New article: Z1U Free Run TC as Ext. Sync.

[Chris Hurd]

Prompted by the discussion in this thread about time code, I've assembled a quick tutorial which outlines the process for sustituting the Z1U's Free Run time code for an external clock sync. Questions or comments welcome.

The link is Z1U Free Run Time Code as an External Sync.

This feature is unique to the Z1U and will not work with the FX1.

[Gary McClurg]

Chris, can you jam sync a dat to the pro model?

[Alex Raskin]

Thanks Chris!

Question: what currently available workflow for PC allows NLE to display Z1U's timecode?

[Chris Hurd]

Gary,

There is no dedicated time code input or output on the Z1U, so I don't see any way to jam-sync a DAT recorder to it, sorry.

Alex,

I'm not sure. We'll have to get some other folks to chime in on this.

[H.P. Tschupp]

Since I will use the HDR-FX1 casually to record live concerts (mainly classical) I plan to use my VS2400 (Mixer/40 GB Hard Disk Recorder) for the sound recording with up to 8 simultaneous channels (96khz/24Bit).

Since I'm not aware of any direct SMPTE output of the HDR-FX1, I have to find a way for synchronization of video and audio. I have a concept.

What I found is a box from Roland (SI-80S) which is able to tap into the LANC jack of the HDR-FX1. LANC port is used for a remote control tripod handle (ON/OFF | SBY/REC | ZOOM). But apparently the port also outputs a time pulse which can be picked up by the SI-80S and then is converted into a MIDI time code (MTC).

The VS2400 has a MIDI time code input and can be synchronized with an external MTC source.

If your DAT recoder can be synchronized via MIDI time code, then this solution should also work for you.

This setup would at least synchronize Video and Sound and the two wouldn't "wander off". This is imortant for a longer take.

I'm not sure if I would have any SMPTE available for editing or not. But since I'm also recording the sound with the HDR-FX1, I have kind of a reference sound track which I could use in the editor to precisely overlay with the recorded high quality mix of the VS2400.

I have done this over and over with DV cameras and it worked well. What you do is magnifying the camera audio wave form of track 1/2 in the editor (Final Cut Pro, Premier, Premier Pro, Vegas) and place the high quality sound file as a track 3/4. Then you shift the 3/4 track until it shows the same wave pattern as 1/2. It's possible to do this with the accuracy of one frame. Since the two signals are synchronized the wave pattern should by 100% congruent during the whole sequence.

Once everything works out, you can delete track 1/2 or mute it.

By the way, instead of using a MIDI cable from the SI-80S (which is with the camera), there is a wireless MidiStream device from Kenton.

So you can use microphones with optimized positions for the recording, move with the camera without any cable connection to the recording device!

[Krisada Reyes]

Another method for doing video and audio sync. is writing LTC into one of the camcorder audio track. While doing the audio live recording using multi-track machine I use MTC/SMPTE sync box to convert MTC from multi-track into SMPTE LTC and record into one of the camcorder audio channel while shooting. When play back feed the LTC sync from camcorder (or deck) into the multi-track for slave sync. or using sync box to convert to MTC before feed to the multi-track. In some case I just record the LTC into one of the multi-track channel (sync stripe) before doing live recording to avoid using sync box. This time the camcorder record the audio sync from LTC channel of the multi-track while shooting. The drawback is the Timecode for Video and Audio are not in the same counting (numbers), just ignore the audio one.

[Dylan Pank]

Would this work?

If you have double system and are running free run record, you could simply clap the first shot of the half day, and tail slate the last, and then use the first slate to check for a TC offset between the camera and the audio recorder, then recheck the last for drift.

Then you can just punch in the offset if your NLE can handle it. In my experience though the TC on sony cameras has been pretty precise.

[Krisada Reyes]

Hi Dylan,

It is not free run at the end. When audio recorder record the live sound it run at it own speed but at the same time stripe its’ reference (time code) into the recording camcorder. At that time the camcorder also running at its’ own speed to capture both picture and sound, anyhow time code signal (as a sound signal) from audio recorder will sync. to the picture. (Imagine that the recorded audio time code represent all the audio recording tracks from the audio recorder machine.) When play back by the same camcorder or other deck or captured AV , the playback play both picture and sound at its’ own speed but the picture and recorded time code still in sync. This sound time code then feed to audio play back machine (the recorder machine) to playback as slave sync. (Not free run) Since the picture frame and audio signal (from camcorder) will never slip (if not intentionally done when editing) the time code as audio signal will never slip (with picture frame). And since the original audio time code are generated in sync. with the real audio recording at recording time. So at the end picture and audio will be in sync.
But as I said both picture time code (Originate by camcorder) and audio sync time code can be offset and slip to each other. When editing we can use picture time code as a reference (audio time code as transparent).

