Baffling Random capture problem

David Newman · March 8th, 2009, 10:20 AM

Yes, also try another FW cable. :) It can happen.

Adam Gold · March 8th, 2009, 10:28 AM

Yeah, hope springs eternal, so I swapped out the cable last night. No joy.

Bob Hart · March 8th, 2009, 12:10 PM

Adam.

I think you may find that the cameras are not entirely innocent in this. If you play the m2t files back in VLC player, you will likely find the capture is corrupted at the same places and that HDlink is faithfully reproducing those defects without adding any or very few of its own.

I have an FX1 which does what you experienced about 10 to 20 minutes into a long capture like an event or music gig. You may also observe with in-camera playback of the FX1 vision, a sort of resync or reset jump after about 12 minutes of a long take, sometimes later, sometimes not at all. This may be a tape dropout. I have never found the cause.

A Z1P has not done this to me at all.

A JVC GY-HD100 randomly captured a 35 minute single music take then flaked out on every other capture I attempted. Despite my doing all the right things, power down everything before touching firewire plug to socket, the JVC's firewire port had gone next time the Firestore was connected.

I routinely capture first to m2t file in HDLink, then convert in a second step.

Dan Kline · March 8th, 2009, 12:35 PM

Quote:

Originally Posted by Adam Gold

Those are all good ideas. I'll try to figure out how to make them work. As I noted earlier, this has happened with both the integrated FW and a separate card I bought and put in my one remaining slot.

Certainly possible it's the RAID card; maybe I could remove and then use mobo RAID control -- wondering if this would mean wiping out RAID data already stored. If so, not an option.

Forgot to mention that this has happened with two video cards as well -- NVidia Quadro FX 1500 before, and now ATI Radeon 4350.

As the glitch isn't visible during capture, only after when you scrub through the file, watching the performance monitor for an hour and writing down any suspicious activity and then checking the captured file will present a challenge -- but it's certainly a good idea.

Thanks for the help.

As it's clear I was barking up the wrong tree by posting in the Cineform forum, I've started a new thread here:

http://www.dvinfo.net/conf/high-defi...s-capture.html

Sorry to effectively double/cross-post -- I hate people who do that -- but I wanted to try to get this in front of a larger audience who might not check out th Cineform forum regularly.

You can setup the perfmon tool to log all the diagnostic data to a "history" log file while you're doing the capture and then you can go back (later when you've identified the glitch in the capture file) and analyze the logs using perfmon tool. The method of doing this is different between XP and Vista but they both should acquire all the data you need. Just make sure you write down your start time and whatnot so you have a way to correlate your glitch time offset in the video file with the timestamps in the log data captured by perfmon. I imagine there are plenty of websites that describe how to set this up (XP or Vista). I think it's worth while to learn how to do (at the very least to add to your tool arsenal for future issues).

Adam Gold · March 8th, 2009, 02:05 PM

Thanks for both ideas above. I'll give them a shot.

Bob, you may well be right. All I can tell you is as I noted, this has occurred with three different playback devices, the tape itself is fine as far as I can see, and the problem doesn't go away when I capture the m2ts in HDLink and then convert. And eventually it captures fine if I try enough times. But I have other cams so I can switch some out and see if if makes a difference over time.

As capturing in HDLink, either as m2ts and then converting, or as CFHD-AVI, provides no benefit, I've gone back to capturing from within Premiere. The last two tapes I tried came in fine, even though they were re-captures of previously corrupted files.

Adam Gold · March 20th, 2009, 05:40 PM

I think Bob (and David, in his very first reply, when he implicated the capture device) was on to something, although perhaps not in the way he intended.

Even though this problem has occurred on three different capture devices, I opted to try a fourth, my HD1000, as it doesn't get much use and these are all continuous, single pass captures anyway.

So far, and I emphasize that this is only preliminary, I've captured the last 11 tapes without a problem using this device. All with the same FW cable, port, settings, PC as the original setup when I first posted.

So maybe it had something to do with the M15, although for the life of me I can't imagine what the problem with two different HC3s would have been, especially as internally they are virtually identical to the HD1000.

