TC difference between tape and FS-4Pro
 

Hi

I´ve just purchased a FS-4Pro (pal) and after reading the operating manual and doing a few basic operations I have always had the following problem: the time code recorded on the hard disc is 3 frames ahead of the one recorded on the tape.

I am recording in Avid OMF format and using Avid XpressDV editing software. Everything works ok but when digitizing footage from the tape and comparing with the clips loaded into the media tool directly from the FS-4, I always get the difference of three frames between both recorded time codes.

Any ideas? The camera used is the Sony DSR PD150P.

Thanks

Pio Rospide

Unrelated perhaps but I can do you one better. When we record in Raw DV format to the drives, we can look at the files playing back on the HD unit and the TC seems to match the tape TC. When we drag the folder of clips to the desktop of our Avid systems and then do an import to Avid as .dv files, all the time codes on the clips reset. Every clip has a begining TC of 00:00:00:00 ! What's up with that? I have tried importing to Vegas, Sorenson Squeeze and Premiere Pro. There is apparently no TC info in the Raw DV files.

Odd.

Sean McHenry

It's probably more to do with your deck setup in Avid. You can adjust timecode latency in the setup.

Sean - reading Shannons thread this seems to be a problem with the HD version of the product. However there is also another problem - Avid does not support timecode on import unless it is already in a format that it understands, like mxf or omf. In fact within the Avid import settings you can adjust the starting timecode for imported clips manually. It also does not preserve timecode on export except in mxf and omf or using Automatic Duck.

Well, I guess I won't come down too hard on the folks at Focus then. It would be nice to hear from one of their Engineering staff that TC is actually encoded into the Raw DV and other formats. We do see a log file on the FS4 which actually has the TC start times of the clips. If we get really hard pressed to set the TC we can manually set them with the log file but wow, still sucks. I wouldn't want to do that with 100 clips.

You would still think they might have tested this with an Avid and maybe mentioned this before we bought 4 of them? Maybe before telling the world DTE was the way to go without a little * to denote that TC doesn't transfer when actually using the DTE system on thew worlds most popular editing systems. Can we edit, yes we can. Can we match 1000 clips back to tape if we ever need to recall these projects. Nope.

Once again we find that equipment at this level may work fine for single editing setups but when you are depending on them for more than single shot one time projects, there are always issues. Some you just can't overcome.

Don't even ask me what I think about the JVC GY-HD100U camera anymore.

Sean

Sean McHenry

Sean - I think you'll find DTE has not been supported at all in HDV. In DV mode you should be able to encode your clips directly to Avid OMF container (or at least this was an option on the FS-4 Pro). I think Focus are working on DTE solutions for HDV - which I guess would involve working in Avid's MXF format. To do that Focus would have to wait for Avid to support all the HDV formats (at the moment they don't support 720P24 or 25, but they do support 720P30 and all the 1280i formats).

I think everyone is working to catch up with the technology.

What are your problems with GY-HD100U?

Hi,

Based on this thread as well as some others, I wanted to add a note about TC support on FS-4/DR-HD100. I have asked one of our engineers to comment.

The log file is strictly used for diagnostics. It is not used during
any aspect of the playback process, or import process to any NLE.

The "Raw" .dv format is the exact byte stream the camera sends to
FireStore. It tends to be used for applications other than production.
The timecode and the user bits and the audio (shuffled) are all encoded
into this single stream.

Raw .dv is not a particularly convenient stream for editing. Hence file
formats such as OMF and Quicktime and AVI and MXF have evolved, allowing for better access to the data in the file through the use of metadata, frame and sample indices and so on.

FireStore manages the metadata, creates the file wrappers, unshuffles
audio, provides indexing, establishes a naming convention, and provides
the ability to work with TC from the camera or 'restripe' TC using
different rules.

The purpose of the OMF format as specified by Avid is to provide a
standard way to carry in metadata such as timecode, unique IDs, indices,
links, etc. It also allows for separated audio carried in separate
files.

We would also like to note that QuickTime HDV support is due at the end of this month or early next month. The .m2t file format can be used directly by Adobe Premiere Pro 2.0 and Canopus Edius.

Hope that helps,

Matt McEwen

Nice of you to chime in here Matt. Focus seems to be strangely quiet on the issue of timecode in HDV. Can you clarify the following:

Does the DR-HD100 work like the FS-4 Pro and wrap in Avid and Quicktime formats for standard def DV?

Is timecode actually supported in HDV or is the problem with the editing systems using the .m2t streams?

