How to Sync Audio and HDV video

[Steve House]

Originally Posted by Mike Demmers

What is gained is that the Horita sends timecode, which is specifically designed to be sent over any audio path. The Horita is small, and a small cheap wireless transmitter could be attached to it. So it can be pretty portable.

I suppose you could wirelessly transmit the video output from the camera, but that would require a video transmitter - more expensive than something to transmit audio, and not as likey to be in an average kit. Lots of people probably have less-than-great old wireless systems around that they used before they got something good...but would work fine for this purpose.

Work clock is not, like time code, really designed to be sent over an audio path. So while I suppose you might find some device to translate video to word clock, and maybe manage to send it over an audio link, it woud be a hack at best. Also, timecode readers are usually designed to handle transmission errors gracefully. Word clock inputs - not so much.

With timecode, even though the exact timecode will have offset, at least there *some* there and the Tascam can use it to stamp the file.

Also, the device on the remote end may be something else than a Tascam, it could be a computer, perhaps a laptop. With my tiny JLCooper’s PPS-2 timecode to midi timecode converter, I could (loosely, but sometimes good enough) lock anything that will lock to midi timecode (almost any audio/midi production program), with this small box instead of a larger piece of rackmount gear. Timecode/midi timecode is very flexible stuff, you can use it to control lighting, smoke machines, etc. I have a tendency to push the technical edges a bit on productions...

So it's about smallest, most portable, most flexible, easily wireless, cheapest. The cheapest way to do something I don't expect to have to do very often, but which could be very useful from time to time.

Does this make better sense now? Its really about the wireless option. Cheaper than a whole lockit system.

Even the window dub feature might be useful for cueing purposes in certain situations. Sent to any cheap video monitor you get a timecode display almost for free.

-Mike

I understand what you're saying but I'm still concerned about several issues. Firstly, do the majority of the prosumer cameras our posters are using (Sony FX1, Canon XHA1, etc) actually output enbedded timecode in the composite video stream in real time? 2nd, is the Horita extracting and converting video VITC into stable LTC? And the biggy - so we get LTC, what do we do with it? Most of the audio recorders do not slave their audio sample clocks to it. Getting a slate is no problem - the primary issue is getting the camera and the audio recorder clocks to share a common timebase, using the camera as the master and slaving the audio recorder to it. Timecode, smimecode - unless they share a common timebase, after a certain period of time sync will drift. Does generating LTC with the Horita accomplish anything in that regard? Even recording the LTC on a spare audio track in the recorder doesn't do much for us since unlike Avid, 99% of the editing software (Premiere Pro, Vegas, FCP, etc) most DV folks are using doesn't recognise LTC on an audio track or do anything with it if its present (and even if it does, LTC doesn't control the playback rate of a BWF file.)

[Mike Demmers]

I understand what you're saying but I'm still concerned about several issues. Firstly, do the majority of the prosumer cameras our posters are using (Sony FX1, Canon XHA1, etc) actually output enbedded timecode in the composite video stream in real time?

Highly unlikely. And completely irrelevent, because I am concerned with the clocks being locked, not the absolute value of the time code.

The horita is locking to the horizontal and vertical timing pulses imbedded in the NTSC video put out by the camera, which are locked to the cameras master clock. It does not know or care about VITC or tri-level sync.

The Horita MUST lock to these in order to insert characters in the video stream to do the window dub, and to generate the 4800Hz clock for the time code it outputs, which must be in sync with the video also. Therefore any timecode it creates will have its clock locked to the video coming in from the camera.

That is the purpose the device is sold for, after all - to create a window dub while generating frame locked linear time code, the sales burb says that pretty clearly. It would be pretty useless if the timecode were not locked to the window dub.

I got lots of window dubs made this way sent to me in the 1980s, to do video sweetening on my analog 24 track. If it were not locked, I would have been screaming loudly. ;-)

2nd, is the Horita extracting and converting video VITC into stable LTC? No. They don't exist for it. A more sophisticated (and expensive) device might be able to do this, if the cameras output VITC. But I doubt they do.

It is creating its own timecode, but with the timecode clock locked to the incoming video clock.

And the biggy - so we get LTC, what do we do with it?

Well, we have time code with an embedded clock that is locked to our master camera. We can feed it to a recorder that will sync to it, convert it to word clock, convert it to midi timecode, convert it to video black burst, anything. Anything we convert it to wil be locked to the same clock as our original camera.


Most of the audio recorders do not slave their audio sample clocks to it.
But some do. Buy one that does. ;-)

Or buy a device that will convert it to wordclock or back to blackburst. Or use a Motu Ultralite, which will lock to it and output locked audio in digital form, which can be fed into any recorder that can record digital audio, or a computer. It will all be locked to the master camera.

