DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   JVC GY-HD Series Camera Systems (https://www.dvinfo.net/forum/jvc-gy-hd-series-camera-systems/)
-   -   Broken Clips (https://www.dvinfo.net/forum/jvc-gy-hd-series-camera-systems/65632-broken-clips.html)

Dave Beaty May 4th, 2006 06:51 AM

Quote:

As far as I know, the only workarounds besides capture and output from premiere using the main concept plugin are:

1) capture the m2t stream with DVHSCap or HDVxDV, and use HDVxDV or MPEG streamclip to transcode in to an editing format such as Apple Intermediate Codec or DVCProHD (beware of scaling in the latter).

2) Use a capture card with HD analogue component inputs (Aja or Blackmagic) to capture as DVCProHD or uncompressed (the latter if you have tons of space or a little material). Use the serial connection on the BRHD50 for timecode.

I hope we can get to the bottom of this.
I did some tests changing the prefs for FCPro warnings during capture.
If on TC break, it's set to create new clips as the default is, it does just that. If warn after capture, it actually stops the capture at each break and warns that the TC is discontinuous or gives a stream error at these breaks.

It makes sense that dubbing the tapes via firewire would not repair the problem, but I'm going to test it anyway.

Dave

Ben Brainerd May 11th, 2006 07:18 PM

So, are we simply doomed to capture using 3rd party utilities until doomsday (Or a FCP update)? I'm currently sitting on top of 15 hours of footage that is, at a rough estimate, broken into 100 odd segments. (My 2nd cameraman was a bit trigger-happy. On. Off. On. Off. Repeat ad nauseum)

My cameras are both (A) models, (At the risk of beating the dead horse, I thought this was fixed in the update?) and I'm using the Sony Mini Digi-Master tapes (PHDVM). All the footage is 30p. FCP 5.0.4

I tweaked to capture settings to heck and back, with no results. Changing "On Timecode Break" to anything but "Make New Clip" dumps the transfer with a stream error. AIC doesn't make a difference either. So, erm... Yeah. What's a guy to do?

(Against the likelyhood that the response to that last question will be "Capture with DVHSCap and Streamclip", I'm off to start that now.)

Andrew Young May 11th, 2006 07:35 PM

Quote:

Originally Posted by Ben Brainerd
(Against the likelyhood that the response to that last question will be "Capture with DVHSCap and Streamclip", I'm off to start that now.)

Hi Ben,

If you end up capturing and transcoding entire tapes with this method, I'd love to know how your sync holds up.

A very unfortunate situation, indeed...

Yoochul Chong May 11th, 2006 07:39 PM

i've noticed that if i capture with imovie, and dump the clip that shows up in imovie's bin to my desktop, then import this file into fcp, it captures using aic. no breaks, but its the only alterntive right now.

funny how imovie can capture and fcp can't.

Ben Brainerd May 11th, 2006 10:15 PM

Well, I was hoping to be back with an answer to Andrew's sync question. But, alas, it was not meant to be. Instead, I come begging for anyone who might have a faster bloody way to do this. The capture to .m2t via DVHSCap went fine, if a bit slow. (Next corporate purchase: Firestores)

But the transcode is just bogging down. I set it up to transcode to "Apple HDV 720p30", and let it rip. An hour later, it wasn't even halfway through the 53 minute clip. Streamclip was reporting a data rate of like 2mb/s. Since it's obvious I'm not going to get a lot of productive capturing done, I canceled it, and started experimenting.

Running it through at AIC reports 30-odd mb/s, but it's still still working at effectively realtime. Which kinda irks me. I'm ok with running the tapes through in realtime once to capture, but running at effectively half-realtime is a bit tedious.

Now, so I don't have to get out my fireproof undies, I know transcoding is a beast. My concern is not that transcoding takes forever, which I more or less expect it to. My concern is that I have to transcode in order to get the files into FCP effectively.

I know there are a bunch of folks out there who are making it work for them (Tim? Nate? Paulo?) and I'm wondering if I'm just doing something totally moronic and wrong, or if they're just infinitely more patient than I am.

Carl Hicks May 11th, 2006 10:47 PM

Ben,

Are you seeing the clip segmentation issue on tapes that were shot after the camera was upgraded to "a" version?

FYI: Our product engineers are looking into this issure as we speak.

Regards, Carl

Ben Brainerd May 11th, 2006 11:03 PM

Quote:

Originally Posted by Carl Hicks
Ben,

Are you seeing the clip segmentation issue on tapes that were shot after the camera was upgraded to "a" version?

FYI: Our product engineers are looking into this issure as we speak.

Regards, Carl

Carl,
All of the footage was shot last weekend. I actually only bought the cameras around 3-4 weeks ago. (Actually went into my dealer the day they shipped their stock out to get upgraded)

Apologies if my last post sounded a bit testy. I've been beating my head against this since I got back into town last night and found out what had happened when my roommate captured the tapes.

