VHS recording has fringing on left edge?
 

Hello, I recently had my Panasonic AG-1980 VCR repaired and so far everything seems fine. However, I've been trying to capture a VHS of RoboCop and for some reason there's something odd going on on the left edge. It looks like its out of sync or something. I recently purchased another RoboCop tape incase maybe the tape was the issue since the one I was using is kinda old and used but the problem is still there. I tested a VHS of NeverEnding Story on the same VCR and saw no problems. To give somewhat of an idea of what I mean here's some screen shots.

The shot with no issues from a different VCR (no TBC)

Attachment 15254

The shot captured from the AG-1980 (look at the B)

Attachment 15255

Any thoughts???

Looks like a H sync adjustment, I'm not familiar with Panasonic but JVC almost eliminated all those analog adjustments by using system on the chip, If it looks like it has a lot of those video boards crammed with electronic components chances are it has that analog horizontal adjustment, The service manual should answer that question.

I see the same behavior with my AG-1980. Here's what I've experienced.

When I first noticed this kind of thing on certain tapes, it drove me crazy for a while trying to understand it's seeming randomness on which tapes did it and why. I went so far as to send it back to the dealer that refurbished the deck to have this reviewed. Ultimately, what I was told was this is how the deck functions when the TBC switch is on and the head switching noise (the random bar of noise you see along the bottom edge) is more than X pixels high. For me X seems to be around 10 or 12. The height of this is not anything I've been able to change. How much noise you see seems to be a function of what machine made the recording. Perhaps it could be changed by tweaking the machine, but that's beyond my experience, and I wouldn't want to mess with a machine this expensive. It is also more common on multi-gen copies, because the head switching noise from the first generation copy is combined with the second generation's copy to go over the height threshold. Although some otherwise great 1st gen/master recordings can also have high head switching noise.

If you turn the TBC switch off, you should not see this effect. Of course, you loose the line TBC and Digital Noise Reduction (i.e. chroma noise cleaner) when that switch is off. When I work with a video like this in Avisynth and it was captured at 720 x 480, the width of this glitch is 38 to 40 pixels. At times, I've worked around it with some fancy Avisynth-ing to try and undo the effect. However, if it is not constant throughout the entire video, which it usually is not for me, then it is beyond tedious to fix only the frames that "jump" on the edge.

In my case, when the head switching noise is over the threshold, it doesn't always happen. Sometimes it's intermittent and seemingly random. Another quirk I've noticed is that it takes several seconds for this to occur after pressing Play. But it would be insane to try and capture the video 10 seconds at a time to try and get around it. This observation initially made me think it was a clue into why my machine was defective. But I now think it's just another detail of how the TBC functions.

Sometimes it can happen on a random frame of a tape with low head switching noise if you have the video jump up a bit vertically due to a timing glitch, because the switching noise has now moved above that threshold of X pixels high.

Even after I had the machine returned to me with the "high head switching" defense, I still carried some doubt that it was a flaw/limitation. I had never seen anyone else mention this quirky behavior. Then one day I saw a video on youtube which said the capture was from an AG-1980, which showed the same intermittent glitchy behavior. The user did not say anything about the glitchy left edge, but it was as clear as day to me when I saw it.

After a long agonizing journey of trying to find a solution and asking why this very expensive machine with such a legendary reputation would have such an awful flaw as this, I ultimately came to terms with it, and have learned to accept it and work around the limitations by either turning the TBC off (sometimes trying to feed the output through a Panasonic ES10 passthru in lieu of the line TBC) or use another machine for playback. The Virtualdub Camcorder Color Denoise filter (i.e. CCD) is good at removing Chroma Noise. It's a nice backup to use when you cannot use the 1980 TBC/DNR, either because of this glitch or if you just have really jumpy video when it's turned on.

In a very odd way, I feel somewhat comforted to finally hear someone else bring this up. Now that I've seen/heard of 3 instances of AG-1980 doing this, including mine, it reinforces my conclusion that it's just a design flaw/limitation of this model's TBC. I think part of it has to do with the complexity involved with this TBC. I believe it's been said on this forum before that the TBC of this model is different from those on say the JVC models. I believe this TBC does more frame level manipulation that must make some kind of assumptions in it's design that head switching won't go above X rows so that it can try to lock on to the actual video frame before it tries to correct the timing. And the design also seems to rely on this left most edge of the frame to lock on to for it's timing assumptions. When the design assumptions are not met, we see this quirky behavior. Anyway, that's my engineering mind trying to make some basic sense out of it all.

Although it's not what you wanted to hear, I'm glad I'm finally able to say something about my experience and not have it be viewed as a solitary case. As I said before, I am really surprised there's not more said about this, since there's so many of these out there and they get so much attention and praise (that is when they've been refurbished properly). You'd think it might make them not quite so sought after when you find out that it cannot handle "high head switching" noise recordings. For what you pay, you'd expect it to handle anything you throw at it.

Thanks for bringing this topic up!

40 pixels shift by the line TBC seems a lot, My JVC machine shifts by about 3-4 pixels when the line TBC is switched ON. If I was the OP I would go through the service manual and look for some spec to how much shift is normal.

Sadly, the service manual is extremely worthless for AG-1980 decks.

You were told BS. For multiple reasons, most commonly bad caps still on the boards.

Yes and no, mostly no.

I've mentioned this several times, and keep getting ignored by 1980 refurb'ers. Inside the maze of board are adjustable pots, and labeled such that I believe these control the image output from the board. I pulled a deck apart earlier this year, and took photos. (I'll reply again later, share those here.) I'm tired of asking and discussing this, and will address it myself in coming months.

