NanoFlash/Tape time code mismatch at DVinfo.net

Francis Gagnon · September 4th, 2009, 01:44 PM

Im recording on tape and the nanoFlash at the same time on a Panasonic HDX900 in 1080I

The nanoFlash is set in .MXF file, HD SDI timecode embeded triggered record 35Mbps Long GOP.

There is a time code mismatch between the tape and the nanoFlash.
Is there a way to fix this ?

Thank you

Dan Keaton · September 4th, 2009, 03:49 PM

Dear Francis,

Could you tell us a little more about the problem.

Is the time code off a fixed amount, such as a frame or two?

Francis Gagnon · September 10th, 2009, 03:14 PM

When i do some record/pause there is missing frames between each clips

here is the time code of the 6 clips as seen on FCP

10:00:01;22--10:00:17;21
10:00:17;00--10:00:26;29
10:00:31;29--10:00:37;13
10:00:41;16--10:00:47;15
10:00:47;22--10:00:57;21
10:00:57;14--10:01:06;00

there is no missing on the tape between each clips

Dan Keaton · September 10th, 2009, 04:47 PM

Dear Francis,

We will test this at our office to determine if we can duplicate what you are seeing.

Dan Keaton · September 10th, 2009, 06:16 PM

Dear Francis,

We have been discussing your timecode lapses.

Since you are running tape at the same time, we wonder if the following is happening:

1. You record something on tape.

2. You stop.

3. You roll again.

At this point, we wonder if you camera repositions the tape, to line up the tape to start immediately after the last take.

In doing so, it may roll timecode (and it may roll backwards, as the tape is backed up slightly to reposition the tape.)

If it does roll timecode, backwards or forwards, we interpret this as a signal to start recording.

The above is just a theory at this time.

We know of other cameras that unload the heads after a pause/stop in recording, then must reposition the tape, and while this is happening, the timecode does roll.

Rafael Amador · September 10th, 2009, 10:05 PM

Hi Francis and Dan,
I don't know if with the tapes is different than when recording in the NANO and in the SxS as the same time.
In this case, the camera and the NANO start and stop recording at different moments (SxS start few frames earlier and finish few frames earlier), but the TC is the same frame by frame.
Cheers,
rafael

Dan Keaton · September 11th, 2009, 02:18 AM

Dear Rafael,

Thank you for your confirmation that the timecodes on each frame are identical, in your testing.

Francis Gagnon · September 11th, 2009, 07:04 AM

Hi there

thanks for the fast answer
It gave me some tips to what to look at, I really appreciate :-)

I've run some othere test this morning

I did a time code burning recording with the NanoFlash.

The NanoFlash is most of the time late next to the burned in time code.

The tape is dead on with the burn in time code.

thanks

Alister Chapman · September 11th, 2009, 07:11 AM

If you have the pre record buffer "on" the timecodes will not match as the timecode is captured directly from the SDi Stream while the video is delayed by 4 seconds in the buffer. As a result there is a 4 second difference between the camera timecode and the NanoFlash timecode.

In addition the NanoFlash needs to see timecode incrementing before it will go into record, so there is a delay of a few frames at the start and end of each recording if you are using timecode to trigger. This can result in some small gaps or in some cases some overlaps.

Dan Keaton · September 11th, 2009, 07:16 AM

Dear Francis,

We will be testing this.

Does you camera, the HDX900 allow you to burn in the timecode?

Or did you record an image of a timecode slate?

We are surprised by the timecode being off as we just completed extensive testing of the timecode portion of our firmware. But, of course, we can not test with every camera.

Tommy Schell · September 11th, 2009, 08:50 AM

Hi,

Yes as Alister mentioned, the timecode is off when pre-record buffer (System->Pre-Buffer) is used. This is a bug in the software right now.
Otherwise, with a window burn of the timecode, our starting timecode should exactly the window burn or maybe sometimes be off by a single frame, when the timecode is viewed in Final Cut Pro.
Is this not the case?

