Vertical stripe caused by TBC on Panasonic AG-1980 - why?

[moxiecat]

I've used the Panasonic AG-1980 for years--it's a great machine. But recently I started noticing a problem: when the TBC is engaged, the unit occasionally causes a thick vertical stripe to appear down the left side of the picture. This stripe is cut off as overscan, which is why I didn't see it for a long time with my DVD recorder-TV setup. I finally spotted it when I ran a bunch of VHS-C tapes into my computer. Here's what it looks like:



The stripe appears down the left side of the picture. It appears to be part of the image, but shifted--you can see how the curb doesn't line up in this example. I believe the stripe is a leftover image from the previous frame, because of situations like this:



The above image was taken from the first frame after a jump cut. The stripe is clearly the image from the previous frame. (I have verified this on other jump cuts as well.)

Here are a few facts regarding this error:

- The Panny's line TBC is the culprit. When it is turned off, the stripe is 100% gone.

- I have tested this phenomenon with more than one Panasonic AG-1980. All displayed the problem on the same tapes.

- The stripe only happens with SP tapes. I do not recall seeing it on EP tapes.

- When the camera is still, the stripe is gone. It appears most often when the camcorder pans from one side to the other, or whenever the camcorder is moving around quickly (it was hard to find a screenshot that wasn't blurry, for this reason).

- Not all tapes show the problem. I transferred a bunch of tapes recorded by the same camcorder, and about half of them showed the stripe. Some were much worse than others. And some tapes in the same collection didn't display the stripe at all. Then I transferred tapes recorded by a different camcorder, and only spotted the problem on about five frames in 30 tapes (all frames in which the camcorder was bouncing around wildly). So depending on the tape, the error ranges from a stripe that you can easily see to an occasional single-frame error that 99.9% of viewers would never spot.

My conclusion is that the Panny's line TBC is a frame behind within this vertical stripe. This would explain why the error is more apparent on scenes with fast motion or on jump cuts (where the frames are very different from one to the next). But it must be more sensitive on certain tapes as compared with others, and I definitely can't figure out why SP is so much worse than EP.

I am here because I know of no one better to address this problem than digitalFAQ. My questions:

1) What is your assessment regarding why this error is occurring?

2) Should I just turn off the TBC on SP tapes? My inclination is to leave it on (the capture is usually more stable and smoother), but spotting frames with these stripes is very hard to do during capture. It bothers me a lot that such a valuable tool on this good VCR can't be used without an error potentially occurring.

3) Hasn't anyone ever seen this error before? I've caught it with a variety of tapes on a few units, so I can't believe that folks who capture to the computer and do more editing than I do haven't noticed this. But I can't find anything online about it.

Thanks in advance for your insight!

[kpmedia]

Hi moxie.

1. Hardware error.

2. If it's what I think it is, then no. Get the VCR fixed. The big reason to own this VCR (aside from the transport) is the full-field TBC. (If I'm wrong, then there's really nothing you can do. It may be a flaw of the tape confusing the TBC -- and that happens, I've seen it many times in the past 10+ years. I keep one tape specifically for it's ability to "mess with" a TBC, and it has a 100% flawless record at doing so.)

3. Yes, I believe I've seen this before. I need to pull a few maintenance books and flip through the pages. I've luckily never had this issue, though I've had similar problems. I can run down the exact error name for you, which you can use to call around to some repair services to ask for a quote on what it would cost to fix that specific problem.

_______

Your first image mostly shows blurs. The tell-tale image, the one that betrayed the VCR and revealed the problem, is in the second image.

It actually looks VERY similar to the frame buffer problem that's been identified in 2011 (and possibly late 2010?) model AVT-8710 TBCs. The buffer is not releasing when the next good frame comes along -- it stays hung on the bad frame indefinitely. This almost appears to be a delayed release to the TBC.

However, I think it's something else -- something to do with the heads or the transport. I know I've seen an image like that before. Amusingly, I think the sample image was very similar, like a person walking a dog down the street.

Then again, if you're able to repeat this flaw on several AG-1980 units, then you either (1) have the same flaws on several VCRs, or (2) it's an inherent weakness of the 1980 TBC when presented with an error as found on these specific tapes.

