Light Gray Border with SignVideo PA-100/Macrovision
 

Copy-protected VHS tapes capture with a 20-pixel high gray bar along the bottom border of images. It's easiest to see in large dark areas (two samples below). The gray bar gradually dissipates during play until 45 minutes or so, when they are too dim to see. Noise reduction has no effect. In smaller, brighter scenes the effect is more difficult to see but it's still there. In the caps below I've brightened the darks so the affected area is easier to see on PC monitors; it's more obvious on TV.

Attachment 2118

Attachment 2119

I know there are other noise problems in the images, but cleanup is a waste of time -- the gray area is persistent (the right-hand gray border in the "Home Video" image isn't on the rest of the video). This occurs on all copy-protected VHS, regardless of player, capture card, or OS. It doesn't happen on home-made tapes, cable recordings, or non-protected tapes.

Is there some internal adjustment in the PA-100 to prevent this? I'm forced to adjust levels, color, etc., with VirtualDub's software capture filters during capture, but it's not as clean or effective as hardware adjustment with a proc amp.

I should add:

My usual circuit for VHS capture is:
VCR > DVD recorder (TBC pass-thru) > PA-100 > AVT-8710 TBC > ATI 9600XT > VirtualDub > YUY2 AVI

The images posted were made with this circuit:
VCR > PA-100 > ATI 9600XT > VirtualDub > YUY2 AVI

I also use an ATI AIW 7500 on an identical PC. Same problem. It doesn't matter where I place the components in the capture setup: remove the PA-100 and the gray bar is gone. OS is WinXP Pro.

Interesting. I'll have to test out my PA-100 with a Macrovision tape and see if I get the same results.

For some reason it doesn't happen on all protected tapes, but on most of them. I realize there's more than one form of copy protection, so this might explain the variation.

It doesn't occur with my old BVP-4 proc amp, but the BVP-4 has another defect so I haven't used it for several years. Output from the PA-100 looks cleaner anyway.

Yeah could be I suppose -- when I get a chance I'll run a few tests, although I don't have a very large collection of copyright protected tapes.

Yeah I had an issue with my BVP-4 as well and sold it several years ago -- hopefully you don't have mine haha :) I would give the video a strange sort of smeared shadow look:

http://forum.videohelp.com/images/gu...82084/bvp2.jpg

I emailed someone about it that worked for them and he gave me a big long explanation of why that happens on some units, I should see if I can dig it up.

There's an internal rheostat you can adjust in the BVP-4. Works sometimes, but often not. Adjusts a gray, noisy border on the left, but often just makes the left border thinner and creates a mirror image border on the right(!). You end up with two borders and all you can do is move them side to side. Ridiculous.

The PA-100 has two jumper clips inside. I don't know what they're for, so I refuse to touch 'em. Have too much $$$ tied up in this thing, and it's great when it works. I don't know why I've seen no other posts anywhere about such a problem -- though I have to say, sometimes you just don't see that bottom border, it's a little vague. On a TV or PC monitor with good detail in dark areas, you can see it.

Now -- I assume what you have shown here is the full frame and with nothing cropped at the bottom?

The reason I ask is because -- are you sure it's not simply the part of the screen in the overscan that the PA-100 (nor PA-200) does not fully cover? As I recall, much like the BVP-4, which doesn't always cover the right side of the frame (as I remember), this device doesn't cover the entire frame length of the frame vertically. Or is this gray line ABOVE that portion in the overscan?

If that's the case, I certainly notice it too, especially on tapes with black frames where I have changed the black level significantly and you get more contrast between the treated and untreated part of the picture. At least that is what I've always assumed -- that it's the untreated part of the picture on the last few lines and not simply some gray bar. I have tons of examples.

Is it present if you just do pass-through?

Of course, since it's all in the overscan, I don't worry about it, but I do all my watching on TV or in VLC on a computer and just mask the frame in the player to hide it on playback.

Correct.

The PA-100 covers the entire frame on un-protected tapes, with no discolored areas. The capture card and the capture software (VirtualDub) don't have overscan areas. They display and capture 100% of the incoming signal, including the a/v timing signal that you can see at the very bottom of the frame. I think you can see from the top image I posted that there is valid image data in the gray area, not just gray noise.

