I've done a whole bunch of televising of Marathons, (NY, Boston, etc.) We turn a bunch (7 to 10) of camcorders loose at 4 or 5 in the morning with free-clocking time code, don't get 'em back 'till dark, and they seem to be frame accurate 12 or 14 hours later (and heaven knows what the cameramen do to those cameras in the dark when we are not looking!!!!).

>I don't get what all the fuss is either.

>Lew Comenetz - Video Engineer


>If any one has any questions about the lockit box, I would be happy to give the details, but generally things have been said. The problems that occur are mostly broken cables and low batteries (green flash). When the battery goes below a certain voltage there is a LED double blink to warn the operator that batteries have to be changed. If they are not changed then after a time the sync and xtal drift. (still double blinking).

>Older tri-level lockits will go at least 13 hrs, new ones over 24 hrs with good alkaline cells. There is a cable to connect to the camera Hirose 12 volt output for external powering. If power is removed then the Lockit batteries take over seamlessly. generally (at Sony Studios) they always put new batteries at the beginning of the day and that does it, (they had the green flash problem when older batteries were used for more than 1 day). As regards video signals the Lockit box is very low jitter and does all Sony and Panasonic Varicam trilevels as well as composite video PAL and NTSC all correctly sync with TC. You can even do 23.98P trilevel sync to 29.97 TC which you can then put on the cameras audio track.

>As everyone says it is not just the accurate xtal that keeps the boxes in sync but the Xtals can be tuned together so that they vary very little between boxes (0.2ppm). giving less than 1 frame in 24 hrs.

>Anyway any questions welcomed.

>Cheers

>Chris Price Ambient


>Lew Comenetz - Video Engineer writes :

class="style8">>>I've done a whole bunch of televising of Marathons, (NY, Boston, etc.) >>We turn a bunch (7 to 10) of camcorders loose at 4 or 5 in the morning >>with free-clocking timecode

>I did say that I had not ever encountered significant problems that were big enough to ever be memorable. Editors have to work with all sorts of problems that we either deal with or kludge around.

>Jeffery Haas
freelance shooter and editor
Dallas, Texas


>Jan is travelling or I'm sure she'd pipe up.

>This statement that no two SDX900s internal clocks run at the same speed is ridiculous. The only relevant question is what degree of accuracy do the camera's internal clocks have.

>In my experience (long before and long after the last SDX900 hits the museum), the time code clocks in video cameras have a degree of accuracy sufficient to satisfy the needs of production. I have done multi-cam shoots with far less sophisticated gear than is presently available and we never had a problem in production or in post.

>According to the engineering team at Panasonic, the SDX900 conforms to the NTSC spec which is 10 pmm. This translates into less than one frame per hour drift. Are there devices available which are more accurate - yes - Ambient for one - or lock to a GPS signal which is even more accurate.

>The only instance when drift can be an issue is when using the SDX900 or Varicam in a mixed shoot - Cinealta and Varicam for instance because the Cinealta could using 23.976 time code rather than SMPTE time code which is what is always used in Panasonic's cameras irregardless of the capture frame rate.

>Robert Goodman
Author/Photographer
Philadelphia, PA


>Lew Comenetz wrote :

class="style8">>>over concerned-ness of things being "Right" when they don't really >>have to be. I've had editors tell me that as long as time code is in the >>same universe, they can deal with it.

>I'm sorry, but just saying, "they can deal with it in post with 'slip frame edits'" is like saying, "Why white balance? They can make all this stuff shot under these fluorescents look right in post with a little color correction."

>BTW, hell yeah they can "deal with it," since post rates are by the HOUR, not the day!

