View Full Version : What's the best encoding way for 16/9 Tv.
mohak
14th November 2003, 18:58
Hello.
I use GK for encoding dvd's i usually play on standalone player with a 16/9 sony tv 36". My problem is that i want to have in my avi file a movie vertically stretched so i can switch manually my tv on 16/9 mode (this is not supported automatically by avi container)and have a proper aspect ratio. To do this i tick in GK the input ratio to 1:1. The problem is that with this value my compress test return a % needing lot of space than if i let it at 16/9 ratio. I don't understand why?
SeeMoreDigital
15th November 2003, 00:20
Hi mohak,
Welcome to Doom9's forum.
Personally I can't see the fascination with GK. I generate my encodes using either MPEGmediator or VirtualDubMod. And most of them are 720x480/576 anamorphic 1:1 and are viewed on my 16:9 TV!
I suppose when sending the output video from your PC to your 16:9 TV it all depends on the graphics card you use. But you should never have to ever stretch your encodes vertically... only horizontally!
Cheers
r6d2
15th November 2003, 01:41
Originally posted by mohak
To do this i tick in GK the input ratio to 1:1. The problem is that with this value my compress test return a % needing lot of space than if i let it at 16/9 ratio. I don't understand why?
If you use 1:1 in GK, it will not resize. It will use the anamorphic nature of your source, which contains much more detail than the properly resized frame, so it compresses much less.
However, if size is not an issue, by encoding 1:1 you'll retain much more quality, caeteris paribus.
mohak
15th November 2003, 14:03
@Seemoredigital:
I'am using a standalone dvd player which is mpeg4 compatible(avi=all divx codecs, xvid ; mp3, ogg, wma, ac3). GK is easy to use and deliver a good quality and is always updated with the best tools. And i like the team spirit around it.
I wanted to know when you use 1:1 720x576 what size you get for a standard mvie of 2 hours? Me it's around 2gig.
@r6d2
What i know is that the anamorphic is an optical system. Pictures are vertically stretched during the shots and V shrinked during the show with anmorphic lenses restoring the true aspect ratio. But there is no difference in the size betwwen the two pictures cause it is the same picture (i'am talking about 35mm and 70mm films). Concerning digital movies i don't really understand: When the source is 16/9 pal, the pictures recorded in the vobs seems are bigger at 1:1 then at 16/9 input. Where are the missing pixels when you swith to 16/9 or 4/3. Is is just an electronic zoom and the pictures have the same size. If it is the same why do i need more space for encoding.
@all
and thank you for the welcome.
manono
15th November 2003, 14:49
Hi-
The reason you need a larger file size to achieve the same quality when encoding anamorphically is that you're encoding many more pixels at 720x576 than if you had cropped and resized to 704x304 or something like that. So yes, you might need 3 CDs instead of the normal 2.
...V shrinked during the show with anmorphic lenses restoring the true aspect ratio.
My understanding is that the Vertical resolution isn't shrunk, but the Horizontal resolution is stretched during playback over a 16:9 TV. Please see This (http://www.doom9.org/aspectratios.htm) for more information. In it, Doom9 says:
The studio has made an anamorphic widescreen transfer. It took the full frame from the camera, then horizontally squeezed it to fit the resolution of the DVD. Then they set the 16:9 DAR flag which tells the player to unsqueeze the picture again.
Although I think that if anamorphic lenses are used in the filming, that they do the shrinking. Otherwise you agree, don't you? This is DVDs we're talking about, not how it's projected in the movie theater (same thing though, I think). If you have a normal 4:3 TV set, then there is some Vertical squeezing done. Either that or Horizontal lines are completely removed, depending on whether or not the 4:3 TV can do the "anamorphic squeeze".
Maybe read This (http://www.thedigitalbits.com/editorial/bz100298.html) for more information.
SeeMoreDigital
15th November 2003, 14:57
mohak,
The word Anamorphic means: A distorted image that looks correct when viewed!
In the case of 720x480/576 DVD movie images, this means that all the images stored on the disc are squashed up, from left to right ie: horizontally... not vertically!
A standalone DVD player has user installer settings that will allow you to select either a 4:3 or 16:9 output TV size. And once selected will stretch an 720x480/576 anamorphic DVD image to an correct looking 'true frame' image, automatically!
Don't worry about how anamorphic film images are captured. This is a totally different process entirely to DVD!
When it comes to deciding how much bitrate I need to 'backup' a DVD, it's all down to personal preference! I'm not too bothered about creating perfect looking DVD copies so I usually generate my encodes to fit on an 800MB CD-R.
Cheers
mohak
15th November 2003, 15:59
@seemoredigital
May be we are not talking about the same thing; i mean when looking at a picture (recorded on a dvd at 16/9 PAR) without using a 16/9 viewer, when you look at it's original aspect(ie: in scan monitor) this picture is vertically stretched, it is nearlly a square picture. When you put it through a 16/9 viewer, the viewer squashed it so it looks at a real aspect ratio. In GK i can see it when i tick 1:1 instead 16/9 for the input source. And my question is : is it better to tick 1:1 and encode at this PAR (knowing i'll have to force my wide screen in 16/9 mode) or to use the PAR detected by GK and encode (knowing that i will never have the real aspect ratio on my wide screen).
The other question is when i encode at 1:1 input, at 60/70% i have a huge output size. The same movie at 16/9 input is smallest by half(always 60/70%). So what happened between the two modes.
SeeMoreDigital
15th November 2003, 16:16
Originally posted by mohak
@seemoredigital
May be we are not talking about the same thing; i mean when looking at a picture (recorded on a dvd at 16/9 PAR) without using a 16/9 viewer, when you look at it's original aspect(ie: in scan monitor) this picture is vertically stretched, it is nearly a square picture. Actually what you are seeing is not the source being vertically stretched - The source always occupies either 480 vertical pixels in the case of NTSC DVD's or 576 vertical pixels for PAL DVD's. What you are actually seeing is the source being horizontally or 'anamorphicly' squeezed to occupy 720 pixels!
Cheers
r6d2
16th November 2003, 02:38
Originally posted by mohak
And my question is : is it better to tick 1:1 and encode at this PAR (knowing i'll have to force my wide screen in 16/9 mode) or to use the PAR detected by GK and encode.
@mohak,
There's no "better" or "best" answer. It all depends on what do you want the encode for (archival? a portable device? a HDTV? a PC monitor? an USB pen drive?)
The fact is that if you encode 1:1, you will retain more detail. You make the choice if that is "better" for you!
The other question is when i encode at 1:1 input, at 60/70% i have a huge output size. The same movie at 16/9 input is smallest by half(always 60/70%). So what happened between the two modes.
Rest assured that we undestood your questions and both manono and I have explained to you why the more detailed picture compresses less than the resized one. manono also suggested you to do some reading. Please do. I don't see the point in asking the same questions again.
mohak
16th November 2003, 11:53
@all
sorry guyes i 've understood after reading. I was really convinced that the output was streched verticaaly.
Thanks for all your answers and to have been so patient.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.