Full-Frame Vs. Anamorphic: The Same Quality? (1 Viewer)

[Vince Maskeeper]

No, anamorphic gives "more" resolution when compared to non anamorphic letterbox (since resolution is wasted on bars in the video signal).

Again- 4:3 images would be 720x480. An anamorphic 1.78:1 image would use 720x480 as well. To do 2.35:1, even anamorphic, some normally used picture area is cocupied by black bars, meaning wasted resolution... thus less that 720x480 pixels would be used to make the picture.

The width of the 16x9 picture is stretched across the full width on a 16:9 set--- but it's still only 720 pixels across.

-Vince

 

[Vince Maskeeper]

Again, it would be impossible to answer the question of PQ- as that there is no formal definition of what defines picture quality. So many factors here, it's simply impossible to make a broad statement.

In terms of strict resolution: 4:3 dvd is 720x480, 16:9 anamorphic is 720x480, in the case of wider aspects than 1.78:1, some resolution is used on bars, so it technically has less resolution.

But again, PQ is not always defined by resolution-- so these numbers could mean very little. But it does answer your specific question.

-V

 

[Randy A Salas]

Great feedback, Vince, and an interesting topic. Thanks for expanding my understanding.

 

[David Von Pein]

Exactly what Randy said, Vince.

Thank You.

 

[Brian Lawrence]

On a 4:3 transfer I take it the the vertical lines are spaced closer together than horizontal lines?

Otherwise 480x720 would result in a 1.50 aspect ratio rather than 1.33 ??

As 480 multiplied by 1.5 equals 720.

 

[Dan Rudolph]

Right. The pixel AR (no matter what the displayed AR) is 1.5:1. So neither 4:3 nor 16:9 transfers use square pixels.

 

[Michael Reuben]

WHY isn't it possible to say whether one looks as good (clarity-wise) as the other?

What Vince said.

M.

 

[ChrisMatson]

I hate to beat a dead horse here, but can someone clarify this one...

Let's say you have a 4:3 36" TV. The viewing area is aproximately 28.8" wide by 21.6" high, or roughly 622 square inches. The pixels are spread out at 720 x 480, or 345,600 pixels (555.6 pixels per square inch).

Now, you take this same TV and apply vertical compression, creating a 16:9 image with a viewing area of 28.8" wide by 16.2" high, or roughly 467 square inches. The same 345,600 pixels now cover a relatively smaller area (740 pixels per square inch).

Now, even if some of this 16:9 area is used to draw black bars for a 2.35:1 picture, there are still more pixels per square inch compared to the 4:3 image, resulting in a picture with increased resolution.

Where did I go wrong?

 

[Damin J Toell]

Where did I go wrong?

The pixels per square inch calculation isn't true resolution, it's just the perceived resolution. By your calculations, you could also have a tiny 4x3 set that has a higher perceived resolution, but that doesn't mean there truly is any increase in resolution.

DJ

 

[DeeF]

Chris,

I don't think you're wrong. To my eyes, an anamorphically-enhanced DVD presented on a 4:3 television *looks* better, because it is using all the pixels but squeezing them together, vertically. It is very similar to the rise in "resolution" achieved by Cinemascope, using ordinary 35mm film, but anamorphically stretching the image (and then squeezing it).

It isn't actually a rise in resolution at all, just a rise in picture quality.

Here's a curious conundrum, though: I have a 16:9 television (a plasma). Anamorphic DVDs look great, of course. But 4:3 material looks great too, and both 16:9 pictures and 4:3 pictures fill the screen vertically. So, is my television actually squeezing the 4:3 picture down, vertically? Shouldn't the anamorphic squeeze make enhanced movies look better, because they have been squeezed? It's a conundrum I don't completely understand.

 

[Brad Hood]

I think it depends on how the two transfers were done. If the 4:3 version is just a zoomed-in version of the anamorphic transfer, I would expect PQ to be worse.

If they are two completely seperate transfers, then all bets are off; either could be a better transfer.