It's gotten to the point where I no longer care. These decks are money pits, and I now have several that have been rejected by known refurb'ers. I'm now casting a wider net, and having those (possibly daisy chain sent) to some other parties. With decks now breaking, and hard, what other choice do we have? Break them more? Oh no, now it won't work after not working! :laugh:

And thus completely removing any need to get these decks.

Nope. I used 1980 decks 15+ years ago, and these did not act that way at the time. This is recent behavior. I think the 1980 scene has some % of revisionism going on, because both the users and the repairers/refurb'ers are new to these decks (relatively speaking). I some cases, I know this to be accurate.

Yes, but you're drawing the wrong conclusion. The fact that two units did it simply proves it had identical or similar issues -- not that it's some sort of model-wide defect.

That sucks.

Yep. :congrats:

It's a field TBC, aka multi-line, aka "infinite window". Not frame, not mere single line.

I know I have, many times, but mostly in private conversations. Though I try to discuss such things in public, for others. These decks are a PITA.

Again, not a rule.

Yes. :pullhair:

That's unacceptably huge. I see no such think on my decks.

Again, the service manual sucks. No mention of this.

If the head switch affects how much area shown on the left side of the frame then there is an adjustment for the head switch, According to the service manual on Section 3-1 (Electrical adjustments procedures), It's done as follows, you put the oscilloscope CH1 on the video out, CH2 on test point TP2001 (Head switch) and you adjust the pot VR2001 as shown in the illustration, But a test tape VFM8080HQFP is required to generate the signal needed for adjustment.

http://www.digitalfaq.com/forum/atta...1&d=1654162892

On VHS decks you can usually adjust that relatively well just by eye rather than needing a scope (different story on some beta and possibly vhs decks where it's set with 2 trimpots). The switching point will also vary quite a bit depending on the recording on the tape being played back, so the exact adjustment in the service manual is mainly to make recordings be exactly at factory spec. Putting it too far down can cause the vertical sync signal to be distorted giving tbc/capture devices issues so can't go entirely overboard with it.

The adjustment pot is accessible from the top of the main circuit board by just taking off the cover on most panasonics as far as I know, so you won't need to disassemble anything to see if adjusting it has any effect.

I agree that it seems odd for it to be caused by head switch position though. It's normal for a TBC to blank a few pixels on the edges (more noticeable on PAL than NTSC), but that seems a bit excessive.

Does anyone have a pic of where this switch is?? I am by no means someone with any skills of tackling the inner workings of any machine let alone a VCR.

-- merged --

So I did a test with the TBC turned off and I guess this confirms the suspicions of it being related to the TBC. As you can see the glitch isnt here and yes this is the same tape that the glitched screenshot came from. Well this sucks cause as Lordsmurf mentioned the TBC is more or less the whole practical reason to use this unit.

Attachment 15261

I've seen that on my AG-1980P as well, but only in preview mode so I assumed it was a software glitch in the VC500 driver or VirtualDub. But for mass-produced tapes I make it a habit of repacking the tape before I capture (stop, FF all the way to the end, let the VCR rewind automatically) and I've seen that change the number of lines of head switching noise.

I’ll give it a try. Can’t hurt I guess.

-- merged --

Took your advice, tried the copy I just purchased off eBay and so far nothing with the recording is standing out. I did a fast forward through on the recording I made just to see if something popped out and haven’t spotted anything yet. I’ll do a full test watch and report back.

Looking closer, it looks to me like even the "good" screenshots (no TBC/TBC Off) are wrong, because they are both missing the right edge of the "O" before the "B". Does repacking fix that error also?

Also, the pixels on the left edge of the front face of the "B" seem to be a slightly different shade than the rest of the "B", and I don't see that in the following "O" or the "C", at least not to the same extent. Maybe somehow that fooled both VCRs, or is an artifact caused by the same problem.

This is why it’s a little hard to explain. This is what I was referring to when I said it looks like the left edge is out of sync. What you’re seeing with the O appears to be from a previous frame where that edge of the O was visible.

You just need to be aware that "repacking" can be very bad. Some tapes are "one and done", too degraded for much handling. So you'll waste your one chance at getting a quality capture by doing a FF/REW, and the tape will be vastly degraded when you want to actually capture it now. I've seen this so many times in the past 5 years.

Obviously not a concern with mass produced easy-to-find retail tapes. Just get a new copy.

This needed to be mentioned. I don't want to see somebody destroy their one good video of (for example) grandma at a birthday party, by following "repack" advice. Capture first. Then feel free to repack and capture again.

In most cases, "repack" has zero positive effect on capture quality.

Oh I agree. If anything I’m more worried about the effect “repacking” will do to the machine then the tape since most of the tapes I’m using I can replace anyway. I’ll definitely just do tests as is and when the problem surfaces decide if I should “repack” or not. I’m also in the process of finding a good VCR/DVD combo with a dubbing feature to supplement this unit while I make my recordings. Keeping this unit going is such a PITA that I’m convinced even with all the money I’ve spent to repair it will not last long enough to get everything done that I want done. That’s just life unfortunately but I try to have proper backup plans as well.

Would you mind if I ask who repairs your VCR's? I am looking for good reliable sources on people that repair these machines. Thanks.

He is a gentleman named Stefan in Tom's River NJ. His ebay account is sionescu1995.

-- merged --

Anyways finished watching my recording. Nothing major that I noticed. Just out of curiosity I decided to use this method with the old tape that I initially spotted the problem with and at first there was no problem until I decided stupidly to stop recording a few minutes in and start over and the problem came back. Took the tape out, turned the machine off for a few minutes, turned it back on, fast forward the tape one last time to the end and back, made another recording which I’m testing right now. So far I see no issues but as Lordsmurf said don’t try this unless you really have to and if you do do it make the recording count!

This implies bad caps still in unit.

Well I’ll try to make do for now. It’s a frustrating unit but at least this problem doesn’t seem to occur with every tape.