Still confused over 16:9 vs 4:3
 

I have been researching past posts regarding shooting 16:9 vs 4:3 in my XL2 and am even more confused.

I am mostly shooting for a government access channel and decided to start shooting in the 16:9 setting. I understand that most people who shoot in 16:9 anomorphic (?) do so for dvd or internet. So am I barking up the wrong tree by limiting my viewing audience? Those who don't have wide screen? Or is it possible to render the project in PP 2.0 for dvd then let the government channel air it through that format. My biggest fear is that those who don't have wide screen will not see the broadcast as it was meant to be seen.

Am I reading in to this all wrong?

Thanks....

If you shoot, render, and deliver it out as 16:9, then when played on a 4:3 TV it should just have a black strip on the top and bottom. Most should see the full screen, just in a letterbox mode.

Mike

I'm not sure that's true.

My 4:3 TVs are wholly ignorant of 16:9. If they are presented with true, anamorphic 16:9 material, it will appear squeezed and occupy the full screen.

I shoot in true 16:9 and, for viewing on my 4:3 CRT TVs, I have to do a true conversion from 16:9 to letterboxed 4:3.

That would be anamorphic which comes from Greek words that mean "changed shape." When you shoot in 16:9 mode it squeezes a widescreen image into the standard 4:3 proportion. Then on playback, your monitor needs to understand how to stretch it back into the correct 16:9 proportions.

If you burn a DVD correctly, the DVD player will handle this. If connected to a 4:3 TV it will letterbox the video with black bars like Mike says. But the letterbox comes from the DVD player itself, and not the TV, as John points out.

So I think the only way to really answer your question is by speaking to the tech people at the public access channel. If your material is 16:9 they may want you to render a 4:3 version which has the black bars on it already. Or they may just prefer 4:3 full screen material.

There are many applications out there where you can change the aspect.

For example, I shot a 16:9 feature, changed the aspect before encoding into MPEG2 to achieve a 720x480 file with a 16:9 video within it. You may want to try MainConcept.

Thank you for the replies.

There are several shows that I have watched on cable or network that appear to be wide screen. I will note that I am viewing them on my wide screen at home. Though these shows have the black bar on the top and bottom am I being tricked into thinking that it is widescreen when in fact it is only 4:3 with the black bands?

Most likely, yes. There are a lot of transmissions now of letterboxed video.

I'm not the brightest on this, I'm just learning. But, the first thing I think I see is the use of the word anamorphic, which is generally understood to be the use of a lens to change the aspect ratio. The XL2 shoots true 16:9, not anamorphic or modified out of the camera, I'm pretty sure. When you shoot 16:9 you are using more pixels or sensor area then when you use 4:3.

If you shoot 16:9 and keep the action more central, and use your title safe areas for titles, your footage should be fine on those few systems that can't show it in letterbox.

If you have a widescreen 16:9 TV and you are seeing black bars at the top and bottom, you are seeing a widescreen theatrical version, or whatever the hell they call the damn thing. Personally, I hate that!

Check with you station. Being in Port St. Lucie myself, and knowing someone at the Martin Co. TV station, I'm curious as to what station you are working with?

Best of luck----Mike

Depends on your editing DVD program. If you place a "FLAG" during authoring, it will communicate to a DVD player. The DVD player will then convert an anamorphic widescreen picture to a letterbox format depending on how the DVD menu was set up prior.

I just got off the phone with the director of our local government access channel. He said there shouldn't be any problem since he has aired other projects that were shot 16:9 wide screen. Of course, the videos displayed the black bars on top and bottom.

I apologize for using the term anamorphic. I am shooting with my XL2 set to 16:9 and capturing into PP2.0. In PP2.0 I have captured using the 16:9 setting so that doesn't seem to be a problem. I guess you would call that flagging.

I have a JVC TM-H1700 broadcast monitor that is set to 16:9. The video is looks great in post and does display the black bars on top and on the bottom.

I would be happy if viewers with 4:3 sets view the video in that manner.