Different players do different things to frame borders. This same tape plays on one of my VCR's with the bottom 8 pixels completely distorted (but not on home-made tapes). You can't attribute this to overscan anyway; overscan includes all 4 borders, not just the bottom. If you play an unprotected tape or a home-made digital video (DVD) thru the PA-100, there's no gray area. I have a copy of this same movie recorded to DVD over cable; I captured it from disc via the PA-100 to AVI to reprocess it and remove some cable transmission noise. That capture has no gray bars anywhere, except 4 pixels of a/v sync signal in black at the top of the image.

Whether I'm using other components or not, the gray bar disappears as soon as I remove the PA-100 or push in the "Unity" button (which bypasses all internal processing in the PA-100, effectively turning the unit "off" and making it useless). My old BVP-4 had no gray bar at the bottom, but it had too many other problems and was unreliable and inconsistent.

Neither the capture card nor the capture software (VirtualDub) have an overscan area. They display and capture 100% of the incoming signal. When you say "all my watching", are you referring to tape? DVD has no overscan. VHS tape doesn't actually have an "overscan" area, either, but many VCRs do place various disturbances in a portion of what most people think of as the overscan, and some even shrink the image to make room for that junk, and others have the image off-center, still others have grunge (wrinkles) on all 4 borders. The VCR I use most of the time has no border noise except 4 pixels of blinking signal junk at the bottom of the frame on all tapes. If there is any unusual border noise, it's coming from the tape or another VCR, not this one. I have an old rebuilt SONY VCR that produces a similar image, but with 4 pixels of grayish junk on the right border. 4 pixels on the top and/or bottom I can tolerate, as there's no data there anyway. In processing it gets covered with 4 pixels of RGB 16 black and the image is centered. But I won't work with 20 pixels of discolored video. So I don't use the PA-100 when it responds to Macrovision in this way.

You shouldn't have 10% overscan around all 4 sides of anything you watch on TV or in media players. If you do, something's wrong somewhere. My VHS captures are 640x480 and rendered full-image to 720x480 MPEG, with only 2 to 4 pixels cropped and filled with black border to cover the 4 pixels of sync noise. Some of the tapes play with an 8-pixel black border on the left, but there's no image data missing or covered with gray bars.

Here are the same two scenes posted earlier, from a different capture using the same setup as before but with VirtualDub's capture filters instead of the proc amp's -- but without the PA-100. There are 8 pixels of black at the right (because the original film's aspect ratio is 1.3777:1, not 1:333), and 4 pixels of sync noise at the bottom. I expanded low-end brightness to be similar to the earlier pics. No de-noising yet or any other processing.

No gray bar anywhere.

http://www.digitalfaq.com/forum/atta...1&d=1323142273

http://www.digitalfaq.com/forum/atta...1&d=1323142273

What I meant to say was that it was my understanding that the PA-100, like the BVP-4, does not apply it's correction over the entire frame on any tape, leaving a small area on the bottom of the frame that is not treated by the device, resulting in the discoloration that is visible. I definitely see what you mean though on the comparison. I'll have to run some captures on mine and see how it performs -- I actually don't have a PA-100, but the Studio 1 PA-200, although I assume it's all the same electronics.

I know what you're saying, but consider this: Would you pay $95 for a Pinnacle USB copy adapter with horrible proc amp controls that cover the entire image, or $205 for an AVT-8710 TBC with somewhat decent basic proc amp controls that cover the entire image, or would you rather pay $449 to $649 for a SignVideo studio-grade proc amp that discolors the bottom 20 pixels of all your movies? IMHO, I think it's some (most?) copy-protected tapes that cause this problem (they also have screwed-up color and odd noise patterns for the first 30 minutes of tape).

