Hmm... ok. Thank you for the reply.
Is there any way to fit my video to be within the outer safe zone then? I can live with the possibility of the 10% between the outer and inner zones being lost, but not that plus the 10% outside of the outer zone.
Can i snap my video and DVD menu screens to be within the outer zone?
Thanks.

guess i'll be taking wider shots from now on too... ;)

You can 'bleed' the edges of your image to those outer zones, and they SHOULD fill the entire computer preview screen. It's titles and graphics that you need to keep inside the safe zones, because different televisions will vary in what they show and don't show, due to any number of reasons. DV in and of itself is designed to never show all of the image anyway. A full 5 percent of the image is intended to be unseen.

While every TV is different you can test your own set(s) to see what fits within the safe area.

1. turn on the safe area borders above the preview window.

2. drop the Border filter on the video or image on the timeline. Size border so it just touches the outher safe area border

3. Use the color tool to paint the border a color that's a strong contrast to the test video so it shows up clearly

4. Render the clip and burn either a test CD or just play back from your camera or you can just test on your external monitor

If you see any of the border color on the TV obviously your set shows more of the picture. Likely you won't get a perfect margin all the way around. For example on my lexternal monitor my "safe area" is larger that the preview window safe area by about a 1/4 inch top and bottom, but both the left and right borders on dead on the outer marker and don't extend.

Its really just a quick exercise just to see, the varations should be minor and not really anything you'd want to fit that close to the end with that extra little bit of real estate if yours has it.

>Is there any way to fit my video to be within the outer safe zone then?

Yes. You can use pan/crop to shrink your video as much as you want. For example, you could shrink it by 10% so that your entire picture shows up inside the outer safe area.

Two things will happen: (1) You'll get a drop in quality because of the pixel interpolation. It can be a significant drop in quality, too. (2) Different TVs have different visible regions, so if a TV had only 5% cutoff, you'll have extra black around your picture and will have been "wasting" resolution.

>guess i'll be taking wider shots from now on too... ;)

Yup, this is the key. The things they don't tell you about but you really need to know, eh :-)

I've always been surprised that some TV manufacturer hasn't latched onto this overscan concept by now. Imagine if Sony began marketing TV sets that were "10% sharper" simply because they tightly controlled overscan just like your PC monitor does. There would be some sort of calibration procedure that let the user adjust the image to just reach the edges of the screen, thus reclaiming the 10% that was usually never seen.

I would imagine the new DLP rear-screen displays could do this now, and of course any front-projection device never has any overscan. Or does it?

RMD - You said Pal! - This is a hard lesson I've learnt . . . . I can see WHAT I've filmed within the LCD of my XM2 . . and I can SEE all of that stuff in the Preview . . but, but, but this isn't what I get when I edit and PTT and then put out to VHS tape and onto a TV.

Yes BB, I do get the same cropped version when I Preview via f/w . . . this has been a hard won experiment. I've come to conclusion that I need to film taking into account a 5% wastage around my framing while on shoot . . . and I am not going to P/C to get things into the area - lack of quality and then the obverse of this would be to get a nice black border on "other" scan TVs - hah! . . . . "Yup, this is the key. The things they don't tell you about but you really need to know, eh :-) ". . too right RMD

Grazie

I guess I never tire of explaining this. Practice makes perfect? When I can get it down to a few sentences I'll be happy.

Here goes.

First of all, bear in mind that when video signal specs were developed all of the equipment used tubes. If you ever had a tube TV you can remember waiting for it to warm up before the picture would look right.

Okay, so a video signal is analog-just a waveform. The hardware takes that signal and uses it to paint a line or raster across a TV screen. At regular intervals that signal drops to a very low voltage and the tv returns to the begining of the next line. The signal then rises in voltage an is painted across the next line. This happens over and over until the line ends at the bottom right of the screen and must be reset to the top left corner.

In all cases it takes time to drop that video signal to a low level, refocus the emmiter to the start of the next line, and then raise the voltage back up to signal level. Even with current technology this takes time. In the days of tubes it took quite a lot of time. You could actually see a soft edge at the right and left sides of the picture. This is where the signal voltage ramps up and down.

You can still see this soft edge in a lot of current electronics. I see it all the time in captures from VHS decks. Even some high end DV cameras do it a little (just a black line at the edge of frame)

To account for this, the specs say that only a center portion of the frame must be viewable. The rest, at either side, is there for this ramping. So there goes a small amount of your frame. (I say "the specs" but honestly, I can't quote them for you. It's based on more knowledgable sources)

In addition to this, a lot of TVs have a certain amount of side to side framing slop. In BBs example you might see very different results on different TVs. Some will favor the left, some will lean to the right.

So the upshot of it is that you can't compose your shots close to the sides of the frame in video. This is the opposite of film where the sides are constant because that's where the sprocket holes are-it's the top and bottom edges that are at the mercy of the projector and projectionist. The point is that TV's have sloppy framing and so must you.

Computers and video projectors aren't sloppy. A video projector will/should display the whole frame, including the fading edges at left and right.

Now about DV and sampling. Let's say we're sampling a TV signal. There ain't no pixels in an analog video signal. Each line of signal lasts for a specific time, ramps to zero, resets to the beginning of the next line, and ramps back to signal strength. When we sample (or digitize) we sample for a specific time at a specific rate. For DV25 and some others this gives us 720 samples for every block of time that a line of video uses. 720 samples per line. This includes the ramping time.

As it turns out, the block of time that must be available to show on a TV works out to almost a 704 pixel span. A TV can show less but it's never required to show more. TV's usually show a little less because, well, you just never know so you add a little more leeway. In the wild days of tubes it was horseshoes and handgrenades. (Now, with modern electronics, we play a much more accurate game of horseshoes).