There is a lot of differences of opinions when it comes to this topic. I was up until 3:00 A.M. researching it and finally went to bed with more questions. My fear was that I didn't want the 16:9 image to be squeezed into a 4:3, making objects look tall and thin.

I appreciate each and every one of you for your opinions.

Thanks

Sorry Mike, but you're dead wrong there. See my post above, "anamorphic" means that you have squeezed a 16:9 image into a 4:3 frame for the purpose of recording it. That is the only way to shoot 16:9 DV. All DV is 720x480 whether it's 4:3 or 16:9. It doesn't matter whether you use a special lens to create the anamorphic effect or if the camera is doing it digitally.

The XL2 has a 960x480 sensor so it can capture a 16:9 shaped image at full resolution. However that image is digitally processed by the camera and squeezed into a 720x480 frame.

But "anamorphic" just describes the format of the image, not the quality; it means that you need to stretch it back to the correct proportions when you view it. For example, the PD-150 can create a proper anamorphic image, but the resolution is lower because the CCD's don't have enough pixels.

Actually your use of anamorphic was correct, no need to apologize :-)

the 'video within it' may not be the best way to think about it.

720x480 is a D1 or full D1 size video, which is standard definition (for NTSC .. leave off pal for these purposes). So, you have 720 pixels horizontally and 480 pixels vertically.

Each pixel will have an aspect ratio ... that is, the shape in which the pixel is.

So, you can shoot in 16:9 in SD, which will generate a SD file of 720x480, however each pixel will have an aspect ratio of 1.78:1, to make your picture look correct (e.g. so a circle would be a circle, not an oval).

Your authoring program would 'flag' the video the correct way to let your DVD player know how to correctly display the video on the TV monitor (since your DVD player would know what the video stream is and also be set to know what type of display you have..it's the device that could then to the math correctly).

So, your video is a certain definition (SD, HD, etc.) and the aspect ratio then helps that get displayed to match how it was shot. (that's why the dropdown in an authoring prog. can set the proj. as 720x480 or 720x480 widescreen, for example..they are both 720x480 in terms of resolution, just different dimensions to the pixels).

So when the above proj. was encoded, it wasn't really encoded to fit in a 4x3 SD format, otherwise you'd lose some of your detail, and when you showed it on a 16:9 monitor, it will be 'double lettterboxed', e.g. your player would think it was 4x3 and would letter box the sides and the top would have already been letterboxed. Know what I mean? I think I have this right.

I don't know the particulars, but if you are broadcasting video it's going to most likely crop your 16x9 image because it's broadcsting 4:3. Have you seen the broadcast? What's it doing to the image?

The XL2 doesn't do any anamorphising, the chip is a 16x9 chip, that's what is known as 16x9 native.

The video and TV world began as 4:3, so most of the televisions and computer monitors have that boxy rectangular shape. With the advent of HD and as we get loser to adoption of the HD standards, things are getting more 16x9, which is a long rectangle. More and more video is conforming to the 16x9 shape. When you are capturing, or playing back, the NLE, or monitor needs to know if the signal is 4:3, or 16x9.

You have the choice of shooting in 16x9, or 4:3 on the XL2. if you shoot 4:3 on the XL2, you are loosing a good chunk of your image.

Go here for the XL2 CCD block overview, it may help you understand. If you read all of the different articles on the Watchdog site, I think you will be up to speed on the basics of your camera. And keep shooting, playing and experimenting.

I think we're just into semantics here, but the XL2 has to "anamorphise". The only way to shoot 16:9 on DV is to make the image anamorphic - to squeeze it to fit the 4:3 standard DV frame size.

Mark, the project is not ready for broadcast. I am just trying to head off a disaster and a waist of time. I love the look of widescreen but if it is bound to only dvd, then I will change direction.

Here is a tutorial regarding widescreen. It is located on the digital juice website. I did notice that he mentioned that the yes, anamorphic is squeezing the widescreen video to 4:3, then the DVD player will recognize it and convert it back to display widescreen.