>For a spot, slipping may not be that big of a deal, but for a feature where the audio post department has to re-conform audio MANY, MANY times to accommodate continuing picture edit changes, the PITA factor adds up to $$$ spent.

class="style8">>>We turn a bunch (7 to 10) of camcorders loose at 4 or 5 in the morning >>with free-clocking time code, don't get 'em back 'till dark, and they >>seem to be frame accurate 12 or 14 hours later

>"Seem to be frame accurate" ?? Cut in the middle of some dialog and see if it "seems" frame accurate. This barely even qualifies as anecdotal evidence. By the way, the only way to check for accuracy would be to intercut cameras shooting the same thing _from_the_same_distance_.

>Don't forget that approximately 33 ft. equals one frame of audio delay.

>That's right. Line up audio & video from two cameras; one 33 ft from the subject and another 100 ft from the subject and your audio from the camera mics will be two frames apart when compared in post.

>John McDaniel
Sonic Arts
http://www.mix2pix.com


>Goodman wrote:

class="style8">>>This statement that no two SDX900s internal clocks run at the same >>speed is ridiculous. The only relevant question is what degree of >>accuracy do the camera's internal clocks have.

>I think that the REAL relevant question is :

>"To what degree is the accuracy of an internal clock as a _single_specification_ reflected in a day's worth of real world use that includes power up/down cycles, battery changes, draining batteries to the point of shutdown, and temperature swings?"

>THAT'S the question IMO.

>John McDaniel
Audio Post Facility Owner
Sonic Arts Digital Audio Services, Inc.
Cincinnati, OH USA


>In the real world, lip sync is flexible. It has been since the day sound arrived on set. Cameras in the same environment are likely to respond the same way to the conditions. If you want perfection, lock the cameras using sync. Otherwise a decent editor can handle a frame slippage here and there.

>Robert Goodman
Author/Photographer
Philadelphia, PA


class="style8">>>I'm sorry, but just saying, "they can deal with it in post with 'slip frame >>edits'" is like saying, "Why white balance? They can make all this stuff >>shot under these fluorescents look right in post with a little color >>correction."

>You miss the point : Its not an "Aw, we'll just fix it in post" thing. Its that we in the live event business don't have any choice: If we're covering a Marathon we slam the cam-corder's time codes together at 4 in the morning at the airport, the cameras go on the helicopters and motorcycles, and we don't get to touch them again until after dark, but when they cross the finish line, the TC gens are clocking away within 5 frames of each other.

>Some of these shows are turned around and edited for air in 2 hours, and nobody has any issues. How do you think they televise the "Tour De France?.

>By the way : We always shoot single system sound in TV, so the audio is on the tape, so there are no audio sync issues.

>Lew Comenetz


class="style8">>>seem to be frame accurate" ?? Cut in the middle of some dialog and >>see if it "seems" frame accurate. This barely even qualifies as >>anecdotal evidence.

>Okay. Let amend that. THEY ARE ACCURATE. How's THAT! MMKayyeee? [from "South Park"]. I've done the New York Marathon 9 times over, the Boston Marathon Twice, the two Marathons at the Atlanta Olympics. I've done this over and over again.

>So its not anecdotal. I see it every year my own little eyeballs. MMmkayyeee?? I prep & release the cameras from the airport or broadcast centre in pre-dawn darkness. MMMKaayyyeee?? They go up in the helo's and on the motos, MMMkayyyeee??? They don't come back 'till its almost dark out, mmmkayyyeee?? and I check the TC Gens mmmkayyyeee?? and they're within frames of each other: The second counter ticks over simultaneously on all of them; MMkayyeeee?? You can visually see errors within about 4-5 frames just by looking at the second counter, MMkayyeee?? or momentarily running a jam cable to see of the code "hiccups". (I do. It doesn't)

>Lew Comenetz - Video Engineer


>John McDaniel wrote :

class="style8">>>Don't forget that approximately 33 ft. equals one frame of audio delay...

>A good point and I'm always amazed during long-lens setups when a Boom Op shotgun mic's the sticks from 60 ft away having conspired with the 2nd AC to give him loud sticks.