Getting a slate is no problem - the primary issue is getting the camera and the audio recorder clocks to share a common timebase, using the camera as the master and slaving the audio recorder to it. Timecode, smimecode - unless they share a common timebase, after a certain period of time sync will drift.

They do. The timecode generated by the Horita had its clock derived from the video sent by the recorder, it is locked to it. So anything you derive from that clock is still locked to the master camera.

Derive the 4800 Hz clock from the timecode. Divide that by 160, you have 30 Hz - video vertical interval. Multiply it by 10, you have 48000 Hz, which just happens to be wordclock.

All still locked to the original source.

Does generating LTC with the Horita accomplish anything in that regard?

Yes, see above.

Even recording the LTC on a spare audio track in the recorder doesn't do much for us since unlike Avid, 99% of the editing software (Premiere Pro, Vegas, FCP, etc) most DV folks are using doesn't recognise LTC on an audio track or do anything with it if its present (and even if it does, LTC doesn't control the playback rate of a BWF file.)

That would be useless, yes. It doesn't matter though if you are moving files recorded to a disk or tape.

What you are missing is the frame oriented nature of both the audio and video. We are just moving blocks of data around.

Lets see what happens one step at a time. (simplifying a bit, ignoring color, etc)

In the camera you record exactly 30 frames on video in one second. You also, in the camera, may record audio - exactly 48000 frames of audio per second. Or 1600 audio frames per field of video. That relationship never changes, because in the camera the clocks that created both were locked together. In the computer, it just plays one frame of video in exactly the same time as 1600 audio samples. It only knows that is 1/30 second because the camera told it so, in the file. It does not know if the camera is telling the truth, if it is really on speed.

What if the camera is wildly off speed from the computer, say running twice as fast as it should? When you import the data, and play it back, it will run the video at half speed, because the file from the camera said it was 30 frames per second video, even though it really wasn't. The audio from the camera will also play back at half speed, because its data said it was 48000 audio, even though it really wasn't.

The absolute speed is wrong for both, but because both were originally recorded syncronized to the same clock, they are still in sync with each other. There are still exactly 1600 audio frames per one video frame. Even if you seperate the audio from the video, save it as a file, and then reload it, they will still be in sync because when the computer reloads the file, it still just sees it as 1600 audio frames to matchup with each video frame.

The exact same thing happens if the audio was recorded to a separate recorder but which was locked to the cameras clock. So imagine the same fast running camera running twice speed. It is really clocking twice as fast as it should be, but it creates a file that says '30 frames = 1 second'. The audio recorder is synced to this clock,which is actually running twice too fast, but it doesn't know this, it still writes a file that says '48000 frames = 1 second'

So when you load the video file into your computer, it sees 30 frames and plays it as one second of video. It loads the audio file and calls 48000 audio frames one second (even though it was actually running twice as fast). So again, both run exactly at half speed in the computer, and are perfectly in sync wth each other.

The computer plays both audio and video with its own clock. It believes whatever is in the file, that is all it has to go on. 1600 audio per 1 frame video.

When dealing with files and frames, pure digital data, ony the original clock rates they were recorded at must match. That is all that matters. The computer is always playing back both with its single clock.

If you have a Tascam, you can feed it the linear timecode directly.

With one of the Sound devices units, you would need to convert it to word clock. A (small, inexpensive) Motu Ultralight could do that, and function as a mixer and preamp as well.

All should sync up with no drift.

If you consider something like the Motu as just 'the other side of the chain', and add in the cost of a really cheap wireless link, I am pretty sure you are still well under the cost of a lockit system, and the Motu adds many other useful functions (preamps, mixer, computer interface).

I don't know what you would do with this, but here is what I would do:

Buy not the ultralight, but the next one up, the Travelor. Now I could use this with my favorite multutrack audio recording software on a laptop to record a whole band, multimiced just like in the studio, locked to camera. Live, portably, inexpensively, and in sync. I could even run lighting and effects for the band through midi at the same time. ;-)

-Mike

[Mike Demmers]

Originally Posted by Mario Vermunt

I am too looking for a small and inexpensive device that can convert Blackburst to wordclock. My SD702 accepts only wordclock. I do not want to invest in an expensive (and mostly complex) device like the Aardvark since most of the times the audio stays in sync anyway. If Tascam can build such a device in a recorder that is far cheaper than my SD702 it should be possible to have one for say $500.

The PluralEyes software that automatically syncs audio sounds great but I think it will only work to synchronize startpoints and not hold sync for a prolonged time. Some sort of automatical slate therefore (am I right?).