But....... once again, this is for DVD. Broadcast is not mentioned.

Here is the link http://www.digitaljuice.com/djtv/seg...how=all_videos

Here's a little background on the anamorphic process - from one of my favorite websites, The American Widescreen Museum: http://www.widescreenmuseum.com/widescreen/squeeze.htm

It was originally an optical process which allowed existing film cameras to shoot widescreen movies with the help of a special lens. Later the same technique was used with video cameras. But today, cameras with high enough resolution CCD's can capture the full image natively, then use a digital process to provide the squeezing. This is what the XL2 does.

When I shoot 16:9 I don't squeeze anything to 4:3, nothing. The XL2 shoots true 16:9 right? You are not cropping a 4:3 image? Right. If it reduces the size slightly, to me that is not anamorphizing! Just me maybe!!!!

My thesaurus says: Anamorphic===Pertaining to a kind of distorting optical system! Optical-----"Read Lens!"

Are you saying that there is no different term to be used with the XL2 or other camera's process, if so maybe we need to start one before people start to look to buy an anamorphic lens for their XL2's and other cameras.

From Canon's and every reseller’s description------"Canon XL 2 shoots in a true 16:9 ratio without artificial letter-boxing or vertically squeezing a 4:3 image."

So maybe we need a new term!!!!!!!!!!!!!!! I'm all for it.

I have nothing against me being wrong, "dead" or otherwise, but just want to clarify our terminology. Just think that that term does not belong in this discussion.

Mike

Edit.

Check your ------http://www.widescreenmuseum.com/widescreen/squeeze.htm, and almost every paragraph mentions "Lens." So let's coin a new term for it!

Mike

Your camera is doing the squeezing, you just don't realize it :-) And things change fast in the digital world, so a thesaurus isn't always right. Just plug your camera into a regular 4:3 monitor or TV. What do you see? Everything is squashed, so it's too tall and skinny. That's exactly what the XL2 has recorded on tape. You need a 16:9 capable monitor to unsqueeze it.

Read back through some of the posts above... Dave explains it very well. The number of pixels in 4:3 and 16:9 video are exactly the same, but the pixels themselves are differently sized. This has to do with the way the display device interprets the data. On a 16:9 screen the same 720 pixels fill a wider space.

This is the very definition of "anamorphic"... you have changed the shape of the original image. As I mentioned above, this was originally done optically on film cameras. Now we can do it digitally and get the same effect.

When we talk about "true 16:9" or "native 16:9" that just indicates the camera's CCD's have enough pixels to capture a full resolution widescreen image. So we're talking about the method of acquisition there. The anamorphic process is the method to storing the image in a format which is compliant with DV's 720x480 spec. In other words, it's a bit of a "kludge"... a way to fit a rectangular peg into a square hole if you will.

We don't need a new term. "Anamorphic" just means you've changed the shape, it doesn't say HOW you changed the shape. You can do that with optics, analog circuits, or digital processing. That term is actually the very heart of this discussion, and it's the correct one to use.

Don't feel bad, it has caused a lot of confusion in the past and it took me awhile to wrap my brain around the concept also :-)

Maybe this old thread will help?

http://www.dvinfo.net/conf/showthread.php?t=29087

Wrong! Neither me nor my camera squeezes anything to 4:3, it squeezes it to 16:9!

As weird as it may seem, Mike, that's exactly what happens, as Boyd explained.

The 'shape' of the CCDs on the XL2 are natively wider than tall (960H X 480V). When you record in 16:9, the XL2 takes those 960 horizontal pixels and maps them onto a 720 pixel horizontal grid. It has to do that to conform to the DV spec of 720X480.

It flags the video as 'anamorphic' as it goes to tape. When you get ready to import to the NLE, it typically doesn't know about the flag so you have to inform the NLE that your video is 'anamorphic'. This tells the NLE to display your video with the 720 pixels elongated to make the picture wide (16:9) as you work with it. However, there is still only 720 horizontal pixels in your video.