>Post slip frames are necessary even with an atomic clock inside the cameras - or if you want things perfect, you do need genlock.

>Maybe people with a video background are shocked that when sound is not recorded directly to the video tracks - there might be some SLIGHT discrepancy at times - usually less than a frame - it still bugs them.

>Relax, its usually ok (except for the exceptions outlined in this thread). This is a much easier issue than being way off white-balance and recorded 3:1:1 and hoping you can fix it for the film-out.

>And at a certain point, the most important thing on-set is to get the shot, and record the sound. I think most here agree its lame to hold the roll because the TC slate needs new batteries or a jam for one take. If you're giving up even one piece of coverage due to rejamming something, I say bang the sticks and we'll figure it out between the next setup - the Editor will spend all of 30 seconds manually syncing that take, and I'm ok knowing that 99% of the stuff will sync fine just by TC.

>Mark Doering-Powell
LA based DP


> Lew Comenetz wrote :

class="style8">>>and they're within frames of each other:

>You obviously are not an editor, Lew.

>Film people try to sync to a standard of a quarter perf (that dates back from cutting on film). Checking by trying to eyeball several cameras at once barely qualifies as anecdotal evidence.

>Here's the test. Jam sync the cameras. Shoot all of them shooting a slate with a clap. Let the cameras sit for 12 hours, with the clocks running. Shoot the slates again. Load the tapes into an editing system, and line all the cameras up using time code as a reference. The first slates will all line up, unless something is very wrong. Let us know if the slates recorded 12 hours later line up, within 1/2 frame.

>Jeff "fighting sync problems since 1968" Kreines


>John McDaniel wrote:

class="style8">>>Why white balance? They can make all this stuff shot under these >>fluorescents look right in post with a little color correction."

>Not necessarily. Being off a few frames with timecode can be corrected in post without a great deal of effort or cost, but recording the wrong color information to tape (e.g., not enough blue channel) can be too costly to fix.

>Often, it really is less expensive to "fix it in post". Knowing when that is the case and when it is not is what keeps the money up on the screen.

>Noel Sterrett
Admit One Pictures


>Noel Sterrett wrote:

class="style8">>>Often, it really is less expensive to "fix it in post". Knowing when that is >>the case and when it is not is what keeps the money up on the screen.

>Testify, brother. I've got no problem with that.

>John McDaniel
Audio Post Facility Owner
Sonic Arts Digital Audio Services, Inc.
Cincinnati, OH USA


>Chris price wrote:

class="style8">>>Anyway any questions welcomed.

>Are you aware of any testing of current professional digital video cameras that answers the following:

class="style8">"To what degree is the accuracy of an internal clock as a _single_specification_ reflected in a day's worth of real world camera use that includes power up/down cycles, battery changes, draining batteries to the point of shutdown, and temperature swings?"

>IOW : does the real world reflect the testing lab?

>TIA,

>John McDaniel
Audio Post Facility Owner
Sonic Arts Digital Audio Services, Inc.
Cincinnati, OH USA


 

John McDaniel wrote:

class="style8">>>"To what degree is the accuracy of an internal clock as a >>_single_specification_ reflected in a day's worth of real world camera >>use that includes power up/down cycles, battery changes, draining >>batteries to the point of shutdown, and temperature swings?"

>Kinetta uses similar technology to Ambient, and our cameras oscillators can be finely tuned just like Ambient's. We're compatible with Ambient and Aaton devices.

>Jeff Kreines


>Mark Doering-Powell wrote :

class="style8">>>I'm always amazed during long-lens setups when a Boom Op shotgun >>mic's the sticks from 60 ft away having conspired with the 2nd AC to >>give him loud sticks.

>Mark, ahem, I think it's good practice. What's wrong with that?

>>I think most here agree its lame to hold the roll because the TC slate >>needs new batteries or a jam for one take. etc

>I at least don't agree. I'm still after all this years amazed that the camera department finds it lame if they have to wait like 30 seconds for sound. Totally kills me...but maybe you were just joking, right?