Using the cam as a deck also solves another minor annoyance I've experienced with the M15, which is that device control isn't great -- it FFs and REWs so fast that it frequently way overshoots the mark when batch capturing and never recovers, leading to small captures that are a minute or more off (and when the clip is only 20 secs, that means you don't get any of what you want).

I wonder if somehow they are related.

Anyway, pretty clearly not a Cineform issue but thought I'd post the progress here as this is where most of the brainstorming occurred.

March 8th, 2009, 10:20 AM	#16
David Newman CTO, CineForm Inc. Join Date: Jul 2003 Location: Cardiff-by-the-Sea, California Posts: 8,095	Yes, also try another FW cable. :) It can happen. __________________ David Newman -- web: www.gopro.com blog: cineform.blogspot.com -- twitter: twitter.com/David_Newman

March 8th, 2009, 12:10 PM	#18
Bob Hart Inner Circle Join Date: Dec 2003 Location: PERTH. W.A. AUSTRALIA. Posts: 4,476	Adam. I think you may find that the cameras are not entirely innocent in this. If you play the m2t files back in VLC player, you will likely find the capture is corrupted at the same places and that HDlink is faithfully reproducing those defects without adding any or very few of its own. I have an FX1 which does what you experienced about 10 to 20 minutes into a long capture like an event or music gig. You may also observe with in-camera playback of the FX1 vision, a sort of resync or reset jump after about 12 minutes of a long take, sometimes later, sometimes not at all. This may be a tape dropout. I have never found the cause. A Z1P has not done this to me at all. A JVC GY-HD100 randomly captured a 35 minute single music take then flaked out on every other capture I attempted. Despite my doing all the right things, power down everything before touching firewire plug to socket, the JVC's firewire port had gone next time the Firestore was connected. I routinely capture first to m2t file in HDLink, then convert in a second step. Last edited by Bob Hart; March 8th, 2009 at 12:14 PM. Reason: error

March 8th, 2009, 10:28 AM	#17
Adam Gold Inner Circle Join Date: Jan 2007 Location: Woodinville, WA USA Posts: 3,467	Yeah, hope springs eternal, so I swapped out the cable last night. No joy.

March 8th, 2009, 02:05 PM	#20
Adam Gold Inner Circle Join Date: Jan 2007 Location: Woodinville, WA USA Posts: 3,467	Thanks for both ideas above. I'll give them a shot. Bob, you may well be right. All I can tell you is as I noted, this has occurred with three different playback devices, the tape itself is fine as far as I can see, and the problem doesn't go away when I capture the m2ts in HDLink and then convert. And eventually it captures fine if I try enough times. But I have other cams so I can switch some out and see if if makes a difference over time. As capturing in HDLink, either as m2ts and then converting, or as CFHD-AVI, provides no benefit, I've gone back to capturing from within Premiere. The last two tapes I tried came in fine, even though they were re-captures of previously corrupted files.

March 20th, 2009, 05:40 PM	#21
Adam Gold Inner Circle Join Date: Jan 2007 Location: Woodinville, WA USA Posts: 3,467	I think Bob (and David, in his very first reply, when he implicated the capture device) was on to something, although perhaps not in the way he intended. Even though this problem has occurred on three different capture devices, I opted to try a fourth, my HD1000, as it doesn't get much use and these are all continuous, single pass captures anyway. So far, and I emphasize that this is only preliminary, I've captured the last 11 tapes without a problem using this device. All with the same FW cable, port, settings, PC as the original setup when I first posted. So maybe it had something to do with the M15, although for the life of me I can't imagine what the problem with two different HC3s would have been, especially as internally they are virtually identical to the HD1000. Using the cam as a deck also solves another minor annoyance I've experienced with the M15, which is that device control isn't great -- it FFs and REWs so fast that it frequently way overshoots the mark when batch capturing and never recovers, leading to small captures that are a minute or more off (and when the clip is only 20 secs, that means you don't get any of what you want). I wonder if somehow they are related. Anyway, pretty clearly not a Cineform issue but thought I'd post the progress here as this is where most of the brainstorming occurred.