Now, let's say you get ready to author to a DVD. Once again, you have to tell the DVD software to place an 'anamorphic' flag bit on the DVD as it's burned so the player will know what to do with the video when playing back. Graphics files such as menus or animations, don't have any of this anamorphic stuff cause they aren't limited by any spec as to number of pixels. So if you create widescreen graphics or animation, you use 853X480, or more simply 16:9 reduces to 1.78:1. Take 480 and multiply by 1.78 to arrive at the correct number of horizontal pixels.

NOW HERE"S THE RUB. Most folks don't know enough to go into their player's video setup menu to instruct it as to what kind of tv they have attached, 4:3 or 16:9. And for some reason, many players default to 16:9.

So if you have a 4:3 set, and your player sees a DVD with the anamorphic bit set, it will react correctly by placing black bars top and bottom if the player knows you have a 4:3 screen. If it thinks you have a 16:9 screen, it will do nothing, and on a 4:3 screen this would show as a squished image (faces are very thin), but on an actual 16:9 tv it will appear correctly.

Long winded I know, but sometimes explanations from a different angle make things clear. Hopefully I was able to help. If not, reading a couple different explanations might work.

regards,

-gb-

Excellent reply there, Greg... spells it out beautifully.

Some folks seem to be confused about the XL2... yes it is native 16:9 on the CCD block, but it *must* go to tape as 720 x 480, because if it didn't, well then it wouldn't be DV. That's where some people are getting hung up. Some monitors will show it as a Spaghetti Western (that is, squished), if the monitor can't add its own letterbox.

See, now we are having fun, and everyone is learning. But, if it squeezes it to 4:3, why does my 16:9 TV show it as 16:9? I thought it was 4:3? It should have black bars on the sides!

Mike

Nope. It would have black bars on the sides only if it was shot in 4:3.

Your widescreen monitor is smart enough to recognize the 16:9 flag, and un-squeezes it for proper display.

But older 4:3 monitors don't know what to do with it and therefore show it squished (Spaghetti Western).

Mike, as we've all been saying, the DV spec only supports 720x480 pixels. This was originally intended to be interpreted by a TV in the 4:3 aspect ratio. So if you're going to record something in the DV format, it has to be 720x480.

Now when you view that 720x480 image, it's up to your TV to fit it onto the screen in the correct proportion. Your 16:9 TV will have several different display modes which you can choose with the remote. On my Panasonic plasma they're called FULL, ZOOM, JUST and 4:3. The default is for the screen to choose FULL mode, where it fills the entire screen with whatever you're sending it. Now this TV (as well as the 3 other 16:9 LCD's that I have) doesn't seem to care whether you're feeding it 16:9 or 4:3 material. It's up to you to choose the mode you want to view it.

So if you're watching widescreen material from your XL2, it will be in the correct proportion if the screen is set to FULL. But now playback a 4:3 tape... it will still fill the screen but everything will be stretched too wide. In that case, it's up to you to choose 4:3 mode to squash it back into the correct shape and add the black bars on either side (known as "pillarboxing").

Michael,

I couldn't have said it better. This is why these boards are great, we can take the opinions of several and come up with some common factors. I have learned a lot from this post. (Disclaimer: Not to say this will be my last post)

Thank you everyone! I am going to have to leave now since my wife is calling for me to watch some Ben Afflack movie (getting sick!). Let's see if it is in widescreen!

Thanks again guys!!!

Heh, yours is smarter than mine then :-) None of my 16:9 screens seem to recognize this (Panasonic plasma, Samsung LCD, Sony LCD and Gateway LCD). You have to choose the desired mode yourself.

Now this only applies to standard definition video though. High definition would always be correctly displayed as 16:9

Exactly, and this goes back to the original question that Gary Gonsalves was asking!!!!

He was asking about those with only 4:3 sets! And, I said that it would be squished down and stretched out with black bars on the top and bottom. Right or wrong? I originally answered this based on his question, with no concern for anamorphic lenses or anything, just how others would view his footage.