It's good to hear that a solution is being worked on. And it's definitely refreshing that you guys take the effort to cruise these boards.

Yoochul Chong May 12th, 2006 12:31 AM

i'm seeing the same problem. i have the a version as well. i bought the cameras about a few weeks ago also.

Jim Giberti May 12th, 2006 09:28 AM

Quote:

Originally Posted by Carl Hicks
Ben,

Are you seeing the clip segmentation issue on tapes that were shot after the camera was upgraded to "a" version?

FYI: Our product engineers are looking into this issure as we speak.

Regards, Carl


Carl I've been on location and haven't had much of a chance to check on this, but apparently yes...all our tapes now being captured in 30p with the "A" model are still exhibiting the broken clip syndrome.

Jonathan Ames May 12th, 2006 09:58 AM

We just finished capturing the 6-camera Tunica shoot from late April and not one issue discussed here was found after 18 hours of tape. We're using Adobe's PPro2.0 so is it really a JVC issues or an FCP problem?

Carl Hicks May 12th, 2006 10:25 AM

Quote:

Originally Posted by Jonathan Ames
We just finished capturing the 6-camera Tunica shoot from late April and not one issue discussed here was found after 18 hours of tape. We're using Adobe's PPro2.0 so is it really a JVC issues or an FCP problem?


Jonathan,

I've heard of the issue only with Final Cut. We have engineers now looking at this to isolate and resolve it.

Jim Giberti May 12th, 2006 10:29 AM

Quote:

Originally Posted by Jonathan Ames
We just finished capturing the 6-camera Tunica shoot from late April and not one issue discussed here was found after 18 hours of tape. We're using Adobe's PPro2.0 so is it really a JVC issues or an FCP problem?


Well clearly it's a JVC/FCP issue that JVC identified and thought they had fixed with the "A" upgrade.
This is from the HD100 "A" upgrade site FAQ page:

"Will this address the problem of segmented frames during capture?

Certain NLE's have experienced issues with scenes being broken into shorter segments (clips) during capture. This upgrade addresses this issue"

It was one of the couple of key fixes with the upgrade, and apparently it didn't fix the problem...universally at least.

Ben Brainerd May 12th, 2006 10:40 AM

Quote:

Originally Posted by Jonathan Ames
We just finished capturing the 6-camera Tunica shoot from late April and not one issue discussed here was found after 18 hours of tape. We're using Adobe's PPro2.0 so is it really a JVC issues or an FCP problem?

I see it as a combination of both. JVC has a responsibility to get their camera working with the various NLEs out there, and Apple has a responsibility to make FCP work with all the various cameras. I've sent emails to Apple, and have yet to see a response, whereas JVC (Personified here as Carl) made a response only a few hours after my post. I have more hope of a solution from JVC at this point.

As an update for those who are struggling with this problem, I fired up the HDVxDV trial this morning, and tried transcoding my giant 53 minute clip. After a few false starts, I got an AIC transcode running. Final verdict? It's kinda faster. About 25-30 minutes. Which is better than 55-60 minutes for a 53 minute clip. And for Andrew: The Sync looks ok. It's an extra-wide shot of the entire ceremony, so honestly I can't tell that well. I'm burning it off to check on my projector right now.

So, I guess this raises a question: Do I really want to pay for a piece of software to do something that FCP should do for free? (Well, Ok, the $1200 I already paid) If I had 24p footage, I'd be all over it. But I don't.

Ben Brainerd May 12th, 2006 12:29 PM

Quote:

Originally Posted by Ben Brainerd
AIC doesn't make a difference either.

This is me admitting that I'm a moron. :P

Apparently, in my high-stress state last night, when I captured the tapes using AIC in FCP, I didn't actually bother LOOKING at them. I saw the file breaks, and said "Hey, this didn't work."

However, upon actually viewing the clips, and doing a bit of reading on how AIC handles captures, I find, amazingly enough, that this is exactly how AIC works. It is supposed to break the clips at start/stops. The difference being, AIC actually captures everything. So yes, there are still a bunch of little files, but they're complete little files.

I'm ok with this. It's not optimal, and being able to capture/work natively would be nice, but it's not the end of the world.

Andrew Young May 12th, 2006 07:59 PM

Quote:

Originally Posted by Ben Brainerd
And for Andrew: The Sync looks ok. It's an extra-wide shot of the entire ceremony, so honestly I can't tell that well. I'm burning it off to check on my projector right now.

Thanks for the sync info Ben. Let me know if you see any of your transcoded material drifting.

As for your broken clips, it sounds like they are at start/stops where they're supposed to be, not in the middle of clips as some others are experiencing. Is that right?


All times are GMT -6. The time now is 09:07 PM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2025 The Digital Video Information Network