Deva Timecode is gonna get me sacked

[studiomprd]

" Oh i was tuning the alexas crystals to the acc501 "

Well, that makes you the fall-person...

Unless the DP specifically requested, and there was a problem, and it can be determined that the Arri Alexa TC (by Ambient) is clearly out of spec, but even then, I would never dare to do something like that to another department's gear; how would you like it if the 1AC re-tuned your Lockit ??...

1 frame in 6 hours is industry standard, and has been easily dealt with for years!

messing with the Camera, now that could get you sacked!

Nope, unless it is determined out of spec, and we all ( as in: DP, and camera owner) agree to any attempt to re-tune...

IMHO that would be not just going overboard, but seems a bit AR.

Edited September 12, 2012 by studiomprd

[Jeff Faber]

HOPKIN:

Yeah, the procedure you describe in your original post is pretty far out of the mainstream. Unheard of actually. I don’t think most people who read your original post understood what you were doing. Hence my question to you in post #101.

In your post #10 you quote David Hall, but he’s actually doing something different:

David Hall explains it better. This is taken from his website: you will have to replace Ambient Lockits with Arri Alexa

"I use Ambient Lockits and the newer ACC501 Master Controller. I am careful to tune the ACC501 to the Deva/Fusion (using the tune to LTC (Linear Timecode) option) and then the camera Lockits are tuned to the ACC501. I then have the ACC501 plugged into the Deva/Fusion where I can keep an eye on any developing timecode difference. Once or twice a day I hit the Jam button on the Deva but the difference is rarely more than a few “hundredths” of a frame. (I say “hundredths” but to be precise due to the niceties of SMTPE timecode the ACC501 is probably displaying 80ths or 160ths of a second)."

He’s using the Ambient ACC501 Master Controller to tune the Lockits which then in turn are attached to the cameras to provide time code and trilevel sync.

He’s not actually tuning the internal clocks on the cameras themselves.

[HOPKIN]

The arri alexas crystal is the same as an ambient and so can be tuned in the same manner. Can someone tell me what the difference is

[Philip Perkins]

The arri alexas crystal is the same as an ambient and so can be tuned in the same manner. Can someone tell me what the difference is

The diff is that in my experience you are way out of your lane. Even if you are a qualified Arri tech, it seems unlikely that they hired you to do this work on the cameras. Most of us have worked successfully for decades worth of TC shoots without tuning anything, and you are trying to tune everything in sight. Admirable, but this exposes you to the risk of running afoul of something you don't know about, esp re cameras. I hope the camera dept and the camera rental house are cooperating with you with all this.

phil p

[studiomprd]

" Can someone tell me what the difference is "

The Arri's crystal is in the CAMERA!

as I asked before, how would you feel about an AC (or even the camera operator or DP) starting to fiddle with tuning-up your wireless ??

[HOPKIN]

John if your not tuning the Alexa, what is the point of tuning a lockit box thats not c.jamming the camera. The ACC501 can tune the ambient timecode by simply plugging the unit into the camera and referencing the timecode in the tuning menu. This was explained to me by the hire company on my last job, when I asked about tuning external lockit boxes. Which we didnt have so this was the procedure to go with.

[old school]

For the life of me, I truly don't get this thread.

CrewC

[John Blankenship]

John if your not tuning the Alexa, what is the point of tuning a lockit box thats not c.jamming the camera. The ACC501 can tune the ambient timecode by simply plugging the unit into the camera and referencing the timecode in the tuning menu. This was explained to me by the hire company on my last job, when I asked about tuning external lockit boxes. Which we didnt have so this was the procedure to go with.

I'm confused by your questions as they don't seem to reference what I've actually said. Maybe it's because I know nothing about the ACC501 (all my master time code devices are Denecke). When you quote what I'm doing you seem to confuse it with your methodology, which we now find is quite different, as you keep mentioning "tuning."