Tommy

Francis Gagnon · September 11th, 2009, 09:49 AM

Hi

Im not using the pre-record buffer.

I record from a Panasonic HDX900 1920x1080 60i
drop frame in both the HDX and the NanoFlash.
Tested in Record run.

Rafael Amador · September 11th, 2009, 08:25 PM

Hi Francis,
You are not using the NANO buffer, but you may have activated the one in the camera (Pre-recording function). That one could hold 7 seconds picture.
rafael

Francis Gagnon · September 14th, 2009, 11:44 AM

hi There

I just got another unit
Probaly the other one was defective i dont know we'll see.

I'll keep you informed

Steve Brown · September 14th, 2009, 05:28 PM

The HDX900 does allow characters to be inserted in the SDI stream. So, one could record with characters "on" and play back to see if time code displayed on the nano agrees with time code in the burned window. This is something I will check when I can, since I don't have an editing system with which to make side-by-side comparisons.

In fact, one has to be careful not to end up with characters on nano files accidentally. I find it best to use the "Monitor Out" (on the side of the camera) instead of the "Video Out" on the back of the camera. The Monitor Out is more difficult to change (requires a visit to the menu), but the Video Out characters can be added with the flip of a switch... very dangerous. Also, the Video Out can be changed from HD-SDI to SD-SDI (or even composite) very easily if the wrong switch is moved.

September 4th, 2009, 01:44 PM	#1
Francis Gagnon New Boot Join Date: Jun 2009 Location: Montreal, Canada Posts: 7	NanoFlash/Tape time code mismatch Im recording on tape and the nanoFlash at the same time on a Panasonic HDX900 in 1080I The nanoFlash is set in .MXF file, HD SDI timecode embeded triggered record 35Mbps Long GOP. There is a time code mismatch between the tape and the nanoFlash. Is there a way to fix this ? Thank you

September 4th, 2009, 03:49 PM	#2
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Francis, Could you tell us a little more about the problem. Is the time code off a fixed amount, such as a frame or two? __________________ Dan Keaton Augusta Georgia

September 10th, 2009, 04:47 PM	#4
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Francis, We will test this at our office to determine if we can duplicate what you are seeing. __________________ Dan Keaton Augusta Georgia

September 10th, 2009, 06:16 PM	#5
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Francis, We have been discussing your timecode lapses. Since you are running tape at the same time, we wonder if the following is happening: 1. You record something on tape. 2. You stop. 3. You roll again. At this point, we wonder if you camera repositions the tape, to line up the tape to start immediately after the last take. In doing so, it may roll timecode (and it may roll backwards, as the tape is backed up slightly to reposition the tape.) If it does roll timecode, backwards or forwards, we interpret this as a signal to start recording. The above is just a theory at this time. We know of other cameras that unload the heads after a pause/stop in recording, then must reposition the tape, and while this is happening, the timecode does roll. __________________ Dan Keaton Augusta Georgia

September 11th, 2009, 02:18 AM	#7
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Rafael, Thank you for your confirmation that the timecodes on each frame are identical, in your testing. __________________ Dan Keaton Augusta Georgia

September 10th, 2009, 03:14 PM	#3
Francis Gagnon New Boot Join Date: Jun 2009 Location: Montreal, Canada Posts: 7	When i do some record/pause there is missing frames between each clips here is the time code of the 6 clips as seen on FCP 10:00:01;22--10:00:17;21 10:00:17;00--10:00:26;29 10:00:31;29--10:00:37;13 10:00:41;16--10:00:47;15 10:00:47;22--10:00:57;21 10:00:57;14--10:01:06;00 there is no missing on the tape between each clips

September 10th, 2009, 10:05 PM	#6
Rafael Amador Major Player Join Date: Jun 2009 Location: Vientiane (Lao PDR) Posts: 349	Hi Francis and Dan, I don't know if with the tapes is different than when recording in the NANO and in the SxS as the same time. In this case, the camera and the NANO start and stop recording at different moments (SxS start few frames earlier and finish few frames earlier), but the TC is the same frame by frame. Cheers, rafael