David, to answer your original question, there is nothing inherently better about anamorphic transfers, but it does maximize the number of pixels in an OAR image when displayed on a 16:9 or "squeezed" 4:3 TV.

 

[DeeF]

Here's an interesting idea:

For future, the standard becomes 16:9. Perhaps the frame size on DVDs becomes 16:9 as well.

Why not enhance 4:3 movies *horizontally*? In other words, use the entire 16:9 picture area, and stretch the movie side to side, and then squeeze it on the monitor to its proper shape.

 

[Brad Hood]

I would assume 16:9 will be the standard for HD-DVD.

 

[DaViD Boulet]

hate to beat a dead horse here, but can someone clarify this one...

Let's say you have a 4:3 36" TV. The viewing area is aproximately 28.8" wide by 21.6" high, or roughly 622 square inches. The pixels are spread out at 720 x 480, or 345,600 pixels (555.6 pixels per square inch).

Now, you take this same TV and apply vertical compression, creating a 16:9 image with a viewing area of 28.8" wide by 16.2" high, or roughly 467 square inches. The same 345,600 pixels now cover a relatively smaller area (740 pixels per square inch).

Now, even if some of this 16:9 area is used to draw black bars for a 2.35:1 picture, there are still more pixels per square inch compared to the 4:3 image, resulting in a picture with increased resolution.

Where did I go wrong?

Vince has not been trying to be obtuse...he's expressing the difficulty of answering this question the right way.

You see...as for which version (4x3 or 16x9) of an image looks "better" depends on other variables.

As you mention...compressiong a 16x9 WS image on a 4x3 display will pack the scanlines (vertical pixels) closer together increaseing the clarity beyond what a normal 4x3 iamge would have *on that display*.

You're right!

But the image is also *smaller* than a 4x3 image filling the whole screen. If you moved up to the image so the apparent heigh of the 16x9 image was the same as the apparent height of the image when you would normally watch 4x3 images from your couch...now the image has no more clarity because you've moved closer.

Starting to make sense?

the greater "clarity" of 4x3 over 16x9 (using the same 720 x 480 pixels) depends on the image size and viewing distance when you switch from one image to the other.

If I had a 16x9 TV and switched to a 4x3 program...suddenly my 720 horiztonal pixels are compressed into the center 4x3 area of the screen. The image looks sharper as a result. Since the image is *smaller* than a 16x9 image that normally fills my 16x9 screen...it appears higher-resolution as a result. That's just the opposite of what you experienced on the 4x3 set with the 16x9 sqeeze!

If I move closer to my 16x9 screen so the *width* of the image is now apparently the same as when I normally sit back on the couch to watch a 16x9 film...now the image has no additional clarity because it's larger.

Can you see how the pixel density alone can't really tell you how the "visible resolution" will appear? It depends on screen size, viewing distance, and how various aspect ratios are displayed compared to one another.

BTW, for this reason sometimes "resolution" of video images is referred to as ratio of screen-width resolution to screen height. The idea is that if the height is kept constant then you can compare how the horizontal resolution changes.

That too has problems because it makes all widescreen material appear to have less resolution than 4x3 material which packs the same 720 pixels into a narrower image.

Easiest way to think about it then is that any 720 x 480 resolution simply has 720 x 480 resolution. Whatever "shape" the image is in is a secondary issue and how it appears depends on factors in your playback chain (like viewing distance and display aspect ratio).

Bottom line...you've got a certain number of pixels...you want to use as many as you can to store the image area of the movie wasting as few as possible on any "black bars".

If it all sounds confusing that's because it is!

 

[DaViD Boulet]

For future, the standard becomes 16:9. Perhaps the frame size on DVDs becomes 16:9 as well.

Why not enhance 4:3 movies *horizontally*? In other words, use the entire 16:9 picture area, and stretch the movie side to side, and then squeeze it on the monitor to its proper shape.

Dee. That's what a 4x3-encoded image on DVD already is.

