WARNING: Firmware 1.0.64 and tape based cameras at DVinfo.net

Louis Block · October 6th, 2009, 06:21 PM

After extensive testing with the nanoFlash recorder and both the Panasonic SDX-900 and HDX-900 cameras, I’ve determined the cause of the intermittent audio sync issues we’ve been experiencing in recorded files.
At seemingly random points during our production day, audio would drift out of sync as much as 3-4 seconds, without any consistency. On these days, we were also recording tape in camera as a backup medium. Although for the most part, we would check playback using the nanoFlash, occasionally we would also check tape playback. It turns out that this is what affects audio sync. After playing back tape, or using the “cue” button on the camera to cue tape up, any subsequent takes are out of sync.
Power cycling either the camera or the nanoFlash recorder returns recordings to proper synchronization.
Unplugging or turning off the recorder during tape playback avoids the issue altogether. I’m not clear whether this behavior occurred in earlier firmware than 1.0.64.
I’ve spoken to Tommy in tech support at Convergent Design, and he has been very supportive. He is working on a firmware fix. I am posting this info to the dvinfo forums with his knowledge.

Mike Schell · October 6th, 2009, 07:20 PM

Quote:

Originally Posted by Louis Block

After extensive testing with the nanoFlash recorder and both the Panasonic SDX-900 and HDX-900 cameras, I’ve determined the cause of the intermittent audio sync issues we’ve been experiencing in recorded files.
At seemingly random points during our production day, audio would drift out of sync as much as 3-4 seconds, without any consistency. On these days, we were also recording tape in camera as a backup medium. Although for the most part, we would check playback using the nanoFlash, occasionally we would also check tape playback. It turns out that this is what affects audio sync. After playing back tape, or using the “cue” button on the camera to cue tape up, any subsequent takes are out of sync.
Power cycling either the camera or the nanoFlash recorder returns recordings to proper synchronization.
Unplugging or turning off the recorder during tape playback avoids the issue altogether. I’m not clear whether this behavior occurred in earlier firmware than 1.0.64.
I’ve spoken to Tommy in tech support at Convergent Design, and he has been very supportive. He is working on a firmware fix. I am posting this info to the dvinfo forums with his knowledge.

Hi Louis-
Thank you very much for posting this information and for your help in tracking down the problem. We now suspect that the HD-SDI signal out of the Panasonic camera may not have time-code and or audio while going into a cue point. This is perfectly allowable by the HD-SDI spec. We now need to compensate for this issue with some changes to the nanoFlash firmware.

We are tesing new code now and hope to send a beta copy to you by Thursday of this week. If this initial firmare does not fix the problem we do have a backup plan now underway with a different approach to fixing this sync issue.

Best-

Louis Block · October 6th, 2009, 07:46 PM

Thanks, Mike.

The majority of my testing was with the SDX900, a standard def system. The SDX900 does not embed timecode in its SDI stream, only audio. Though the HDX does embed both, I'd lean toward the audio theory.

Louis

Steve Brown · October 10th, 2009, 01:40 AM

Louis,

I have both of those cameras and I am certain the SDX900 does include time code in the SDI stream. There may be an option for that in the menu somewhere, although I can't imagine why one would ever want to turn that off.

Have you looked at the SDI stream of the SDX900 on a digital scope, such as the Leader 5750? If you do that, you should see time code displayed.

Louis Block · October 14th, 2009, 03:07 PM

I have both the SDX and HDX900 cameras here with me now. I've run them into both the nanoFlash recorder and a Leader 5750 scope. Neither the recorder or the scope shows timecode on the SDI output of the SDX. In the "TC/UB" menu of the camera, timecode output is set to TCG/TCR. I know of no other menu that would control what is embedded in the SDI output.
Both the scope and recorder show code correctly from the HDX.

Steve Brown · October 14th, 2009, 03:10 PM

Louis,

I will check mine to see what gives. I know I have used that function many times watching TC on the 5750 as well as audio.

Are you going through the nano, before going to the scope... or are you going to the scope directly from the SDI output?

Louis Block · October 14th, 2009, 03:36 PM

Thanks, Steve.

I am running directly from the SDI output of the camera to the 5750. When testing with the recorder, I am running directly to that as well.

Louis Block · October 14th, 2009, 04:31 PM

After extensive testing today, with both the SDX and HDX 900s, I've come to the conclusion that the aformentioned playback anomaly (audio and video going out of sync) only seems to affect the SDX900. Although I have encountered out of sync files from the HDX900 in the field, I have been unable to recreate the circumstances that caused them, here on the bench.