Anyway not to complain though. I used to work with a Uher 4200 taperecorder with a built in sync module that could sync perfotape to the flashoutput of my 8mm filmcamera!

If you don't mind a wired connection, all you need is a simple wordclock generator that will lock to video. Now, why can't I find a cheap one?...

This might work:

MOTU | MIDI Timepiece AV (USB/Serial) MIDI Interface | 1156

$ 499.95

Makes sure it does this function in standalone mode though.



-Mike

[Mario Vermunt]

Wired to the camera is not that much of a problem, but wired to 220V is not that nice ;-)

[Nelson Alexandre]

Ok...

If you are using Sony cameras I can give you a tip that will solve your issues.

It's called TTL and it's from Fisher Robotics.

It was designed to function with Canon cameras also but unfortunately, it seems that XH-A1 has troubles with it. I know... I have one!

What this thing does is convert the embedded LANC TC signal from Sony and Canon LANC terminals to LTC. It works like a breeze... On Sony cameras! Not so on Canon. Strangely, it works perfectly on my old Canon XM2 (the US GL2), but not on my newer XH-A1.

I have it working on a Sony though and it slaves my Tascam HD-P2 with no hassles at all.

If you care, talk with Fisher himself. I'm almost certain that if many XH-A1 users insisted with him he probably would have solved this Canon issue long time ago.

Hope that helps

[Mike Demmers]

Mario - It probably runs on 5 volts internally. Renove the bulky transformer, replace with a battery.

[Jay Massengill]

Another device that might be found more cheaply on the used market: The ProTools Video Slave Driver (circa early 1990's). Its sole function was to create WordClock locked to incoming video. And yes, it was also AC powered and rack-mounted but not very large or heavy. Sorry if this device has been mentioned before, I didn't read the whole thread.
It could be helpful for someone setting up a DIY locked audio solution for minimal cash. I still have mine but none of the other gear from that now antique system.

[Ben Moore]

Originally Posted by Nelson Alexandre

Ok...

If you are using Sony cameras I can give you a tip that will solve your issues.

It's called TTL and it's from Fisher Robotics.

It was designed to function with Canon cameras also but unfortunately, it seems that XH-A1 has troubles with it. I know... I have one!

What this thing does is convert the embedded LANC TC signal from Sony and Canon LANC terminals to LTC. It works like a breeze... On Sony cameras! Not so on Canon. Strangely, it works perfectly on my old Canon XM2 (the US GL2), but not on my newer XH-A1.

I have it working on a Sony though and it slaves my Tascam HD-P2 with no hassles at all.

If you care, talk with Fisher himself. I'm almost certain that if many XH-A1 users insisted with him he probably would have solved this Canon issue long time ago.

Hope that helps

VERY COOL!!, to bad more sub $1000 recorders dont have BNC Word Clock/Video Ref In.

Here is a link to the specs.

http://www.fisher-robotics.com/PDFs/...OLLv1Flyer.pdf

[Mike Demmers]

Originally Posted by Ben Moore

VERY COOL!!, to bad more sub $1000 recorders dont have BNC Word Clock/Video Ref In.

It's not the lack on sub $1000 cameras that bothers me, it's the lack of any kind of sync above that price point. You literally have to go to the very top of the line - if even there - too get any kind of sync at all.

They put really crummy audio sections into these cameras, with the excuse that if you want good audio you are going to go double system anyway. But then make it very hard to do so.

The lack of any kind of sync input means if you want to switch cameras live, you have to put a TBC literally on every input to the switcher, a very expensive proposition. All for the lack of simple backburst sync, something the very first 'consumer' camera I ever used had, back in the video stone age of black and white cameras.

I hear the excuse that the markets are not big enough, but really, this is useful to anyone who is using multiple cameras recording anything longer than 30 minutes or so, or who has to switch live.

I see people doing video for churches here wanting to do this. I see people from corporate environments wanting to do this. These are small markets?

The main market for $5000 cameras is grandma, to make nice videos of her grandchildren? Maybe I am wrong, but that just seems unbelievable to me.

Or why not come into the 21st century and just automatically lock sync to GPS (now appearing even in cheap consumer gear) or use some other kind of wireless sync? These cameras have so many frills yet are so lacking in some basics. They can put complicated auto everything in, but to get a simple switch to manual mode I have to spend much *more* money? Just to be able to change a lens, I have to spend thousands more? It costs so very much more to put a lens mount on a video camera than a still camera?

This market looks so irrational to me.

Very frustrating.

