I assume that you're talking about the "squiggly" few lines at the bottom of the frame. It can't be disabled, it's always been there, you just never saw it because of the overscan on televisions. You just didn't become aware of it until you saw the full frame displayed on a computer monitor. You can mask it out by placing a black bar at the bottom of the frame to cover it up, or you could zoom in on the image slightly (but this is a less-than-ideal solution.)

And they're not "tracking bars," the correct term is "head switching noise."

John is correct. I too vote for masking because it avoids any degradation and people are used to small black bars and the top/bottom of the picture. Also, masking takes virtually no additional render time, whereas zooming is very resource intensive, especially since you should always use the "best" setting in the Render As dialog whenever re-scaling video, and this is a very slow setting.

Thanks for the answers. I am not sure I explained myself all that well. The issue was that the Samsung VHS player generated graphics on the screen that show that the device is automatically changing the tracking. Anyway, I pulled out an older VHS player that does not automatically change the tracking and that seemed to solve my issue.

Oh, well, that's very different! The Samsung probably has a way to defeat the display of the auto-tracking information buried in a menu somewhere. (Auto-tracking is generally a good thing.) Nevertheless, I'm glad to hear you got it worked out.

I wouldn't count on it. Most non-pro VCRs will display what they want to display on the screen no matter what and the displays can't be disabled.

I had a VCR for a while that, when you hit the play button, would show "PLAY" large enough to cover about 1/3 of the screen for 5 seconds. Then a smaller "Play" would appear in the corner for about 5 seconds after that. Then the triangle > icon would show up on the screen for another 10 seconds. Why??????? It was easy to see the word "Play" on the display on the front of the VCR. For that matter, even if the VCR wasn't visible, we could clearly see the movie playing on the screen.

Just makes me wonder what was going through the design teams' heads (if they had heads). If we didn't see the Play indication on the screen were we going to panic 10 seconds into the movie thinking we had accidentally hit Stop instead?

But, back to the original question, both of my remaining functional VCRs do the same thing, informing anyone watching that it is "auto-tracking" in big bold white letters across the bottom of the screen. Both of them also say that on the displays on the VCRs as well, so the on screen display has to be nothing more than an ego boost for the manufacturer proclaiming "aren't you glad you paid an extra $30 for a model with auto tracking? see how well it's working for you?" Oy. I turn off the auto tracking, set it manually for the first part of the tape, and capture for as long as i can until the tracking becomes unusuable. Then i'll stop capture, manually adjust the tracking again, rewind back before where the picture went bad, and start capturing again. Time consuming, but worth the effort.

Perhaps your VCR has a "monitor" output and a normal output. My sony SLVR5 has three outputs, one always contains the monitor info which you would connect to your monitor, the "output 1 & 2" are for recording.

If I may piggy-back on this thread, I'd like to ask about the masking technique. A friend has loaned me his Canopus device, and I'm now capturing very old home movie VHS tapes on my hard drive (pretty amazing how different I looked twenty years ago!), and yes, I'm seeing the squiggly noise at the bottom as well.

I had planned on simply using track motion to place my track such that the squiggly line wouldn't be visible, but if I understand the comments here correctly, that's more resource intensive than using a mask.

Could someone explain how to do that? I realize it's a beginner question, but then, I'm very much still a beginner.

Thanks!

I had planned on simply using track motion to place my track such that the squiggly line wouldn't be visible, but if I understand the comments here correctly, that's more resource intensive than using a maskI wouldn't use track motion, since that will mess with the resolution. The whole idea is to not cause any of the pixels in the underlying video to be re-sampled.

If you are going to display the final result on a normal TV set, you probably don't need to do ANYTHING. If you didn't see the overscan before, you definitely will not see it in the final result. Other than perhaps getting a slightly cleaner encode, because the fixed number of bits per frame now don't have to be wasted trying to encode that garbage, I don't think there is any real upside to masking or zooming or anything else.