There is no "stretching" or "sqeezing" on a DVD. It's a common misperception and one that even industry execs and professionals continue to promote by the use of the term "anamorphic" regarding 16x9-encoded DVD images. DVDs are digital images--they are not "streched". they simply "are". They have 2 paramaters that are relevant to this issue: 1. pixel resolution and 2. aspect ratio.

DVD has no "anamorphic" feature. Folks...listen up! DVD (NTSC-oriented) discs store 720 x 480 images in either of 2 aspect ratio shapes: 4x3 or 16x9.

It's that simple. Neither one is the "standard" on the DVD format and neither one is the stretched or "anamorphic" version. They're just 2 aspect ratios in the MPEG2 system and DVD is designed to work with both.

Terms like "anamorophic", "streched", or "sqeezed" only have *relative* meaning in an analog architecture given one aspect ratio as the "standard" and all other aspect ratios as "deviant".

If we considered 4x3 to be the "standard" then we could (arbitrarily) label all other aspect ratios as "anamorphic" because they *look* stretched when we display them in native form on our 4x3 screen. But what if we considered 16x9 to be the "standard"? Then the same 16x9 image on the DVD that had been labeled "anamorphic" in the previous context suddenly becomes *normal* and the 4x3-encoded image now looks "stretched" when we display it natively on our 16x9 display--We have to select a speical "4x3" mode on our 16x9 display which sqeezed or digitally resizes the image to a 4x3 area in the middle and pastes black bars on the left/right sides. Or we might have a DVD player that auto-adapts the image to our native 16x9 display (like my DVD player does with progressive-scan output) so we never have to think about what's going on...(just like 4x3 viewers with their DVD player set to "4x3 lbxed" never have to think twice whether a WS movie is 4x3 lbx or 16x9 encoded).

In either case described above...the digital image on the DVD disc has been exactly the same...either a 4x3 or 16x9 image. What's changed is the playback context and how we labeled the image....which is purely semantic and utterly arbitrary.

This is why I eshew the use of the word "anamorphic" in any technical discussion about 16x9 DVD...it confuses people and somehow holds up 4x3 as some deemed "standard" and assumes 16x9 to be secondary. Not in my book

So considering DVD (accurately) to be merely a digital medium that stores a 720 x 480 image in either a 4x3 or 16x9 aspect ratio...

All DVD players are required to be able to convert 16x9 images to 4x3 for backwards-compatibility with 4x3 TVs. However, DVD players are *not* required to convert 4x3 encoded images to 16x9 because when the DVD spec was created no one ever thought about 16x9 TVs as being common enough for this to matter. This is why those of us with 16x9 players that "lock" into 16x9 mode with 480P signals need to seek out DVD players that *do* convert from 4x3 to 16x9 as well.

My Panny rp91 can electronically resize 4x3 lbxed images to 16x9 and it can also "windowbox" 4x3 full-frame images to 16x9 preserving full resolution of the 720 x 480 image in the 4x3 window area.

 

[Jack Briggs]

Really, after Vince and DaViD have explained this so clearly I don't see what the problem is. Key to this misunderstanding is a desire, I think, people have to make things more complicated than they really are. Add to this mix the inappropriateness of the word anamorphic when applied to DVD, and the thinking that non-16:9-encoded DVDs have, by default, "33 percent less resolution," then there you go.

 

[Vince Maskeeper]

This discussion is of interest however- at least to me- because I'm often amazed at how many people seem to misunderstand, misinterpret, or just plain miss the boat on anamorphic enhancement as it pertains to DVD. If we, the tech geeks and home theater nuts, have such a hard time understanding it- it makes me wonder what that will mean for the "regular folks".

I have a pal with a 36 inch WEGA with the squeeze- so he leaves his dvd player in 16x9 mode and uses the squeeze as needed. I can't count the number of times I've come in to find his roomate watching anamorphic discs without the squeeze on, with the actors tall and skinny. He seemingly can't tell the difference.