So what about those TV safe lines? They're guidlines, not laws. They're also adjustable. Vegas defaults to very conservative numbers of 10% and 20% but some software uses 5 and 10%. Do BBs test. Render some footage onto a VHS tape and play it on every TV you can. Annoy/bore your friends with it. Then decide what's safe. Essentially, the title safe line should guarantee that you'll see you titles on any museum grade TV you happen to run across. 20% for title safe pretty much guarantees success.

Rob Mack

So, if we assume that ALL CRT TVs will block at least 5%, does it make any sense to crop the outside 5% to save those bits for elsewhere? Or should we keep all of it, figuring that soon everyone will have LCD, DLP, or Plasma based TVs with no overscan?

"This is the opposite of film where the sides are constant because that's where the sprocket holes are-it's the top and bottom edges that are at the mercy of the projector and projectionist. The point is that TV's have sloppy framing and so must you."

As a former projectionist and repair man, I would like to point out that movie theatres are not examples of proper aspect ratios. With several different film aspect ratios available there are many theatres that use a 2:1 ratio for everything. Cinemascope (2.35:1) is cropped by 17% on the sides, flat (1.85:1) is cropped 7% top and bottom. Also many older theatres with steep projection angles crop the sides of the image to hide the keystone. No matter what you are shooting, never place important action or details at the edge of the frame.

Charlie

I'm kind of amused everyone doesn't automatically consider this when taking the shot and practing proper framing. For example you could have poor uncle Bob in some group shot with part of his arm cut off or worse his ear. The outside border area of any shot you frame, 10-15% at least... should be considered throw away area. Its just being professional. ;-)

Dating myself... When I was about 5-6, my aunt Helen was the first to buy a TV. It was a 'giant' screen 12 inch and the picture tube was round. Imagine all that was cut off.

Check out this state of the art beauty circa early 50's.

http://inventors.about.com/gi/dynamic/offsite.htm site=http://www.tvhistory.tv

I've often been tempted to mark out a 10% safe zone rectangle on my camcorder's LCD. Probably a grease pencil used lightly would do the job ok without damaging the LCD. I just haven't gotten around to trying it yet.

Perhaps Scotch™ tape would work well too, slightly blurring the overscan area. It seems like this would get messy as it ages though.

Any other suggestions?

Well, I think that someone with an artistic bent might want to compose on the edges of the frame.You can do it but a little experience will tell you that the important parts need to be well in the frame. It makes it inconvenient to insert still photos into a movie because people often do frame at the edges in stills.

Usually when you are actually composing a shot you'll point things into the frame. Uncle Albert can be at the edge but you might want him looking into the frame...

BB, I think you've got 10-15 years on me but I clearly remember our TV-it was old when I was born I think. I loved to stare at the little white dot in the center and wait for it to fade away. Wish I had that kind of patience now.

Rob Mack

Good information there.

I was just thinking about adjusting the gate on a 16mm projector. You can roll it up or down, even past the frame lines. There's just no side to side adjustment.

I used to project at my college's film festival. We were very, very careful to do the best job we could. Commercial theaters? That's sometimes the worst place to project.

>does it make any sense to crop the outside 5% to save those bits for elsewhere

That's a darn good idea. You wouldn't actually "crop", because you need to keep the same pixel resolution. But you could place a black frame on the top track. (Not black black, the greyish TV-friendly black, of course.) That would free up the mpg encoder to spend its bits on the inner details. Or you could reduce the bitrate for an equivalent quality.

I've never seen this mentioned. But it sure seems logical. Great suggestion.

Nope, doesn't make any sense at all. Your frames will always be a set size. I suppose if you take that area and make it black you might save a tiny bit in compression. It's kind of like removing your mirrors to increase your gas mileage. There's no real percentage in it.

And part of the point is that, while you don't usually get to see all of the frame, you can't know which segment of the frame you will get to see.

Rob Mack

I've used colored gels. Let's you see through but shades the frame. It's more useful on field monitors to simulate 16:9.

It's worth it to use a camera with a good black and white viewfinder with a reticule built in.

Black and white will help you focus. I was lighting a National Geo shoot early this year where the three XL1 cameras used color viewfinders and and manual lenses. The operators couldn't see well enough to tell that the backfocus was off on one of the cameras.

Reticule:
1. [n] a network of fine lines, dots, cross hairs, or wires in the focal plane of the eyepiece of an optical instrument
2. [n] a woman's drawstring handbag; usually made of net or beading or brocade; used in 18th and 19th centuries

Of course I meant #2.

I wouldn't think it'd make a big difference except maybe to render times.

A test is in order here.

Rob Mack

Now that "reticule" came up jarred an old memory. They are common on microscopes, telescope viewfinders, even build into the modern combat chopper targeting systems... how come not common in reasonable priced digital cameras?

Some enterprising person could easly develope some clip on basic system that would snape on LCD viewfinders.. Hmm....

Well, it increases rendering time by about 3x! I can't notice a quality difference going from AVI to a MPEG2 at 4,000k/s bitrate.

Chienworks:

I now have two Sony camcorders, a little TRV8 that I've worked to death, and a brand-new VX-2000 (wow!). Both camcorders crop the image about 5% in both the flip-out LCD and the viewfinder.

Riredale, i've got a Panasonic Palmcorder. I haven't run any exhaustive tests yet, but that 5% figure sounds about right for this camera too. I'd like to add a 10% line as well though. Maybe a 16:9 frame would be nice too.