However, if you watch on a device that has no overscan (like your computer monitor), then the following masking technique will probably be worth doing.

Someone else will have to describe whether the following is the correct approach, but here's how I'd do it:

Create a 655x480 project in your favorite photo editor (this is for NTSC, and square pixel 655x480 is equivalent to 720x480 0.9091 NTSC DVD non-square pixels video). Make the image perfectly white, and then place horizontal 100% black bars at the top and bottom. The width of those bars will determine the size of the mask. Then, create a new track above your video and place this still image on that track. Extend the right edge of the image so that it covers the entire video below. Finally, go to the track header for the mask (NOT the video) and set the "Compositing Mode" to "Multiply."

Here's a link to the mask PNG file I used for this example:

Retrace Mask

Here's what a VHS image looked like before the mask (the retrace "tracking" garbage can be seen best in the lower right):



And here's the same image from the Vegas preview window after application of the mask as described above:



[Edit] In looking at the Before image, I'll admit that it isn't too easy to see the problem area near his left cuff (his left, your right), because the image above is so small. However, it is there. The small black cropping bars in the second image should be readily apparent.

I just realized, after right-clicking on the above image and downloading the underlying file, that Photobucket had halved the resolution both vertically and horizontally. So, I uploaded that file to YouSendIt, and you can download it so you can see the original scan lines, that is if anyone really cares. Probably not necessary ...

Original File

You can just use the Border FX built into Vegas. That'll possibly mask more on the top and the RH side than needed but it's quick and easy.

Bob.

OK, I just did a performance test. I put a levels fX on the original video (which was captured from VHS tape and saved as a standard NTSC DV AVI file). I rendered five seconds of that. I rendered twice because the second render is often a little faster for short clips because the second render is done from video that has been cached to RAM memory. That second render took 9 seconds.

I then rendered using the approach I described in my last post. This one took 11 seconds. Finally, I tried the border approach that Bob suggested. This one took 12 seconds and ended up with borders on the sides as well as top & bottom, which I don't think I would want.

The best approach I've found is 4 tracks of gen media with track motion to line them up. That way I can judge exactky where I want them. I've found you need to playback the footage to really judge accurately where you want to position them to mask any jitters and noisy edges.
Just using the Border isn't ideal however on a CRT they should be outside picture essential and on a LCD or PC playback they should end up being symetrical. That's one issue I've found that's a bit beyond what we're talking about here. Between different playback devices the centre of the frame shifts. Normally such as small difference as to not be a problem but at others a real PIA if you're trying to get an even border.

Bob.

Wow, thanks for the input. I definitely have some things to try now.

Reading these also got me to thinking (when you mentioned the television). My original is 4:3, but I have a widescreen television. When I've converted our Digital 8 movies to dvd, I've cropped them to 16:9, giving up some sections of the top and bottom to have a widescreen movie, which we generally prefer.

Now I'll have to burn two different dvds to see how much quality difference results when cropping to 16:9, especially since the originals are old VHS, and on top of that, I made the mistake back then of shooting in lower quality to save space (doh, I'll never do that again).

I'd never crop 4:3 to 16:9 no matter how good the original much less something that started out on VHS. Just make a 4:3 DVD, your player / TV should cope perfectly OK. If you really can't cope the with black bars at the sides then the TV will almost certainly have the option to letterbox the footage to 16:9 anyway. This way you've preserved as much as possible from the material and still have the option to view it cropped. Once you crop it to 16:9 the loss is permanent.

Bob.

Bob, thanks for the great points. I can't believe I didn't think through this more clearly, but thankfully I've created only one dvd so far (and I actually created both widescreen and fullscreen versions, because my brother has a 4:3 television).

I don't mind black bars, but for whatever reason it drives my kids nuts, so I thought I'd just crop. Well, stupid me - my television has several different ways of handling 4:3!

Sometimes it definitely helps just hearing (reading) other people's thoughts. Thanks again.