New article: XL2 Free Run TC as Ext. Sync. at DVinfo.net

Chris Hurd · November 23rd, 2004, 04:49 PM

Here's a quick tutorial called Free Run Time Code as an External Sync. This simple procedure for setting frame-accurate identical time codes to several XL2's will greatly aid a multi-cam shoot, and is every bit as good as having an external sync input to each camera. Questions or comments are welcome.

Kevan Forrest · November 23rd, 2004, 08:18 PM

Thanks Chris, very informative! This sounds easier to deal with than running extra cables all over the place.

Aaron Koolen · November 23rd, 2004, 08:33 PM

OK, so all one needs to do now is make a device that can turn regular timecode into an IR signal for the Canon and you're away, you can sync to other devices..

Aaron

Anders Hochheim · November 23rd, 2004, 09:11 PM

Great article about simultanous sync. And brilliant timing. I was thinking about a few hours ago cause I have a job with two cameras in the near future.

Jeff Miller · November 23rd, 2004, 10:11 PM

<<<-- OK, so all one needs to do now is make a device that can turn regular timecode into an IR signal for the Canon and you're away, you can sync to other devices. -->>>

On another video board someone tossed out the idea of using Bluetooth for such a purpose...

A. J. deLange · November 24th, 2004, 01:16 PM

This method has some limitations which depend on the stability of the cameras' internal clock crystals. Typically these are good to 100 ppm with good ones being 10 ppm. Using the more optimistic number the relative stability between two cameras (assumed to be at the same temperature) would be about 14 ppm. This means they could (this is a statistical thing) differ by 1 frame (at 30 fps) after about 2300 seconds (38 minutes). Using the poorer stabilty of 100 ppm this goes down to 3.8 minutes. If the circuit is designed with this kind of use in mind by employing a TCXO an order of magnitude better stability is possible meaning 380 minutes before a slip of 1 frame. As I have no idea about the basic stability of the crystals being used in the camera I can't say how well cameras will stay synced. I'd suggest a test in which 2 cameras' clocks are brought together using the technique described and then set on the shelf (together so they are at the same temperature) for say 24 - 48 hrs hrs. Now take a video of a clapper or something similar on each and compare the timecodes to see how many frames apart the cameras have drifted. If it's 48 frames in 48 hrs then the relative rate is 1 frame per hour, the stability is around 100 ppm and one knows that he must resync every half hour or so to keep the two together. If the slipage is 5 frames in 48 hrs that's about 0.1 frame per hour, the oscillators must be in the 10 ppm performance band and resynch every 5 hours or so would be sufficient.

I've considered only stability here. Crystals also drift as they age and are subject to temperature variation so paired cameras should be kept at the same temperature if possible and should be re-synched if one has been subject to a temperature variation to which the other has not. Also, the test I described above should be repeated every month or so and a change in the relative rate expected until the camera is 3 - 6 months old after which the relative rate should settle out to a more or less constant value.

Chris Hurd · November 24th, 2004, 02:27 PM

Thanks for the detailed input, A.J. I agree that this method should be tested for sync first before being used on a paying shoot. Someone other than myself will have to do this, of course, since I doubt I'll ever have two XL2's in the same place for any real length of time.

Lauri Kettunen · November 25th, 2004, 02:46 AM

I can second what A.J. says, and my experience in the past of running (music) recording machines in sync suggests the idea will not be that practical. (If the time code was lost, the machines were within a few minutes out of sync.)

Does anybody know that the Xl2 has no time code output/input (through the BNC jacket), or is that the conclusion we make for the manual does not say anything about that?

Recall that the Xl1 had color bars although the manual said nothing about them. Furthermore, the first news of the Xl2 in July suggested the camcorder would have something like time code input/output. Besides, without input/output, there is not that much use to the time code, and it's hard to believe the designers would not have been aware of this. I guess I'm not the only one who would like to have an external audio recorder in sync with the Xl2 to make 5.1 (or 7.1) surround sound recording.

James Emory · November 28th, 2004, 04:36 PM

Here is a possible solution

www.spcomms.com/ltcexport

James Emory · December 2nd, 2004, 10:01 PM

I guess one good way to find out if and how much drift there may be is to perform the above sync process and leave the two XL-2s on for 24 hours and take a look at the code the next day for any deviation.