On set, I do not TUNE... I JAM -- either a one time jam, or a continuous jam, as I deem appropriate. I believe that's what most here would say they do. I also use a time code slate (which I have jammed) and have them clap the sticks. Beyond that, depending upon the needs of the production company, I often send a reference track to the camera.

I wouldn't allow the D.P. to re-calibrate my Deva, so I don't re-calibrate, or re-tune, the camera. I will, however, place a sync box on the camera and do a continuous jam, if I think it's needed.

Other than that I'm not sure what to tell you as you seem to be operating from a different mindset. I wish I could help but I think I've offered what I can.

[RPSharman]

Deva software version 7.55 fixes a problem with timecode jumping around by a frame or two randomly when using certain combinations of frames rates and sample rates. I believe 23.98 at 48khz was the main culprit.

-howy

Well thankfully nobody is using that combination very often ;-)

[soundslikejustin]

I'm confused by your questions as they don't seem to reference what I've actually said. Maybe it's because I know nothing about the ACC501 (all my master time code devices are Denecke). When you quote what I'm doing you seem to confuse it with your methodology, which we now find is quite different, as you keep mentioning "tuning."

The ACC501 allows tuning of timecode clocks in OTHER AMBIENT DEVICES. You cannot tune the ACC501 to the Deva clock, and therefore cannot tune the other lockits or cameras to the Deva clock as the Deva doesn't use an Ambient-based clock and is incapable of providing the required reference signal to the ACC501.

So who knows what you're tuning the Alexas and Lockits to, but it sure isn't the Deva's clock.

John B - all Ambient timecode devices (including those in the Alexa, 7-series recorders and Ambient Lockits) can be calibrated (what he means by TUNE) in the field using the ACC501 to within +/- 0.1ppm.

[John Blankenship]

...

John B - all Ambient timecode devices (including those in the Alexa, 7-series recorders and Ambient Lockits) can be calibrated (what he means by TUNE) in the field using the ACC501 to within +/- 0.1ppm.

Thanks for the explanation.

In the wrong hands too much power is a dangerous thing. <g>

[Jeff Faber]

The ACC501 allows tuning of timecode clocks in OTHER AMBIENT DEVICES. You cannot tune the ACC501 to the Deva clock...

This guy named David Hall, who apparently inspired the original poster HOPKIN, claims you can:

David Hall explains it better. This is taken from his website...

I use Ambient Lockits and the newer ACC501 Master Controller. I am careful to tune the ACC501 to the Deva/Fusion (using the tune to LTC (Linear Timecode) option) and then the camera Lockits are tuned to the ACC501. I then have the ACC501 plugged into the Deva/Fusion where I can keep an eye on any developing timecode difference. Once or twice a day I hit the Jam button on the Deva but the difference is rarely more than a few “hundredths” of a frame. (I say “hundredths” but to be precise due to the niceties of SMTPE timecode the ACC501 is probably displaying 80ths or 160ths of a second).

It sounds like you need the LTC option from Ambient (whatever that is) to do that.

[Bash]

Errrrr......

I believe that you can tune an ACC501 to the TC of a camera, or any other device. So... HOPKIN is right that you can do that. However.... if you tune the ACC501 to the camera TC, how can you be sure that the camera's TC is stable ...

I believe (as mentioned previously) that you can tune the Alexa clock to a value given by the ACC501, BUT ... I also believe that the Alexa forgets that tuning value on a power cycle, so this is not a reliable methodology.

Irrespective of all of this discussion.... we are generally talking about single digit parts per million, and if there is any drift involved, and since we are talking about separate units (video cameras or sound recording machines etc...) running plesiochronously, then any drift involved between jams should be in the order of less than a frame even after a good few hours.

I have done numersous shoots with plenty of cameras and sound recorders, where Lockit boxes nor cameras have been tuned, but we know that the amount of drift over a few hours will be negligable, so we simply rejam every, say, 4 hrs.

Denecke TC equipment cannot be user tuned at all, and yet has been proved to be more than accurate over many hours, as, again, we are talking about parts per million. Thousands of shoots over the years have been synced using Denecke kit, and there is no tuning going on at all.