Does Focus have any plans to support the Avid .mxf format for HDV? (or.omf) Avid seems to be slowly abandoning the older .omf format in favour of the more generic .mxf format.

Hi John,

Here is some answers for you...

>Does the DR-HD100 work like the FS-4 Pro and wrap in Avid and Quicktime formats for standard def DV?

Yes, in SD mode, DR-HD100 can record QuickTime DV and Avid OMF DV.

>Is timecode actually supported in HDV or is the problem with the editing systems using the .m2t streams?

Yes, for 720p HDV we record the GOP TC value from the stream. For 1080i HDV, we record the TTC timecode value in the stream. As far as we can tell, Avid utilizes the GOP TC value and we have confirmed this.

We are looking at Avid MXF HDV support for DR-HD100/FS-4Pro HD. Like QuickTime, the audio must be demuxed first and then wrapped in a MXF wrapper...which is quite processor intensive. I cannot comment on whether MXF HDV support is going to happen yet as we have not totally investigated what is involved yet.

Hope that helps.

Matt

Matt - that's really good info and should clarify things for this thread.

I guess the problem with GOP timecode in the .m2t stream is that Avid only looks at this when being played from a deck, not on import.

It seems like a reasonable feature request for Avid or perhaps you guys could build a standalone software tool for post encoding to MXF, to take the load of the DR-HD100 hardware?

I think you already have some kind of conversion suite... it could be incorporated into that. Not as quick as DTE but should be almost as fast on quick machine.

Hi John,

Thanks for the suggestion. I think if we can get the MXF wrapper working, this would be the best solution.

Matt

Matt,

I just ordered the DR-HD100, and I've been using Premiere Pro 1.5.1 with the Aspect HD 4.1.2 from Cineform. Am I going to have this timecode trouble? I know that the Cineform software produces m2t files as well as a special version of avi files, but so far I have not found a way to work with them in PPro 1.5.1. The avi files do just fine, most of the time, but m2t get a "file type not supported" error message on import. Apparently an upgrade to 2.0 is becoming imperative, but will it be enough?

Hi,

Premiere Pro 2.0 does support direct .m2t import. The .m2t should also import into Cineform (for wavelet based conversion).

Hope that helps,

Matt

Our FS4/DR-HD issues
 

Matt and others. I had sent a rather lengthy report to Jim but that was about 2 weeks ago now and nothing heard back on TC issues here.

My findings in a 30+ minute test show odd results. Here is the jist of the test to recap for others to compare:

I recorded about 34 minutes of material to the DR-HD100 product via 1394 in the synchro/slave mode directly off a DSR-45 Sony deck. In looking at the footage we noticed the following oddities.

TC at the first of the three 9+ minute breaks was off by 6 frames. While the video did not actually loose any frames it was a mystery to us where the 6 frames went. Did the device drop them on the first clip? Did it drop them in the transition? Was the entire recording off by 6 frames +/- ?

In examining the tape TC recorded simul on the deck and the DR-HD100 unit, we saw the following table of info -

(frame before 1st 9min break) DR-HD100 time = 00;11;26;00 was actually 00;11;26;03 on tape.
(frame after 1st 9min break) DR-HD100 time = 00;11;26;06 was actually 00;11;26;04 on tape.

(frame before 2nd break) DR-HD100 time = 00;20;39;07 was actually 00;20;39;04 on tape.
(frame after 2nd break) DR-HD100 time = 00;20;39;08 was actually 00;20;39;05 on tape.

(frame before 3rd break) DR-HD100 time = 00;29;52;11 was actually 00;29;52;09 on tape.
(frame after 3rd break) DR-HD100 time = 00;29;52;12 was actually 00;29;52;10 on tape.

So, to recap, the tc from the frame before and after on the DR-HD100 at the first 9+min break is 6 frames different but the video was right frame to frame, as it was on tape.

On the drive, other than the odd 6 frame issue on the first break, the unit seems to be several (2-3) frames off throughtout the recording.

While this seems to not be a big issue with smaller one-off, single suite/NLE projects, in a large post house such as my day job, this is a big issue. We cannot use these on long form projects with frame accuracy. EDLs will need to be adjusted for each transition, redigitizing tapes is going to be an issue if we need to revisit a project down the road, audio sync will be an issue with tape vs hard drive material from multi camera shoots, etc.

Not sure where the issue lies but it is an issue.

Sean McHenry

Just thought I would check these issues nearly a year later. Is this still an issue or have others not seen any of this? Perhaps nobody here on the forums is attempting to use these i a true Post House situation or as a solution in Multi-Camera shoots?

We are now looking at the nNovia units as reliable replacements.

Sean McHenry