J. Michael Oyster · June 13th, 2007, 05:42 PM

I'm interested in knowing if the same procedure published by Chris Hurd for syncing two XL2's would work equally well on a pair of HV20s?

Also, I considering purchasing one HV20 and one XHA1 as my basic camera kit and two used occasionally for two camera shoots. In order sync up the two camaras using the procedure in Chris Hurd's article with two different camera models it would seem that either the HV20 would be able to respond to the remote on the XHA1 or the XHA1 would be able to respond to the remote on the HV20. Is this possible?

Generally I shoot live classical music performance and edit in FCP. Because I don't switch cameras live, genlock seems to be overkill and free run time code may be accurate enough for my application. I'd rather purchase an XHA1 and a HV20 than two HV20s and don't want to spend the money for two XHA1s so would like to know if there's a way to sync the two different cameras via Free Run Time Code. (within limits, I believe minor amounts of drift can be covered with cross-dissolves).

I capture the audio separately to Protools on a Mac and use a clapboard to match up audio and video.

I've done a little two camera shooting before (XL1 + GL1)but I'm a complete newbie to HD.

Chris Hurd · June 13th, 2007, 06:13 PM

Sorry, there's no Free Run time code feature in the HV20.

This method will work for Canon camcorders that offer Free Run time code, which include the current XL and XH series camcorders. One remote should trigger them all -- I don't think the remote codes are different between them.

Not possible with the HV10 or HV20 though.

November 23rd, 2004, 04:49 PM	#1
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,366 Images: 513	New article: XL2 Free Run TC as Ext. Sync. Here's a quick tutorial called Free Run Time Code as an External Sync. This simple procedure for setting frame-accurate identical time codes to several XL2's will greatly aid a multi-cam shoot, and is every bit as good as having an external sync input to each camera. Questions or comments are welcome. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

November 23rd, 2004, 08:33 PM	#3
Aaron Koolen Trustee Join Date: Apr 2002 Location: Auckland, New Zealand Posts: 1,727	OK, so all one needs to do now is make a device that can turn regular timecode into an IR signal for the Canon and you're away, you can sync to other devices.. Aaron __________________ My Website Meat Free Media

November 24th, 2004, 02:27 PM	#7
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,366 Images: 513	Thanks for the detailed input, A.J. I agree that this method should be tested for sync first before being used on a paying shoot. Someone other than myself will have to do this, of course, since I doubt I'll ever have two XL2's in the same place for any real length of time. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

November 28th, 2004, 04:36 PM	#9
James Emory Regular Crew Join Date: Aug 2002 Location: Atlanta, Georgia Posts: 1,892	DV Time Code Extractor Here is a possible solution www.spcomms.com/ltcexport

June 13th, 2007, 05:42 PM	#11
J. Michael Oyster Tourist Join Date: Jun 2007 Location: Encinitas, CA Posts: 1	Free Run Time Code on XH A1 and HV20 I'm interested in knowing if the same procedure published by Chris Hurd for syncing two XL2's would work equally well on a pair of HV20s? Also, I considering purchasing one HV20 and one XHA1 as my basic camera kit and two used occasionally for two camera shoots. In order sync up the two camaras using the procedure in Chris Hurd's article with two different camera models it would seem that either the HV20 would be able to respond to the remote on the XHA1 or the XHA1 would be able to respond to the remote on the HV20. Is this possible? Generally I shoot live classical music performance and edit in FCP. Because I don't switch cameras live, genlock seems to be overkill and free run time code may be accurate enough for my application. I'd rather purchase an XHA1 and a HV20 than two HV20s and don't want to spend the money for two XHA1s so would like to know if there's a way to sync the two different cameras via Free Run Time Code. (within limits, I believe minor amounts of drift can be covered with cross-dissolves). I capture the audio separately to Protools on a Mac and use a clapboard to match up audio and video. I've done a little two camera shooting before (XL1 + GL1)but I'm a complete newbie to HD.