HOPKIN....

Have you asked post prod a few pertinent questions like....

Is the amount of difference constant?

If yes then how many frames?

If the amount of difference is changing (i.e. drifting), then is it better just after you have jammed all TC units, and does the amount of difference increase over time after you have jammed?

If the TC is drifting, then by how many frames over how many hours?

Are all cameras 'out' by the same amount with respect to the sound recorder?

Kindest regards,

Simon B

[soundslikejustin]

Apologies. It appears I have my info wrong. You can't tune TO devices without the Ambient clock though, I got that much correct. My guess is a combination of Alexa's forgetting the tune setting after a power cycle, and the Deva timecode error after power cycling (referenced from another thread - improved with latest firmware) buildup are both contributing to the mis-match.

[HOPKIN]

Sound like justin you can tune the acc501 to the devas crystal. I understand the Deva doesnt have an ambient but it can be tuned using the TUNE Internal menu. ( i say tune because that is what ambient call it)

After using sound devices for one day everything is back in sync 100%.

Edit happy pressure off. Autosync workflow fully operational.

Will report back on any developments.

Thanks to everyone for all their help.

I should add that I still prefer the Deva. Unfortunately the timecode on this job has to be 100% all of the time. Which the Deva cannot do. Look forward to a firmware update soon so I can get back on it.

[soundslikejustin]

Sound like justin you can tune the acc501 to the devas crystal. I understand the Deva doesnt have an ambient but it can be tuned using the TUNE Internal menu. ( i say tune because that is what ambient call it)

Got it, thought you had it trying to go the other way for a bit, got confused.

Anyway, good to know your workflow is working now with the new unit.

[Matt Mayer]

The ACC501 allows tuning of timecode clocks in OTHER AMBIENT DEVICES. You cannot tune the ACC501 to the Deva clock, and therefore cannot tune the other lockits or cameras to the Deva clock as the Deva doesn't use an Ambient-based clock and is incapable of providing the required reference signal to the ACC501.

Hopkin beat me to it.

[John Blankenship]

With everything said, the culprit may be partially due to the (previously discussed) Deva's jitter. However, with the workflow that most here employ, it wouldn't be an issue. That might account for the disparity.

I want to thank Hopkin for hanging in here and continuing the dialog. This can be a tough crowd for anyone who approaches things differently and he's been steadfast and forthcoming in the discussion. There are some who would have gotten defensive and fled in terror as we challenged their workflow. In my opinion, he deserves a good deal of credit for not letting our various comments rattle him a bit.

Our community is enriched by such discussions.

[larry long]

Well said John, and I concur.

[Soundwil]

With everything said, the culprit may be partially due to the (previously discussed) Deva's jitter. However, with the workflow that most here employ, it wouldn't be an issue. That might account for the disparity.

I want to thank Hopkin for hanging in here and continuing the dialog. This can be a tough crowd for anyone who approaches things differently and he's been steadfast and forthcoming in the discussion. There are some who would have gotten defensive and fled in terror as we challenged their workflow. In my opinion, he deserves a good deal of credit for not letting our various comments rattle him a bit.

Our community is enriched by such discussions.

I agree. Discussion is a wonderful thing.

[davidm]

Hi Hopkin,

Here's how to test if your recorder has random offsets in the BWF timecode stamp.

Setup the recorder for internal timecode use. Take the timecode generator output and feed it into audio channel 1 with the TC at about -10dB. Make a series of recordings of at least 1 minute in length. 20 takes would be a good number. You could just feed timecode onto a spare track of your production recordings on a typical work day.

Next, you need two items, the Ambient ACC501 Controller which you have and 2nd, a friendly audio post facility as you will need a Pro Tools HD system or equivalent that can output 23.976 timecode and is locked to stable sync

Import the audio files into Pro Tools and snap them to thier timecode locations on the timeline. You'll need cables from the Pro Tools timecode output to the ACC501 input and from the Pro Tools audio channel output to the ACC501 input. You may need to adjust the audio level output so that it's a level acceptable to the ACC501.