One of my AG-1980P VCRs is completely messed up right now. The front panel LED is about gone, which isn't really important. It emits buzzing into the audio, which is a problem -- something that started after the fuse was replaced. And most problematic, the tape starts out looking fine, but chroma playback fades away within 1 minutes, leaving you with a B&W image. I've never seen anything quite like that error.

Now that the local network is re-established (long story), I'll be able to upload the AVT-8710 error samples. I think you'll notice some striking similarities in the bad footage.

[moxiecat]

Quote:

Originally Posted by kpmedia

It actually looks VERY similar to the frame buffer problem that's been identified in 2011 (and possibly late 2010?) model AVT-8710 TBCs. The buffer is not releasing when the next good frame comes along -- it stays hung on the bad frame indefinitely. This almost appears to be a delayed release to the TBC. . . . Then again, if you're able to repeat this flaw on several AG-1980 units, then you either (1) have the same flaws on several VCRs, or (2) it's an inherent weakness of the 1980 TBC when presented with an error as found on these specific tapes.

"The buffer is not releasing when the next good frame comes along -- it stays hung on the bad frame indefinitely." That sounds exactly like what is happening, although I would replace "bad frame" with "last frame" (unless there's a lot of bad frames!).

It's definitely not a hardware error with the one deck, because I did make a point of trying it with a few units and could reproduce the problem easily. But I think your (1) and (2) are both right. The model itself has a certain flaw in certain situations, AND some specific tapes trigger the flaw more than others. Most likely because of how the tape itself was recorded, IMO.

It's not a problem when transferring tapes to DVD, both because the flaw lies outside the overscan, and if it doesn't, it's also hard to spot on most tapes when you're just watching them for pleasure and the video keeps rolling along. It's more of a problem with tapes that are transferred to the computer for editing, because the entire picture is visible and you're more likely to see individual frames when you're editing.

One last note...the tapes that were most problematic were the batch that included that VHS-C tape that you worked on a few months ago. So we already know those tapes had various tape transport flaws that had been caused by the camcorder. The tape I used for these screenshots transferred fine, except for this issue. But this particular batch of tapes, and the camcorder that recorded them, may be some of the worst that the AG-1980's TBC ever has to deal with...and so the error may happen much less on most other tapes. (Although like I said, I did catch it briefly on others--always with fast motion shots.)

Any thoughts on the SP/EP thing though? That's one thing I can't really figure out. What makes EP playback so different that it doesn't trigger a frame buffer hang?

Thanks for your help. I know I mentioned this issue ages ago but never got around to getting those screenshots until today!

[kpmedia]

Quote:

replace "bad frame" with "last frame" (unless there's a lot of bad frames!)

I would have to argue this -- I'd say it's a bad frame. And since we're talking about the nasty little creature known as VHS-C, it's very likely to have been an abomination of a tape, filled with errors. The transports of the compact VHS cameras were simply lousy, even in the higher grade S-VHS-C cameras.

However, all of that may be a moot point anyway, as the Panasonic AG-1980P TBC is not a full frame TBC with frame synchonization abilities. It's a full field TBC, which is basically a multi-line TBC. (Just don't mistakenly think it's better than the JVC line TBC, which is a technically weaker TBC. However, most advanced users know it can outperform the Panasonic most of the time.) It may be a relative of the error found on the AVT-8710, but I doubt it's identical.

Quote:

problem with tapes that are transferred to the computer for editing, because the entire picture is visible and you're more likely to see individual frames when you're editing

Just mask it. By default, you should drop a matte on tape conversion work to hide any possible noise that may be found in the overscan. There's always something, and it very often changes exact position over the course of the tape -- especially if the tape was used for multiple recordings, or even multiple cameras/VCRs. It's rare that you need to really get in there and save that last 5% of image (forensic video work).

Quote:

with fast motion shots

You know, it's always possible this is how the tape was recorded. Don't automatically assume the player is always at fault -- very often (most often?) errors are a direct result of the camera/VCR that created a tape, and not the one that's currently playing it back. Assuming the current VCR is good, of course -- which is the case for the Panasonic AG-1980.

Quote:

Thanks for your help.

Always glad to help.

When I find some time to browse those tech manuals, I'll let you know what I come across. I'll see about flipping through some pages tomorrow for you.