This is really the key. Regardless of how your arrange the pixels in output- you have a fixed amount.

-Vince

 

[DaViD Boulet]

"Anamorphic squeeze" is a function of 4:3 sets, not 16:9 sets like yours. When people discuss the "squeeze"- it means they have a 4:3 tv set with a "16x9 mode" which allows you to get the extra resolution of anamoprhic dvd, even if you don't have a 16:9 set.

Actually...to be even more specific, the "anamorphic squeeze" only refers to displays that are ultmiately *analog* in nature (CRT) which opt to adjust their scanning-raster in order to compress or expand the image to alter aspect ratio. This is true for most 4x3 direct-view sets with the "16x9 mode". Even some 16x9 CRT RP sets merely "expand" the vertical raster when zooming 4x3 lbxed material though many of them digitally center 4x3 material and scan the entire horizontal distance so they can display gray-bars on the sides to minimize burn-in.

Some CRT displays (like my 16x9 Proscan Monitor) and all digital displays (like LCD, DLP, Plasma etc.) adjust aspect ratio by digitally reconfiguring the image to fit in the same scanning-area.

It would be possible to have a 16x9 CRT display that would present 4x3 material by literally "squeezing" the horizontal raster in tighter to bring the full 16x9 area into the 4x3 window just like 4x3 CRT sets sqeeze their vertical raster into a 16x9 area.

Most high-end displays do these aspect ratio adjustments in the digital realm either because the displays are digital themselves (and have a fixed-resolution pattern to which the image needs to be "mapped"--This is a fine solution as most digital displays offer far more than 720 x 480 resolution and so mapping the 720 to the center 4x3 area does not discard any information in the process) or in cases of CRT displays, the scanning beam is calibrated for maxiumum "fill" at a single rate/adjustment and then all images are digitally modified to fit this rate. This ensures optimum image quality regardless of aspect ratio without one aspect ratio being favored while other might risk scanning-beams overlapping or leaving unfilled gaps.

 

[Jack Briggs]

On a very, very side note: A couple of years ago I had a co-worker whose boyfriend had just purchased an NTSC-only Sony KV-36FV26 WEGA. Her boyfriend just wanted a "big screen." I told her that the only good reason for purchasing that set (when he could have bought into ATSC technology) was to take advantage of its 16:9 mode.

The next day, she told me she had repeated my comments to her beau. He was not even aware he had a 16:9 mode, and didn't know what a 16:9 mode was to begin with.

Sizing up the tech-mindedness situation quickly with my co-worker and her boyfriend, I wrote on a piece of paper what the fellow needed to do (select 16:9 output on DVD player, enable the Sony's 16:9 detection, etc., etc.).

Well, the next day Kristin comes into the office and tells me the results were "fantastic." But the clincher is: She told me she was dreading what the outcome of the experiment would be. She thought the letterboxing would be even more extreme with only a little slit to see. And that was after telling her repeatedly the window would be the same but the picture sharper.

My point? Even when you tell some people about this stuff as simply as possible they still don't get it.

 

[Brian Lawrence]

So I take it that there are different types of pixels used in different applications (as in dvd vs. Tiff, bitmap or Gif) ?

When I do a straight image capture of a 1.33 film image, My powerdvd program saves the image as an uncompressed Windows bitmap image, and it has a 1.50 aspect ratio rather than 1.33.

Are the Bitmap pixels shaped differently then pixels used on a dvd transfer , or are dvd transfers in a way enhanced by having a 1.5 image "squeezed" to 1.33

This image has been converted to a jpeg with heavy compression so as not to be slow loading but the size and dimensions remain the same. As can be seen, when I do a straight image capture the aspect ratio is 1.50 and the image has the full 720x480 pixels



In order to get the image in the proper aspect ratio on a bitmap capture I have to resize the 720 width down to 638



So when such an image is shown on a 4:3 TV, Is the vertical res being lowered down to 638 lines, or does it display the full 720 lines?