Below, three photos with no discernible discoloration at the bottom, except the usual dark side borders added by cable stations and VHS makers to fit 35mmm movie frames into a 4:3 image:

A: A copy-protected tape, without the PA-100, fed thru my BVP-4 (the last time the BVP-4 worked, 5 years ago!). The BVP threw a 16-pixel black border onto the right-hand side and cut 8 pixels off the left side of the image. The next day, the BVP-4 blew a fuse and so did I, ending my association with the BVP-4. But there are no gray bars in the image, and the tape does have Macrovision that required the AVT-8710 in circuit.
http://www.digitalfaq.com/forum/atta...1&d=1323154181

B: Two frames from a non-protected VHS tape, both captured with the PA-100 in circuit. Note black side borders from the retail VHS so that the entire 35mm movie frame fits into a 4:3 box. If there is Macrovision on this tape, it must have faded; I recorded the tape with only the line-level pass-thru (which by itself won't defeat Macrovision if it's there). I used the PA-100 for both pics, but there's no discoloration that I can see.
http://www.digitalfaq.com/forum/atta...1&d=1323154181

http://www.digitalfaq.com/forum/atta...1&d=1323154181

The thing to really remember is that video standards are just that -- standards. Companies came together to form groups, who then decided on how video would work. In several cases, there were competing standards, but standards all the same.

Then came "copy protection".
Or as those of us in the video editing and conversion fields call it: errors and noise.

Garbage is artificially added to the videos, which often causes it to fail 100% compliance with the agreed-upon standards. Yet it tends to play "good enough" in "most" equipment. "Most" is the important keyword. There are certain models of DVD players that can't play some DVDs, and the same was true of videotapes. Signal processors are very strict about input and output, and copy protection has often mangled the signal far too much to be corrected again.

And that's when you see errors like this.

What's truly aggravating is when artificial errors mimic natural errors, and then the overzealous copyright consortiums force device manufacturers to "obey" the anti-copy signals. So what you end up with is a VCR, DVD recorder or capture card that won't play or record your slightly messed-up copy of junior eating ice cream or grandma telling a story.

I've seen far too many of these in the past decade or so.

Very often, it varies from tape to tape. I can play the same movie, and get two different results. Sometimes simply changing the VCR changes how the signal is output (and thus input into the filter device that was having problems). I see this quite a bit with TBCs, proc amps and detailers. There's just no way to avoid it.

While you could blame the device, I'd blame the tape. It has the errors.

Sounds reasonable to me, lordsmurf. I feared for my PA-100, but I think its designers probably know better. Guess I'm stuck with using VirtualDub capture filters (Contrast, Bright, etc.,) or my AVT-8710 buttons. Kinda clunky and start-stop/trial-and-error that way, but better than krud all over the image. At least I feel my pricey proc amp ain't busted (whew!). That PA-100 luminance meter sure made things easier, though.

Thanks lordsmurf and others for giving this a look.

If I were you, I'd use the proc amp. Then open the video in VirtualDub, crop out all of the noise (2 pixels at a time -- NEVER just 1 pixel on interlaced recordings!), and then fill out the frame to 720x480, letting the "good" signal center itself. Feel free to crop some off the top, too, if you want to retain the "safe area" balance a bit. It all looks to be in the overscan, so I would not go to too much effort to "fix" what probably doesn't need much fixing.

Resize filter in VirtualDub:
Attachment 2127

Then resize on this filter.

__________

Your BVP-4 experience is pretty accurate. :)

The BVP-4 is not the easiest device to get along with, but it's powerful enough that you can excuse it. Sort of like dating a Playboy model with a bit of an attitude, if I had to make a colloquial analogy. You learn to cope with small errors for the overall advantages it gives. Or at least most people will.

I've felt your frustration move than once, that's for sure.

I don't know, LS. I could do that, but 20 pixels seems like a lot to chop away. Viewing the tape on HDTV I can see all 20 pixels, and a good part of them on my old Hitachi CRT. I wouldn't have a 4:3 image any more, either. Cropping proportional amounts from all sides, I'd end up with 612x460. Anyway, even if I'd just chop the bottom 20, there's image data down there. That's a lot to lose from Oscar-winning camera work. I can think of two scenes offhand where I'd lose important visual information and a whole bunch of chins and lips. The biggest visual problem with the tape is contrast and slightly but visibly crushed blacks (not to mention a lot of noise that seems to be on the master source); contrast is simple to mildly lower with the AVT-8710. Its controls look very clean, but fiddly. Tint adjustments are usually useless on most VHS, the tint changes with every other scene anyway.