I know I am less than brilliant in this area, but does this not mean that my original answer was correct?

Thanks all----Mike

P.S. This is fun and I hope I'm not pissing anyone off, but we are learning a lot. Tell me to stop if you want!

Mike

Since we're in learning mode (!) - you may also be interested to know that the DV specification supports more than just 4:3 or 16:9...

DV playback equipment *should* send the appropriate display mode signals out on the analog video signal. There are two standards supported:

The IEC standard - three possibilities here:

4:3 full format
16:9 letterbox (centered)
16:9 full format (squeeze)

The ETS standard - eight possibilities here:

4:3 full format
14:9 letterbox (centered)
14:9 letterbox (top of screen)
16:9 letterbox (centered)
16:9 letterbox (top of screen)
>16:9 letterbox (centered)
14:9 full format
16:9 full format (anamorphic)

All, of course, constrained to 720x480 or 720x576.

Did you also know that the DV specification supports Closed Captions and Teletext? My DSR-11 even encodes analog CC signals into the DV stream.

And, since a picture paints a thousand words, here is a true 16:9 DV image represented as a square-pixel 720 x 480 JPG:

http://www.enosoft.net/DVProcessor/native16_9_2.JPG

and its letterboxed (but still 720 x 480) equivalent:

http://www.enosoft.net/DVProcessor/letterboxed_2.JPG

(they are deliberately chosen to show the interlacing).

Here is a layman's description of the Widescreen process:

http://gregl.net/videophile/anamorphic.htm

Have a look at John Miller's post #3 on the first page. This is what happens on perhaps 98% of the 4:3 television sets in the US. Only the very newest ones are capable of letterboxing a 16:9 image. Generally speaking, a 4:3 set just fills its screen with whatever you send it.

Unfortunately, we have to get into "anamorphic" for this discussion. Remember, that simply means that the shape of the image has been changed. The camera recorded a widescreen image, but squashed it into a 4:3 frame to record it on tape. So your 4:3 TV doesn't know any better... it just displays it as though it was a full screen image. The result will be squashed, with everything looking too tall and skinny (which might not be bad, depending on who you're looking at ;-)

If you want the black bars (letterbox), the you will need to provide them for the TV. As discussed, if a DVD player is properly configured so that it "knows" it's connected to a 4:3 TV, then it will scale the image properly and add the black bars. So it's sending a correctly proportioned 4:3 image to the TV, with the 16:9 image in a letterbox.

But if you aren't distributing the video on DVD, then you will need to provide the black bars and scale the video yourself. You can do this with your editing software. Don't know about PC's, but in Final Cut Pro this is as simple as dropping your 16:9 video into a 4:3 sequence. The software will automagically letterbox it.

The downside here is for people who have 16:9 TV's; they don't get the full quality possible. Letterboxed 16:9 only uses 360 of the available 480 scan lines. There are 60 black lines above the image, and 60 black lines below it, so in other words, you're wasting 25% of the available lines. But an anamorphic image uses all 480 lines, and stretches them to the full width of a widescreen TV.

I'm not familiar with the XL2, but the Sony Z1 can be set to provide the letterbox itself when connecting the camera to a monitor. I'm not sure whether the XL2 can do this, I kinda think that it can't. But you would only want to do this if you're viewing a tape from the camera directly on a 4:3 TV.

Great information Chris and Boyd...thanks

Thanks to all, great informative post!

Mike

Most people with 4:3 televisions should have figured out by now to adjust their DVD player for proper widescreen playback, unless they've never watched a widescreen movie. It wouldn't normally be up to the TV to make this adjustment, it's controlled by the player.

Just in case anyone isn't clear yet about anamorphic video, the key is that the recorded pixels contain a non-square portion of the image. That's confusing because we tend to think of pixels as being square, but in this case they're just not. As previously mentioned all DV (NTSC) video contains 720x480 recorded pixels, while basic math tells us that a 16:9 image should have 853.33 x 480 pixels (if the recorded pixels were square). So if you want 16:9 output on a square-pixel display using 720 x 480 recorded pixels, then each recorded pixel has to represent ~1.185 pixels on the horizontal axis of the display. That's the "pixel aspect ratio," which in this case would be 1.185.