November 23rd, 2004, 08:18 PM	#2
Kevan Forrest Tourist Join Date: Nov 2004 Location: Tulsa OK Posts: 4	Thanks Chris, very informative! This sounds easier to deal with than running extra cables all over the place.

November 23rd, 2004, 09:11 PM	#4
Anders Hochheim Tourist Join Date: Oct 2004 Location: Copenhagen, Denmark Posts: 2	Great article about simultanous sync. And brilliant timing. I was thinking about a few hours ago cause I have a job with two cameras in the near future.

November 23rd, 2004, 10:11 PM	#5
Jeff Miller Major Player Join Date: Nov 2004 Location: St Louis, MO Posts: 227	<<<-- OK, so all one needs to do now is make a device that can turn regular timecode into an IR signal for the Canon and you're away, you can sync to other devices. -->>> On another video board someone tossed out the idea of using Bluetooth for such a purpose...

November 24th, 2004, 01:16 PM	#6
A. J. deLange Major Player Join Date: Jun 2004 Location: McLean, VA United States Posts: 749	This method has some limitations which depend on the stability of the cameras' internal clock crystals. Typically these are good to 100 ppm with good ones being 10 ppm. Using the more optimistic number the relative stability between two cameras (assumed to be at the same temperature) would be about 14 ppm. This means they could (this is a statistical thing) differ by 1 frame (at 30 fps) after about 2300 seconds (38 minutes). Using the poorer stabilty of 100 ppm this goes down to 3.8 minutes. If the circuit is designed with this kind of use in mind by employing a TCXO an order of magnitude better stability is possible meaning 380 minutes before a slip of 1 frame. As I have no idea about the basic stability of the crystals being used in the camera I can't say how well cameras will stay synced. I'd suggest a test in which 2 cameras' clocks are brought together using the technique described and then set on the shelf (together so they are at the same temperature) for say 24 - 48 hrs hrs. Now take a video of a clapper or something similar on each and compare the timecodes to see how many frames apart the cameras have drifted. If it's 48 frames in 48 hrs then the relative rate is 1 frame per hour, the stability is around 100 ppm and one knows that he must resync every half hour or so to keep the two together. If the slipage is 5 frames in 48 hrs that's about 0.1 frame per hour, the oscillators must be in the 10 ppm performance band and resynch every 5 hours or so would be sufficient. I've considered only stability here. Crystals also drift as they age and are subject to temperature variation so paired cameras should be kept at the same temperature if possible and should be re-synched if one has been subject to a temperature variation to which the other has not. Also, the test I described above should be repeated every month or so and a change in the relative rate expected until the camera is 3 - 6 months old after which the relative rate should settle out to a more or less constant value.

November 25th, 2004, 02:46 AM	#8
Lauri Kettunen Major Player Join Date: Aug 2004 Location: Kangasala, Finland Posts: 445	I can second what A.J. says, and my experience in the past of running (music) recording machines in sync suggests the idea will not be that practical. (If the time code was lost, the machines were within a few minutes out of sync.) Does anybody know that the Xl2 has no time code output/input (through the BNC jacket), or is that the conclusion we make for the manual does not say anything about that? Recall that the Xl1 had color bars although the manual said nothing about them. Furthermore, the first news of the Xl2 in July suggested the camcorder would have something like time code input/output. Besides, without input/output, there is not that much use to the time code, and it's hard to believe the designers would not have been aware of this. I guess I'm not the only one who would like to have an external audio recorder in sync with the Xl2 to make 5.1 (or 7.1) surround sound recording.

December 2nd, 2004, 10:01 PM	#10
James Emory Regular Crew Join Date: Aug 2002 Location: Atlanta, Georgia Posts: 1,892	I guess one good way to find out if and how much drift there may be is to perform the above sync process and leave the two XL-2s on for 24 hours and take a look at the code the next day for any deviation.

June 13th, 2007, 06:13 PM	#12
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,366 Images: 513	Sorry, there's no Free Run time code feature in the HV20. This method will work for Canon camcorders that offer Free Run time code, which include the current XL and XH series camcorders. One remote should trigger them all -- I don't think the remote codes are different between them. Not possible with the HV10 or HV20 though. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!