Frankly, I've never worried about overscan because, apparently, broadcasters and VHS and DVD makers don't seem concerned either, except on menus. As you know, most digital TV's today have a default "overscan" but they don't chop off any image, they zoom in on it a bit. Since most consumer TV's do a sloppy job of even minor resizing, I disable that hide-the-pixels feature (and won't buy a TV that doesn't let me do it) and use full-justify display. I don't see any cropping of classic movies on retail video or broadcast, though I've seen some exceptions that are annoyingly easy to spot.

I'll just have to live with those few problem tapes that won't let me use the PA-100 (or the BVP-4, for other problems). I'd rather have the full image -- minus the small bit of sync pulse noise in the border, which doesn't have any image data to begin with. If I was just chopping off "noise", I'd do it; but that's part of the movie down there.

That can be fixed with a script, but there's a few ppl who've asked for my help and I haven't done their scripts for a week now... so if I can ever get to finish that, I can look at yours sometime. The simplest thing is to record with/wo this device and use one copy to line up the other, if that makes sense.

Thanks, jmac698, I know what you mean. I still have the tape, so I made 2 new captures last week. A mild contrast and brightness adjustment with the AVT-8710 (just enough to prevent crushed blacks and blown highlights) was mostly what was needed. It's just a pain without the PA-100's luma meter to jockey back-and-forth between views of histograms in VirtualDub before the setup is right for capture.

All I needed was the first half of the movie. By the second half, Macrovision side effects fade away on the old capture. For many problem scenes the 2 new captures look better anyway. So I'll be patching-in scenes from all 3 copies. That's the way I should have done it in the first place. Only problem with that is that a few scenes have audio sync differences of 2 or 3 frames at scene changes, but I know how to fix that.

Can you believe I've worked on this movie on and off since 2007! I started a thread about the "LILI" tape's color and noise problems in the videohelp forum.

That explains a lot. Yesterday I played the same tape with the same capture setup on another Panasonic VCR. It doesn't have 20 pixels of grayed video. Nope, it has 20 pixels of distorted, twisted, stretched, wiggling, unintelligible noise! Played on an old SONY, the 20 pixels have the image data duplicated for the top 4 pixels and about 4 pixels off to the right of the rest of the image. With or without TBC, etc., the effects look pretty much the same.

Why the other two of my old Panasonic VCR's can play the entire image, I don't know (PV-4662, PV-8662, both circa mid-1990's). The images in this post were from the rebuilt 8662. Wish I still had my old 1994 JVC, model number forgotten now, but parts for it disappeared years ago. No TBC, but transfers made from its years-old tapes look pretty nice -- better, IMHO, than the TBC-equipped models from the later 90's. That JVC must have weighed 30 pounds!

That almost perfectly describes head switching noise.

Look at my menu bleed guide:

Attachment 2133

• the BLACK background is the full 720x480 image (@720x540 for 4:3 design purposes), and accounts for the full signal image palette, and the visible black shown here is pretty much guaranteed to never be seen on any television,
• the WHITE area is the standard viewing area on most televisions (the "TV safe" area), and this one is actually a bit conservative, as quite a few tube TVs and cheaper LCDs/plasmas crop far more,
• the BLUE is the safe bleed area, where you want to bleed an image to cover the small variation of overscan edge from TV set to TV set, so the screen is always filled with valid image, and
• the GREEN is the on-screen guarantee area, which is where any on-screen text or imagery needs to be to guarantee it's not on the edge and/or getting cut off.

The important area to notice for this conversation is the white, blue and black. (Ignore the green.) White is about 25 pixels in (25px), while the bleed range is from 15px to 25px. So most televisions are going to cut off all of this noise, save for the most non-aggressive overscans. That's why it's been suggested to crop/mask and then re-center the remaining "good" image.

Even if the TV does have the overscan turned off, I'd rather mask the noise, and watch what's left -- even if it is being postage-stamped by small fraction. Black borders on all sides is far less obnoxious to watch than a video signal that has ants in the pants. Restoring video is sometimes about hiding problems, not trying to fix everything. Keep that in mind. (Otherwise you'll just drives yourself crazy!)