Or to put it another way, if you took widescreen DV video of a wall 853 feet wide by 480 feet tall, each pixel in the recorded image would represent an area of the wall about 1.185 feet wide by 1 foot tall. If you then played that image directly on a square-pixel monitor without adjusting the aspect ratio, it would make the wall look like it was only 720 feet wide, which we know is not correct. But if your video player knows the pixels aren't square it can process them to show correctly on a square-pixel display, making the wall look like it's 853 feet wide again. Again, the key to this is understanding that the recorded pixels do not represent a square portion of the image.

This all applies whether you get the result by using an anamorphic lens to bend incoming light onto a 4:3 sensor, or using a camera like the XL2 which can interpret the data from its widescreen sensor to cram the results onto DV tape. Either way the end result is 720x480 recorded pixels with an anamorphic pixel aspect ratio of 1.185 (for NTSC). The benefit of having a camera like the XL2 is that you don't have to put a special adapter on your lens to get anamorphic widescreen recording, but it's still anamorphic widescreen recording. With the XL2 the "anamorphising" is done electronically after the light hits the sensor, rather than optically before the light hits the sensor.

Also see what Wikipedia has to say about all this: http://en.wikipedia.org/wiki/Anamorphic

Don't forget, though, that 720 x 480 NTSC doesn't use square pixels for 4:3, either!

For square pixels, the dimensions would be 640 x 480 (just like VGA resolution).

Likewise, for PAL, square pixels would relate to a 768 x 576 image.

But, since DV samples at 720 pixels per line, you end up with slightly skinny pixels for NTSC (for 4:3) and slightly fat ones for PAL....


....anyhoo, I always thought a simple solution for watching squished material on a standard 4:3 display would be some inexpensive, anamorphic glasses!

Great post too Kevin! This turned out one very informative thread and thanks to all who contributed. A little coaxing got out the big guns with the best of info.

The sticking point for some, me included, is that we generally never play anything back straight from the camera, or raw, to the old 4:3 TV, it is always burned to a DVD and the player makes life easier for us, and thank God for that! Even my 12 year old little 4:3 TV knows what to do then! I'm sure that the TV stations have no problem with it either.

This post is worth reading several times, and that is what I will do for sure. With all the excellent posts and great links, it almost at sticky level.

Mike

I have a question along the same lines as whats being discussed but with the difference being the resolution not compressed to DV. For example a native 16:9 SD cam that records component out to an uncompressed codec. Would it be 960x480 NTSC, being that it doesn't have to conform to DV spec? Always wondered about this.

Just to throw one more chink into all this:

Many of the HDTVs (widescreen) out there have a button or feature that allows you to choose the aspect ration to view: 4:3, 16:9 (letterboxes), 4:3 expanded (scales and fills up the screen...crops pic, looses resolution), and 16:9 fullscreen.

I'm using FCP and DVDSP. I shoot in 16x9, capture to a 16x9 anamorphic timeline and export for 16x9. DVDSP is marked for 16x9. My question is...will that give me the 16:9 letterboxed or 16:9 fullscreen? I just want to use up the whole screen and not have this smaller version with letterboxes.

By the way, very informative thread!

Jonathan

That will depend on the user's DVD player and TV. If they have the player correctly setup, an anamorphic DVD will fill a widescreen TV at the highest possible resolution. If they have a 4:3 TV then the player will provide the letterbox.

But those different modes for widescreen TV's are only selectable by the person with the TV. You can't force the TV into the correct mode with standard definition footage unfortunately. So the user could set a widescreen TV to 4:3 mode and your anamorphic DVD would appear squashed in a small recatangle with a black bar to the left and right for example. Or if they set the TV to zoom mode the video would fill the width of the screen with a squashed image while chopping off the top and bottom.