Tri-level sync, double-system, the way to go? at DVinfo.net

Jeffery Magat · January 10th, 2008, 01:01 PM

I've been racking my brain over this, searching RAMPS, google and the like. I was wondering if double-system (1 cam) with a lock-it box is really that necessary and safe. I've almost always previously gone with timecode recorded onto track 2 of the camera, timecode slate, bla bla bla, but this time around there's a request for a lockit box.. I think this time around they'll be using a Denecke SB-T.

In the past, I've seen people just jam timecode into an F900 with no tri-level. Seems from reading RAMPS that's really not safe from the 'green flashes'.

We'll be using a Sony HDW-700A,for the shoot, and I was wondering if anyone had any suggestions on this issue. I'd really just like some clarification, knowing that I've seen both no tri-level and tri-level being used.

So what is the most common way people are running with these boxes?

Steve House · January 10th, 2008, 01:33 PM

I don't know anything about that camera so can't help with specifics but just a few thoughts while reading your post ... it's common practice in video for the camera to be the timecode master, in contrast to film practice where audio is the master. I understand the Green Flashes you mention are caused when you just jam code into the camera without sending it tri-level as well.

If you set the camera to use record run and send code from the camera to the audio recorder, you don't need the lockits to go double system. You didn't say what audio recorder you were using but it's true there could be a problem with this method with sync drift on long takes since the recorder's sample clock might not slave to the incoming code - Sound Devices slaves their TC clock to incoming code but not their audio sample clock, for example. Takes of 10 or 15 minutes should be okay, an hour might drift.

To insure there's no drift, use two lockits tuned to each other jammed for identical code. The one going to the camera supplies TC and tri-level sync. The one going to the audio recorder supplies TC and Wordclock.

Wolf Seeberg's books have a wealth of information on this. If you don't have a copy, Trew Audio over on Villiers St carries 'em. Check out Wolf's website - wolfvid.com, he's reprinted some of the chapters online and they might answer your questions.

Jeffery Magat · January 10th, 2008, 01:55 PM

I'm actually renting the box from Trew Audio, so I guess I'll have to chat with them about my issue.

Well, mobility for camera is probably going to be an issue, so not too much tethering can be going on. My question is how dangerous is it to not send tri-level. I've seen some people send it, and some people not.

I've read the excerpts from Wolf's book.. I might end picking it up.

Steve House · January 10th, 2008, 02:37 PM

Quote:

Originally Posted by Jeffery Magat

I'm actually renting the box from Trew Audio, so I guess I'll have to chat with them about my issue.

Well, mobility for camera is probably going to be an issue, so not too much tethering can be going on. My question is how dangerous is it to not send tri-level. I've seen some people send it, and some people not.

I've read the excerpts from Wolf's book.. I might end picking it up.

The Denke box you mention sends tri-level in addition to TC so why not go ahead and use it just to be sure?

Wayne Brissette · January 10th, 2008, 03:25 PM

On the MTV Pilot I worked on right before the strike, we use the Denecke SB-T with the F900. We used the Deva as the master time source. I would Jam the SB-T first thing in the morning, lunch, and again after dinner. The SB-T was used to Jam the slate and both the SB-T and TC Slate handed over to the 2nd AC who jammed the camera every time the battery was changed out, or the camera was switched out.

FYI, Steve, the reason audio is the master time code in film is because there is no time code on the film cameras themselves. The audio is time coded and the slate is used on location. From what I understand, the telecine folks stripe the time code onto the film based on these numbers. My sound reports all have time on them, the film department uses length of film.

Wayne

Jeffery Magat · January 10th, 2008, 03:39 PM

Quote:

Originally Posted by Wayne Brissette

On the MTV Pilot I worked on right before the strike, we use the Denecke SB-T with the F900. We used the Deva as the master time source. I would Jam the SB-T first thing in the morning, lunch, and again after dinner. The SB-T was used to Jam the slate and both the SB-T and TC Slate handed over to the 2nd AC who jammed the camera every time the battery was changed out, or the camera was switched out.