The PSD is in the attached RAR, and each color is on a separate layer. This is a menu template. :cool:

Mm, no. I don't crop 10% of classic Oscar-callibre camera work, nor do I know anyone who does except hobbyists using inadequate gear and VCR's with playback problems. 10% is not a "small fraction" of the image. 4 pixels at the bottom can go black OK because, you see, there's no image information down there. It's amazing how that 20-pixel border of "head switching noise" disappears when I remove the proc amp.

Thanks for the work on the overscan area. I've been using those safe-area templates for SD and HD from Photoshop since 1997. But I only use them for titles and menus for friends of mine who still have CRT's. Do you really make titles, menu images, etc., using a 720x480 frame? That's unusual, as most titlists would realize that their fonts and drawings will distort on playback in proportion to the display aspect ratio. Photoshop has recommendations for frame sizes (or they used to, anyway) that will fit within the safe area and maintain the correct aspect ratio on display.

I realize that many TVs crop the display and won't you set up for the full image. I don't watch those TVs. I find that doing so is visually annoying. I think you might have bypassed looking at the sample captures made without the problematic proc amp on some protected tapes. Why maim a good image? All I had to do was find another way to capture without the Macrovision side effects. The proc amp is less fiddly, but not if I have to cut up the film directors' work.

While clearly some of the 'noise' here is caused by the proc-amp, don't virtually all VHS and SVHS VCRs have some head switching noise at the bottom of the frame? The smallest I've seen is on GShelly's (on Videohelp) JVC SR-W5U in the screenshots he posted. I've never seen a VCR with none though. Any models that you know of?

I know by default that monitors don't have overscan, but I thought the majority of the TV sets out there (including LCD, plasma) applied some overscan to the image. I actually have a couple of HD CRTs (Sony KD-XBR910 and KD-XBR960) which have a noticeable overscan but it can be dialed down to 0 if need be in the service menus.

I didn't say there was no pulse sync or switching noise in the VCR's I used. The captures posted here show 4 pixels of that noise at the bottom of the frame on all tapes:
http://www.digitalfaq.com/forum/18279-post9.html
http://www.digitalfaq.com/forum/18300-post11.html

My deceased JVC 7600 produced a similar frame. As I said, there's no image down there. I mask it with RGB 12 black in TMPGenc or Avisynth. I don't see 4 pixels as being that much of a problem. But 20 pixels, yes. I have one Panny VCR I don't use any more that has a lot of noise down there, but I don't think it was properly aligned and haven't found a tech who would (or could) check it. My experience with most (but not all) a/v techs is that they're clueless when it comes to PQ.

My old HD CRT bit the dust some time back, but I did adjust for full frame. I did that with my old SD CRT in the service menu, but still lost about 4 pixels anyway -- better than the default, though. I do the same for my SONY LCD and Samsung plasma. By default the latter two used a bit of overscan. The SONY does it on all input sources, the Samsung didn't do it with DVD or BluRay but it cut off a bit with VCR input (I wonder why?).

I belong to a local videophile/classic film club with about a dozen members. All of them raise hell when they see overscan. I recall some of us paid good cash to attend a silent-film festival in Connecticut. The first thing we noticed was that the projector was set to expand the image several inches beyond the size of the screen. One member demanded his money back; the manager refused, had no idea what the guy was talking about, so the member walked out. None of us stayed for more than a couple of films.

What I'm talking about is strictly a personal preference. In my PC repair business I visit many homes and see SD material stretched on HDTV's everywhere. The viewers don't see any difference, and I don't say anything (I have other work to do, anyway, and how they watch TV is none of my business). It's just a personal "thing".

Nothing wrong with that at all -- lots of footage is precious and should be maintained at the highest standard possible. I can understand where you are coming from and your expectations for the premium equipment you use.

On a bit of a side note -- how prevalent was Macrovision in the heyday of the VHS business? I assume it only impacted the major studio films due to the fee of using the protection and the potential for lost revenue for the piracy of the film, was it pretty much all of them? I don't work with a big amount of commercial sources, but I do know for the few I do have, most of the non-independent films have the protection scheme.