September 11th, 2009, 07:04 AM	#8
Francis Gagnon New Boot Join Date: Jun 2009 Location: Montreal, Canada Posts: 7	Hi there thanks for the fast answer It gave me some tips to what to look at, I really appreciate :-) I've run some othere test this morning I did a time code burning recording with the NanoFlash. The NanoFlash is most of the time late next to the burned in time code. The tape is dead on with the burn in time code. thanks

September 11th, 2009, 07:11 AM	#9
Alister Chapman Inner Circle Join Date: Dec 2005 Location: Bracknell, Berkshire, UK Posts: 4,957	If you have the pre record buffer "on" the timecodes will not match as the timecode is captured directly from the SDi Stream while the video is delayed by 4 seconds in the buffer. As a result there is a 4 second difference between the camera timecode and the NanoFlash timecode. In addition the NanoFlash needs to see timecode incrementing before it will go into record, so there is a delay of a few frames at the start and end of each recording if you are using timecode to trigger. This can result in some small gaps or in some cases some overlaps. __________________ Alister Chapman, Film-Maker/Stormchaser http://www.xdcam-user.com/alisters-blog/ My XDCAM site and blog. http://www.hurricane-rig.com

September 11th, 2009, 07:16 AM	#10
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Francis, We will be testing this. Does you camera, the HDX900 allow you to burn in the timecode? Or did you record an image of a timecode slate? We are surprised by the timecode being off as we just completed extensive testing of the timecode portion of our firmware. But, of course, we can not test with every camera. __________________ Dan Keaton Augusta Georgia

September 11th, 2009, 08:50 AM	#11
Tommy Schell Regular Crew Join Date: Nov 2007 Location: Colorado Springs, CO Posts: 65	Hi, Yes as Alister mentioned, the timecode is off when pre-record buffer (System->Pre-Buffer) is used. This is a bug in the software right now. Otherwise, with a window burn of the timecode, our starting timecode should exactly the window burn or maybe sometimes be off by a single frame, when the timecode is viewed in Final Cut Pro. Is this not the case? Tommy

September 11th, 2009, 09:49 AM	#12
Francis Gagnon New Boot Join Date: Jun 2009 Location: Montreal, Canada Posts: 7	Hi Im not using the pre-record buffer. I record from a Panasonic HDX900 1920x1080 60i drop frame in both the HDX and the NanoFlash. Tested in Record run.

September 11th, 2009, 08:25 PM	#13
Rafael Amador Major Player Join Date: Jun 2009 Location: Vientiane (Lao PDR) Posts: 349	Hi Francis, You are not using the NANO buffer, but you may have activated the one in the camera (Pre-recording function). That one could hold 7 seconds picture. rafael

September 14th, 2009, 11:44 AM	#14
Francis Gagnon New Boot Join Date: Jun 2009 Location: Montreal, Canada Posts: 7	hi There I just got another unit Probaly the other one was defective i dont know we'll see. I'll keep you informed

September 14th, 2009, 05:28 PM	#15
Steve Brown Regular Crew Join Date: Aug 2009 Location: Tampa, FL Posts: 96	The HDX900 does allow characters to be inserted in the SDI stream. So, one could record with characters "on" and play back to see if time code displayed on the nano agrees with time code in the burned window. This is something I will check when I can, since I don't have an editing system with which to make side-by-side comparisons. In fact, one has to be careful not to end up with characters on nano files accidentally. I find it best to use the "Monitor Out" (on the side of the camera) instead of the "Video Out" on the back of the camera. The Monitor Out is more difficult to change (requires a visit to the menu), but the Video Out characters can be added with the flip of a switch... very dangerous. Also, the Video Out can be changed from HD-SDI to SD-SDI (or even composite) very easily if the wrong switch is moved.