FYI, Steve, the reason audio is the master time code in film is because there is no time code on the film cameras themselves. The audio is time coded and the slate is used on location. From what I understand, the telecine folks stripe the time code onto the film based on these numbers. My sound reports all have time on them, the film department uses length of film.

Wayne

Thanks guys.

Well, I'm guessing the timecode is stable enough on the camera just to jam every few hours, and not have the box hanging off camera the whole day?

I think I'll just have to go full on with both TC and Tri-Level to be safe. Not worth it to go half-assed and just use TC.

Wayne Brissette · January 10th, 2008, 03:44 PM

Quote:

Originally Posted by Jeffery Magat

Thanks guys.

Well, I'm guessing the timecode is stable enough on the camera just to jam every few hours, and not have the box hanging off camera the whole day?

I think I'll just have to go full on with both TC and Tri-Level to be safe. Not worth it to go half-assed and just use TC.

We used the F900, so I can't comment on the camera you are using, but they did not hang the Denecke box off the camera, they simply jammed the TC and every once in a while the 2nd AC would verify that the slate and the camera were in sync. We knew the audio (Deva) was in sync because it was the master, and we would jam the Denecke from it.

Wayne

Steve House · January 10th, 2008, 05:46 PM

Quote:

Originally Posted by Wayne Brissette

...

FYI, Steve, the reason audio is the master time code in film is because there is no time code on the film cameras themselves. The audio is time coded and the slate is used on location. From what I understand, the telecine folks stripe the time code onto the film based on these numbers. My sound reports all have time on them, the film department uses length of film.

Wayne

Yep ... that was my understanding too, although I think it's Aaton/Panavision that has a system that prints timecode numbers on the edge of the negative as it goes through the camera. I think there's a temendous amount of misunderstanding on the role of timecode in double-system sound for video recording and what it can and can't do to maintain sync between picture and sound. There seems to be a belief that somehow putting linear timecode on an audio track in the camera, in the recorder, or both can somehow prevent long shots from drifting out of sync. I won't claim to be an authority, but to the best of my knowledge it just doesn't work like that and the best code does is establish a single line up point between picture and sound. Unless the audio and video sample clocks are slaved as well, there's nothing in the code that will prevent the picture and sound from drifitng apart over time as one moves away from that single lineup point.

Wayne Brissette · January 10th, 2008, 07:01 PM

Quote:

Originally Posted by Steve House

I won't claim to be an authority, but to the best of my knowledge it just doesn't work like that and the best code does is establish a single line up point between picture and sound.

That's exactly right. Broadcast wave files only contains a couple pieces of time code, one of which is the very start time when the mixer pressed record. This is the time Final Cut Pro and other NL editing systems use to align audio and video (assuming the video has time code).

One other important part though. In a feature, there really isn't any drift to speak of because the takes aren't all that long. Most of the time you're rolling maybe 3 minutes tops before you stop the camera, re-slate and go again. These short takes will stay in sync a majority of the time. Where time code goes adrift most often is in the video world where there are some sessions that go multiple hours. That's where you learn how good or bad time code really is.

I had lunch with Jeff Wexler in Los Angeles a few months back and he was telling me how one feature he worked on they actually ignored most of the time code and used the clapper to align the audio and film. Most people locally don't use time code, although I have been forcing it on a lot of them mostly because it can help the post production process, but you certainly don't need time code, it's simply a nice to have.

Wayne

Jeffery Magat · January 10th, 2008, 07:03 PM

Just to follow up.. Not sure if you guys read on RAMPS yet (which has a similar post), but Glen Trew recommended just jamming the camera every four hours from the recorder or vice-versa. Same way how Wayne has mentioned for his previous shoot.