Dan Keaton · October 14th, 2009, 04:36 PM

Dear Louis,

Please note that we are still actively working to find a solution to this problem.

We have our engineers working on this and we have sent firmware updates to a tape based Panasonic camera user that can create the problem on demand.

Since he can create the problem on demand, we are very confident we can quickly find a fix to this problem.

October 6th, 2009, 06:21 PM	#1
Louis Block New Boot Join Date: Oct 2009 Location: Novato, CA Posts: 5	WARNING: Firmware 1.0.64 and tape based cameras After extensive testing with the nanoFlash recorder and both the Panasonic SDX-900 and HDX-900 cameras, I’ve determined the cause of the intermittent audio sync issues we’ve been experiencing in recorded files. At seemingly random points during our production day, audio would drift out of sync as much as 3-4 seconds, without any consistency. On these days, we were also recording tape in camera as a backup medium. Although for the most part, we would check playback using the nanoFlash, occasionally we would also check tape playback. It turns out that this is what affects audio sync. After playing back tape, or using the “cue” button on the camera to cue tape up, any subsequent takes are out of sync. Power cycling either the camera or the nanoFlash recorder returns recordings to proper synchronization. Unplugging or turning off the recorder during tape playback avoids the issue altogether. I’m not clear whether this behavior occurred in earlier firmware than 1.0.64. I’ve spoken to Tommy in tech support at Convergent Design, and he has been very supportive. He is working on a firmware fix. I am posting this info to the dvinfo forums with his knowledge.

October 14th, 2009, 04:36 PM	#9
Dan Keaton Inner Circle Join Date: Dec 2002 Location: Augusta Georgia Posts: 5,421	Dear Louis, Please note that we are still actively working to find a solution to this problem. We have our engineers working on this and we have sent firmware updates to a tape based Panasonic camera user that can create the problem on demand. Since he can create the problem on demand, we are very confident we can quickly find a fix to this problem. __________________ Dan Keaton Augusta Georgia

October 6th, 2009, 07:46 PM	#3
Louis Block New Boot Join Date: Oct 2009 Location: Novato, CA Posts: 5	Thanks, Mike. The majority of my testing was with the SDX900, a standard def system. The SDX900 does not embed timecode in its SDI stream, only audio. Though the HDX does embed both, I'd lean toward the audio theory. Louis

October 10th, 2009, 01:40 AM	#4
Steve Brown Regular Crew Join Date: Aug 2009 Location: Tampa, FL Posts: 96	Louis, I have both of those cameras and I am certain the SDX900 does include time code in the SDI stream. There may be an option for that in the menu somewhere, although I can't imagine why one would ever want to turn that off. Have you looked at the SDI stream of the SDX900 on a digital scope, such as the Leader 5750? If you do that, you should see time code displayed.

October 14th, 2009, 03:07 PM	#5
Louis Block New Boot Join Date: Oct 2009 Location: Novato, CA Posts: 5	I have both the SDX and HDX900 cameras here with me now. I've run them into both the nanoFlash recorder and a Leader 5750 scope. Neither the recorder or the scope shows timecode on the SDI output of the SDX. In the "TC/UB" menu of the camera, timecode output is set to TCG/TCR. I know of no other menu that would control what is embedded in the SDI output. Both the scope and recorder show code correctly from the HDX.

October 14th, 2009, 03:10 PM	#6
Steve Brown Regular Crew Join Date: Aug 2009 Location: Tampa, FL Posts: 96	Louis, I will check mine to see what gives. I know I have used that function many times watching TC on the 5750 as well as audio. Are you going through the nano, before going to the scope... or are you going to the scope directly from the SDI output?

October 14th, 2009, 03:36 PM	#7
Louis Block New Boot Join Date: Oct 2009 Location: Novato, CA Posts: 5	Thanks, Steve. I am running directly from the SDI output of the camera to the 5750. When testing with the recorder, I am running directly to that as well.

October 14th, 2009, 04:31 PM	#8
Louis Block New Boot Join Date: Oct 2009 Location: Novato, CA Posts: 5	After extensive testing today, with both the SDX and HDX 900s, I've come to the conclusion that the aformentioned playback anomaly (audio and video going out of sync) only seems to affect the SDX900. Although I have encountered out of sync files from the HDX900 in the field, I have been unable to recreate the circumstances that caused them, here on the bench.