Each files will be played one at a time. The process is to play the file and jam the ACC 501 from the timecode output. Then switch the cables to the timecode on the audio track. As you are in LTC mode, the ACC501 Controller will display any difference between the recorders timecode output (TC on the audio track) and the timecode on the BWF metadata( from the Pro Tools TC output)) down to about 0.01 of a frame.

I ran some tests over the past week with files from a colleagues Fusion (V7.55) and my 788T (V2.19) at 25fps.

The 17 Fusion takes had a relatively constant offset between the TC on the audio track and the TC from the timeline of 0.35 or 0.36 frames. The 20 788T takes had a relatively constant offset between the TC on the audio track and the TC from the timeline of 0.05 or 0.06 frames. I'd guess some of the offset is due to AD conversion on the audio input. So while my tests indicated the Fusion may have a more significant (but sub frame) offset compared to the 788T, there was no random offset in the current software.

A note on my Pro Tools setup. I use a Pro Tools 10 Native system using wordclock from a Lockit and for timecode, one of the few MIDI-SMPTE interfaces that can handle 23.976, an Ambient ALL 601.

[Bash]

Hi David M,

so the time difference, on PB, between Fusion takes, and 788 takes, is approx 0.3 frames (0.35 or 0.36, less 0.05 or 0.06).

The interesting thing is that this would be a constant difference.

I believe that HOPKIN tried to do the offset on burning wav trick that Fusion can do. From memory I believe he went for 40 ms.

I am on mental arithmetic atm, and post pint in the pub... but isnt a frame at 24fps about 40ms long, so are we not taling about a third of that, (0.3) so about 15ms of difference? So - if HOPKIN tried the 40ms offset it might have been rather too much to put things into about the right ballark?

Apologgies for any errors in the above - put it down to a long day, and a couple of pints of London Pride ;-))

Kindest,

sb

[Jeff Wexler]

Hi Hopkin,

Here's how to test if your recorder has an random offsets in BWF timecode stamp.

Setup the recorder for internal timecode use. Take the timecode generator output and feed it into audio channel 1 with the TC at about -10dB. Make a series of recordings of at least 1 minute in length. 20 takes would be a good number. You could just feed timecode onto a spare track of your production recordings on a typical work day.

David's test as described would seem to provide the definitive answer. If Hopkin were able to do this sort of test I think it would determine unequivocally whether the Deva (or both Devas) that he was using is the source of the errors. David's test seems to indicate to me that there is little or no significant differences in timecode between the 788T and the Fusion (at least the 2 recorders he used) so that again would make me suspect something else in the chain rather than the recorders.

This topic has gone on for quite some time now, 3 pages of posts, and I guess that it is a good thing (as someone pointed out it is a test of patience on Hopkin's part to hang in there and report to us all --- I don't know how he is finding the time while working and under the gun, literally). I believe the last post stated that Hopkin is now using a 788T and all is well. So, what have we learned?

[Richard Lightstone, CAS]

What I've learned is never yell "Deva Timecode is gonna get me sacked" on JWSound. I'm also amazed that with all of Hopkin's purported problems and the complaints from post, he did NOT get sacked as he feared.

In my experience, if you blame the equipment, which we Sound Mixers usually own and provide for the timecode problems, then Producers are more likely to fire you quite quickly, usually in the first week the issue arises.

Yet Hopkin was still posting on his problem for over two weeks and is still gainfully employed on the same project; go figure?

[cmgoodin]

One other thing to consider is that DEVA (at least the last time I examined the files) used a different tag than everyone else in the bext metadata to indicate Time Code Frame Rate. It could be that whatever edit software or other plugin that was being used in post couldn't detect the correct frame rate from the bext data so fell back to whatever default it was last set to. This could explain the difference in interpretation from the Deva files and the 788t files.even if they were both being driven from the same Time Code Clock. Frame rate may be represented the same way in the iXML chunk but a lot of software still pulls the Time Code data from the BEXT chunk first.