[Dylan Pank]

Krisada,

I wasn't responding to your idea directly, just bringing in another possible approach. I think both might be viable approaches, mine would depend on a NLE that could read a timecode offset and take account of sync drift, your approach wouldn't have that problem but of course requires additional LTC timecode generators, as opposed to the built in timecode on the camera and audio recorder.

I think the best thing to do is experiment. I don't have any TC audio equipment to hand (or a Z1U for that matter) but I might try a dummy run with a couple of PD170s that I can use.

[Chris Hurd]

If you do that, Dylan, please report back on how that worked for you. Thanks,

[Dylan Pank]

I will, but don't hold your breath. Projects are coming in here at the Uni so marking is my priority, but if I get a chance to play I'll definitely stick up a post on it - be warned I'm limited to FCP-HD, so all of those saying "Throw your macs in the bin - Edius is here!" are going to have to translate for a Maccite point of view.

:-)

[Dylan Pank]

OK I got to run this experiment with two PD170s to test how close Free Run Timecode will stay in sync between two units.

So this is NOT with a Z1U, I just assume the results would be similar. Also I was not able to do Chris' timecode sync as we only have RM811 remote controls with our cameras, which do not have any way of remotely resetting timecode generation, I just synced by hand (or by two hands!)

Anyway, I ran the experiment like this: I set two PD170s side by side so they would both frame a clapper slate. I set the TC to free run, and then simply pressed "reset" on both cameras at the same time.

I then set the cams recording, then got in front and clapped the slate in the shot. I then switched the record off, went about my business for three and a half hours.

I came back, recorded on both cams and then marked another shot, switched off.

Another Three and a half hours after that, I came back and marked a third and final shot.

So I ended up with about 3 minutes of footage recorded over seven hours.

My Thumbs must have been pretty good when first resetting the TC as the first slates both came in right on 00:01:34:04

Second slate on TAPE A was 03:37:55:09 -- TAPE B was 03:37:55:10

Third slate was TAPE A 07:06:41:10 -- TAPE B was 07:06:41:12

It was sometimes a little tricky to tell but in some cases the clap may actually have been on the preceeding field, so maybe the slates drifted a little less than these numbers imply, but it seems that you can count, on a SONY PD170 at least, on free run to keep in sync for just under 3 hours.

[Shannon Rawls]

Question....

When you set the camera to freerun timecode on the PD170's (or Z1, or XL2, or DVX100a, or whatever has freerun timecode).........and then hit reset, clock starts, turn the cameras off.....

OK, you come back let's say 1 hour later....

As soon as you turn the camera on and take a look in the viwfinder...are those numbers running still? Like a stopwatch? Or is the timecode stamped in the viewfinder the last take you just took, and when you start the recording again, the timecode will 'catch-up' so to speak to the current freerun time??

I sure hope those numbers are just continuously running like a clock on the wall. and I also hope it does this on the Video Display Out, so we can see those numbers in "Video Villiage" when on set shooting a dual camera setup with 2 preview monitors.

Si or No?

[Dylan Pank]

<<<-- Originally posted by Shannon Rawls : Question....

When you set the camera to freerun timecode on the PD170's (or Z1, or XL2, or DVX100a, or whatever has freerun timecode).........and then hit reset, clock starts, turn the cameras off.....

OK, you come back let's say 1 hour later....

As soon as you turn the camera on and take a look in the viwfinder...are those numbers running still? Like a stopwatch? -->>>

Yes, at least it is in the PD170, the PD150 and I'll assume the Z1U and the PDX10. I can't speak for the other cameras.

<<<-- Or is the timecode stamped in the viewfinder the last take you just took, and when you start the recording again, the timecode will 'catch-up' so to speak to the current freerun time?? -->>>

No, the TC will continue to run in the display in camera mode even when the tape is idle, or until you turn the cam into VTR mode (and even then it's still running in the background, so the time will be consistent when you turn it back to cam mode)

<<<-- I sure hope those numbers are just continuously running like a clock on the wall. and I also hope it does this on the Video Display Out, so we can see those numbers in "Video Villiage" when on set shooting a dual camera setup with 2 preview monitors.

Si or No? -->>>

Definitely Si

[Chris Hurd]

<< As soon as you turn the camera on and take a look in the viewfinder...are those numbers running still? Like a stopwatch? >>

For *any* camcorder with Free Run TC: the time code is constantly running, even when the camera is powered down, as long as the time/date batt is fresh.