>Karl Lohninger
Sound mixer, editor, etc.
Los Angeles based


>Karl Lohninger wrote :

class="style8">>>Op shotgun mic's the sticks from 60 ft away having conspired with the >>2nd AC to give him loud sticks.

class="style8">>>Mark, ahem, I think it's good practice. What's wrong with that?

>Only that it will essentially be out of sync if Editor had to sync to clap (and the discussion was about sync). 1 frame every 33 ft. It’s not the valiant effort, just the fact that the clap will only get you close, and unfortunately some sound crews tend not to heed this - the ones that do put a radio mic onto the slate itself (a Sound Mixer's how I first learned about the 33ft=1 frame equation).

>I'd rather have the 2 of them conspire to do quiet sticks when they put it right in front of an actor - that's more relevant on many shots.

class="style8">>>I at least don't agree. I'm still after all this years amazed that the >>camera department finds it lame if they have to wait like 30 seconds for >>sound. Totally kills me.....but maybe you were just joking, right?

>I'm very good to Sound - help them all the time - light their way in and get a decent boom line up to the edge of frame. My saying is always "sound is half". If they had to fix a radio mic or what not its understandable. Like I said, the most important thing to do is for them to record good quality sound. As for sync...perhaps the difference is that I work a lot with short schedules and when a kid's about to turn into a pumpkin, you should roll and know that the clapper is a backup that works fine if you're not using it for every take all day. I just don't think that in a sunset or "panic setup" is the best time to replace slate batteries - do it between setups. I usually encourage "lets do it the old fashioned way on this take" and urge us to shoot.

>Mark Doering-Powell
LA based DP


>Mark Doering-Powell wrote :

class="style8">>>Only that it will essentially be out of sync if Editor had to sync to clap >>(and the discussion was about sync).

>The point is to get the clap sound at all! If the editor knows what he's doing he can and will easily adjust, alas only if the clap is audible, therefore I am one of those asking for a loud clap. And IF there is time for it, I'm happy to slap a wireless lav onto the slate.

class="style8">>>As for sync...perhaps the difference is that I work a lot with short >>schedules and when a kid's about to turn into a pumpkin, you should >>roll..

>Mark, I totally agree with this scenario, I had trouble with your earlier more general statement.

>Karl Lohninger
Sound mixer etc,
Los Angeles etc.
Currently Las Vegas


>Karl Lohninger wrote :

class="style8">>>The point is to get the clap sound at all! ...IF there is time for it, I'm >>happy to slap a wireless lav onto the slate.

>Yup. I agree - better than nothing. Thing is some sound crews don't seem to think of or care of wiring the slate - but whatever - most important thing is that they record good sound. The Editors, or their Assistants actually make it all happen if sync isn't perfect for a couple of long lens shots. No biggie.

class="style8">>>Mark, I totally agree with this scenario, I had trouble with your earlier >>more general statement.

>Sorry, I tried to relate that when you're losing coverage over some little detail that's not critical (and in this case with a clap stick backup) it just seems myopic. There's plenty of times when I have to improvise on the camera/lighting details a bit until take 2 or 3. It just happens when you're forced to move too fast - you gotta roll with the punches (damn, my 2nd bad pun for the 'fistfight thread'... I'd used 'duke it out' previously).

>Mark Doering-Powell
LA based DP


class="style8">>>Let us know if the slates recorded 12 hours later line up, within 1/2 >>frame.

>That is pretty much the experiment I mentioned earlier with my D600s. (Well not exactly, I only let them run for 10 hours - what can I say, I'm impatient)

>Now, a really exhaustive test would include a variety of cameras from different manufacturers, a variety of sound equipment, etc.- all running simultaneously.

>George Hupka
Director/DP
Downstream Pictures
Saskatoon, Canada