And the Sony HDW-700A is basically the 1080i version of the F900 if anyone was wondering. :)

January 10th, 2008, 01:01 PM	#1
Jeffery Magat Regular Crew Join Date: Feb 2007 Location: Toronto, ON Posts: 186	Tri-level sync, double-system, the way to go? I've been racking my brain over this, searching RAMPS, google and the like. I was wondering if double-system (1 cam) with a lock-it box is really that necessary and safe. I've almost always previously gone with timecode recorded onto track 2 of the camera, timecode slate, bla bla bla, but this time around there's a request for a lockit box.. I think this time around they'll be using a Denecke SB-T. In the past, I've seen people just jam timecode into an F900 with no tri-level. Seems from reading RAMPS that's really not safe from the 'green flashes'. We'll be using a Sony HDW-700A,for the shoot, and I was wondering if anyone had any suggestions on this issue. I'd really just like some clarification, knowing that I've seen both no tri-level and tri-level being used. So what is the most common way people are running with these boxes?

January 10th, 2008, 01:33 PM	#2
Steve House Inner Circle Join Date: Mar 2005 Location: Hamilton, Ontario, Canada Posts: 5,742	I don't know anything about that camera so can't help with specifics but just a few thoughts while reading your post ... it's common practice in video for the camera to be the timecode master, in contrast to film practice where audio is the master. I understand the Green Flashes you mention are caused when you just jam code into the camera without sending it tri-level as well. If you set the camera to use record run and send code from the camera to the audio recorder, you don't need the lockits to go double system. You didn't say what audio recorder you were using but it's true there could be a problem with this method with sync drift on long takes since the recorder's sample clock might not slave to the incoming code - Sound Devices slaves their TC clock to incoming code but not their audio sample clock, for example. Takes of 10 or 15 minutes should be okay, an hour might drift. To insure there's no drift, use two lockits tuned to each other jammed for identical code. The one going to the camera supplies TC and tri-level sync. The one going to the audio recorder supplies TC and Wordclock. Wolf Seeberg's books have a wealth of information on this. If you don't have a copy, Trew Audio over on Villiers St carries 'em. Check out Wolf's website - wolfvid.com, he's reprinted some of the chapters online and they might answer your questions. __________________ Good news, Cousins! This week's chocolate ration is 15 grams!

January 10th, 2008, 03:25 PM	#5
Wayne Brissette Major Player Join Date: Aug 2007 Location: Austin, TX Posts: 383	On the MTV Pilot I worked on right before the strike, we use the Denecke SB-T with the F900. We used the Deva as the master time source. I would Jam the SB-T first thing in the morning, lunch, and again after dinner. The SB-T was used to Jam the slate and both the SB-T and TC Slate handed over to the 2nd AC who jammed the camera every time the battery was changed out, or the camera was switched out. FYI, Steve, the reason audio is the master time code in film is because there is no time code on the film cameras themselves. The audio is time coded and the slate is used on location. From what I understand, the telecine folks stripe the time code onto the film based on these numbers. My sound reports all have time on them, the film department uses length of film. Wayne __________________ Mics: KMR 82 i, NTG-1, MKH418S, MKH8040, SR77, QTC1, QTC40, SR30 Recorder: Zaxcom Deva 5.8 & MIX-12. Wireless: TRX900 stereo, Lectro 411

January 10th, 2008, 01:55 PM	#3
Jeffery Magat Regular Crew Join Date: Feb 2007 Location: Toronto, ON Posts: 186	I'm actually renting the box from Trew Audio, so I guess I'll have to chat with them about my issue. Well, mobility for camera is probably going to be an issue, so not too much tethering can be going on. My question is how dangerous is it to not send tri-level. I've seen some people send it, and some people not. I've read the excerpts from Wolf's book.. I might end picking it up.

January 10th, 2008, 07:03 PM	#10
Jeffery Magat Regular Crew Join Date: Feb 2007 Location: Toronto, ON Posts: 186	Just to follow up.. Not sure if you guys read on RAMPS yet (which has a similar post), but Glen Trew recommended just jamming the camera every four hours from the recorder or vice-versa. Same way how Wayne has mentioned for his previous shoot. And the Sony HDW-700A is basically the 1080i version of the F900 if anyone was wondering. :)