I hope Scarlet and maybe the Chinese start giving some of these companys ulcers soon, to match mine. ;-)

-Mike

[Ben Moore]

You make some good points, though what I meant by sub $1000 recorders was not camera's. I meant flash based audio recorders. With the TTL device you could use any Sony cam with a LANC out and then run a BNC cable to a flash recorder such as the Tascam HD-P2 and maintain sync, sorry for the confusion.

Ben

[Mike Demmers]

Originally Posted by Ben Moore

You make some good points, though what I meant by sub $1000 recorders was not camera's. I meant flash based audio recorders. With the TTL device you could use any Sony cam with a LANC out and then run a BNC cable to a flash recorder such as the Tascam HD-P2 and maintain sync, sorry for the confusion.

Ben

Oops, I guess my general frustration level caused me to misread your post. Sorry....

It is too bad about the inexpensive flash recorders, but here at least I can understand these are not really built for the video market.

[Nelson Alexandre]

Originally Posted by Ben Moore

You make some good points, though what I meant by sub $1000 recorders was not camera's. I meant flash based audio recorders. With the TTL device you could use any Sony cam with a LANC out and then run a BNC cable to a flash recorder such as the Tascam HD-P2 and maintain sync, sorry for the confusion.

Ben

That's exactly right Ben. And for what it matters, I think that Tascam's recorder is THE bang for the buck concerning flash based recorders. So why do you need any others? ;-)

[Steve House]

Originally Posted by Mike Demmers

Oops, I guess my general frustration level caused me to misread your post. Sorry....

It is too bad about the inexpensive flash recorders, but here at least I can understand these are not really built for the video market.

I think one of the reasons the Sound Devices 702T and 744T do not slave to video blackburst or incoming LTC is that their clock/timecode generators are actually made by Ambient and are essentially the same as on-board Lockit boxes in their own right. In fact, if you're using a Lockit on the camera, you can tune the two together just as if they were two Lockits and get to within 1 frame per day accuracy. I'm thinking they just didn't think many folks with $5000 prosumer cameras would be attempting much double system sound while people using full-blown pro cameras would either be using house clock or Lockits.

[Mike Demmers]

Originally Posted by Steve House

I'm thinking they just didn't think many folks with $5000 prosumer cameras would be attempting much double system sound while people using full-blown pro cameras would either be using house clock or Lockits.

Well, I generally agree in practical terms, and in their market. Their clocks are good enough even without extraordinary measures to stay in sync for close to an hour (with first class cameras), and just how often do you need to do takes on a movie set or in ENG that are over an hour long?

But I do see in some of the forums the pros asking each other about interfacing to some of these less expensive cameras being used for things like webisodes and reality shows, where these cameras do not necessarily have such accurate clocks. A use maybe not anticipated when Sound Devices designed their stuff.

The reason we spend the big bucks on stuff at that level is that it is robust, reliable, and flexible in ways that lesser gear that might work fine in the audio sense is not, and because we want to never, ever have tell a producer or director 'Sorry, my gear cannot do that'.

In that regard, my own personal feeling is that such a top of the line recorder should be able to sync in any format that has ever been in general use in the industry. That would include not only all the various kinds of SMPTE, but things like 'pilot tone' and 60 Hz as well. And a pro deck should be able to chase lock on set. It's not just about the 1 in 10000 chance you might need this to use in the normal, old fashioned way, but also because having such extra capabilities sometimes allows you to solve problems in creative ways.

I've done crazy things using capabilities that were used in weird ways - like syncing an analog 24 track to a non timecode dat machine with LTC on track 2, to be able to record 20 more takes of a vocal sequentially, using different offsets. Not exactly a 'normal' use of timecode, especially in a music production! It got us through though, and that was all that mattered in the end. I couldn't have done that without the 'extra' ability of my syncronizer to do offsets, something not normally required to lock two decks together for a music production.

If I were making my living doing sound for picture, I'd buy Sound Devices top of the line recorder without the slightest hesitation. But I would also have to buy one of the Tascam units to cover the hole the in the SD capabilities, and I should not have to do that.

But that is just me, considered a fanatic by some. ;-)

-Mike

[Don Xaliman]

It's got me thinking.
I have not noticed much sync drift in my concert recordings but a bit does appear on the +15 min compositions. So in my quest to better use the equipment I have, it struck me that since I often record separate audio on a Tascam DA-38 8 track digital tape recorder that the simple act of hooking a cable between my Canon XH-A1 composite out and the word sync in, on my Tascam deck, I will accomplish much better sync. Assuming that the length won't vary by the time I capture the audio into the computer, mix it in Wavelab and place the mixed wave file into my video edit timeline.
At least it will be an improvement over the machines all going on their own clock.
Thanks for the insight.