Macrovision was not the only anti-copy system in use. There were a number of nasty signal injections that were known to massively screw up how tapes played, even with a basic VCR > TV playback. Disney tapes and Lucasfilm tapes were among the worst. The Young Indiana Jones Chronicles (movie version VHS releases, circa mid/late 1990s) stands out as a great example of tapes that would actually get worse when passed through a TBC. They played with errors directly to a TV.

--

As a photographer, the preset size of film is already limiting, and I almost always crop images to the perfect size. Frames are built for my images, I don't shoot images for frames. I have frames custom built, I don't use ready-make cheap frames from Hobby Lobby.

Videographers are forced to work within a preset space, too. But unfortunately, they don't really have the luxury of cropping. I would not 100% assume the crop is what a director necessarily wanted. Sometimes it's just "close enough" to convey the story (or art, if that be the case). Cropping a few more pixels is honestly still within that "close enough" territory. Not that it should be done if it can be avoided, but sometimes your hand gets forced.

I often have to crop 4:3 to 16:9, when the tape has catastrophic tracking errors. You work with remaining good signal, to produce the most enjoyable image and audio for the viewer (aka out clients). It's a necessary evil.

I can understand the pursuit of perfection, but sometimes it's just not feasible.

^ Interesting! I was always under the impression that Macrovision solely controlled the market and licensed it out to all of those companies.

Eventually, yes, due to mergers and acquisitions (M&A).

To quickly swipe a Wikipedia quote:

Companies such as Macrovision and Dwight Cavendish provided schemes to videotape publishers making copies unusable if they were created with a normal VCR. All major videotape duplicators licensed Macrovision or similar technologies to copy protect video cassettes for their clients or themselves. Starting in 1985 with the video release of "The Cotton Club", Macrovision has licensed to publishers a technology that exploits the automatic gain control feature of VCRs by adding pulses to the vertical blanking sync signal.[6] These pulses do not affect the image a consumer sees on his TV, but do confuse the recording-level circuitry of consumer VCRs. This technology, which is aided by U.S. legislation mandating the presence of automatic gain-control circuitry in VCRs, is said to "plug the analog hole" and make VCR-to-VCR copies impossible, although an inexpensive circuit is widely available that will defeat the protection by removing the pulses. Macrovision has patented methods of defeating copy prevention, giving it a more straightforward basis to shut down manufacture of any device that descrambles it than often exists in the DRM world.

Note the keyword: major. A number of operations have always wanted to do things their own way, so they opted for small lesser-known methods, or developed in-house methods. I sometimes think Macrovision was altered or added to other methods, to create some of the messes I've observed through the years. (And I'd bet it wasn't within the user agreement, if it did happen.)

Most dedicated fans will buy a newer version, if available and unbutchered (no "improvements" or content edits). The best method for studios/producers to protect their income was to just provide quality releases for fair prices, as demanded by their customer base. (For the most part, that's happened, though it took a while. Forced copy "protection" was never so required as to violate the specs of how VCRs produce good images. You can mostly thank the now deceased Jack Valenti for that one.) Sadly, a few odds and ends slipped through the cracks, so we have threads like this. "Help me preserve my favorite movie that's not available in any other format." A reasonable request, I'd say.

Very informative. I'm not surprised that copy protection has many forms. I have a couple of tapes that uniformly display at least one variation: the disturbance at the bottom of the image gradually fades, although it takes about 45 minutes -- halfway thru the tape -- to disappear. At the same time, color and noise gradually improve as well. At first there's a lot of color noise, low saturation, slow but visible color tone shifts. On other tapes, there's a slow but detectable variation in brightness; this also disappears about halfway through the tapes.

Many of these effects are seen in normal playback, but the odd problems with frame borders appear only when I use a proc amp. Sign Video grays out the the bottom 20 pixels, but the BVP-4 does weird things with left and right borders, placing a stripe one one or on both, or distorting the right border video. Why these effects are there with a proc amp, I have no idea. My AVT-8710 has basic proc amp controls, certainly not as sophisticated as the PA-100 or BVP-4, but the AVT-8710 doesn't display those border effects. Other effects on color and brightness are always there, but they look a bit worse without the line-TBC or frame-TBC.

Whatever the cause of the effects with proc amps, it's beyond my expertise. But at least I found a workaround: remove the proc amp and use something nearly as effective but more fiddly and inconvenient. I rely heavily on the PA-100 luma meter; without it, I have to use a VirtualDub histogram and jockey back and forth between viewing settings and contrast/brightness controls, etc., and color balancing on bad video is just more bothersome. But, as long as the workaround works . . .that's what I'll have to tolerate.

In the case of the BVP-4, it's really nothing more than an engineering flaw. It varies highly between the models. One of the BVP's here has almost zero in-image bordering. Others, however, can eat up almost half of the "safely off screen" overscan area.

I've never been comfortable referring to the AVT-8710's color controls as being anything related to a "proc amp". But I lack a good description. The video is not passed through any degree of complex circuits, so these are very basic adjustments.

While VirtualDub is good (the ColorMill 2 plugin especially), you'd probably like Adobe Premiere Pro CS4 or CS5 much better. The color controls are quite advanced, Photoshop-like (not surprising considering both are Adobe products, and the software has lots of crosstalk of features / interoperability between programs).

Give that a try sometime, if you have access to Premiere Pro. :thumb:

I don't expect much from the AVT-8710 image controls. Bright/Contrast is all I attempt to control there.

Unfortunately Premiere is unsuitable for the way I capture. I see no reason to tolerate design mistakes like 64-bit Vista/Win7. My captures are to AVI/YUY2, most of the initial luma and some very basic color correction are done with AviSynth, usually to avoid crushed blacks and blown out brights. A lot of tape damage gets fixed in YUV or YCbCr first, then I move to RGB with VirtualDub and After Effects CS3 for most of the color work. In fact AE is difficult to learn, but I'm finding it essential for correcting many VHS damage problems. I'm not knocking Premiere. It's great stuff. But with VHS, a lot has to be fixed before it gets to Premiere or AE.

I actually ran into an early copy protection system on a Disney Betamax movie, I think it was called CopyGuard. Most capture cards don't pick it up, but it did something funky with the vertical blanking interval. My Panasonic plasma TV showed the video with the VBI in the center of the screen (like the vertical hold on a CRT was stuck), so I saw the bottom half of the frame, black strip (VBI) and then the top half of the frame. Running that video though a TBC completely fixed the problem. Another cheap trick was to turn on the digital effects on my SL-HF860D deck, running any video though its frame buffer completely regenerated video sync like a TBC.

I also had some interesting problems with Macrovision VHS tapes that were dubbed to Betamax. Betamax was pretty immune to Macrovision due to how its automatic gain control worked, so dubbing was possible. What I found out is that there was still residual traces of Macrovision on those tapes (1990s Disney, go figure), it caused my AVT-8710 to jack up the contrast and/or brightness and completely blow out the whites in the video.

Amazing, the number of ways engineers can work Frankenstein effects on media. With so many hobbyists working around these tricks, the industry has finally hit on new ways to prevent private use: discontinue the input/output connections on incoming and playing components (cable boxes, TV, etc.), stop making anything that can make copies of anything, and screw up a/v with weird tricks like using Ethernet twisted-pair wire to make something that they misnamed "HDMI". It's a wonder anything transmits coherent images or audio, whether analog or digital. Back in the 1920's and 1930's engineers figured out how to screw up analog tv with a crippled NTSC that never improved very much for nearly a century. They replace analog chaos with digital logic and discover they created their own nightmare (as usual).

Maybe these guys oughtta read the original Frankenstein. There are reasons why they say the book was better than the movie.

Yes, I recall CopyGuard now. Wow. Takes me back.

The problem of 20 pixels of grayed area along the bottom of the captures with the PA-100 was solved by JVRaines, here:
http://www.digitalfaq.com/forum/vide...html#post44422

The trimmer adjustment was extremely small, and cleared the problem.
:)

Worked great for me too. The adjustment was a very, very light turn with the screwdriver. Took me a little bit of trial and error to hit the sweet spot.

In my case, there was a slight gap between the bottom of the underscan area and the color corrected area, so masking it was losing a little more than I wanted to.