Handbrake help

[robertbiferi]

ok but

Well I do understand what you are telling me I just need to know if I have this part right? I have 2. things I just would like a Yes or No

1.
Eatch Field is a compleat picture in time.

The Top Field is a compleat picture.
And the Bottom Field is also a compleat picture.

But they draw the Bottom Field after the Top Field starts to fade away so the movement is smoth.

I know I have this right but if I don't just tell me?

2.
Now the Top Field and Bottom Field both make up one Frame.
But the two Fields are not ment to be seen on screen togather.
This is why the Top Fields starts to fade away then the Bottom Field is drawn.

But they still say the Top Field is in the Frame with the Bottom Field because it is not compleatly faded away.

Do I have this right?

[admin]

1. No, each field only has half of the lines of a complete frame (a field is also known as a half-frame), so it would look something like this when shown at the correct height, with blank lines in between each line of data.



2. They are not displayed at the same time, but they can be "seen" at the same time, and in fact, that's what your eyes see since the human brain cannot see the 1/60th second fields as separate images due to the persistence of vision effect, on an interlaced display. By design (again, on displays capable of showing native interlaced content), the top field would have faded away already when the bottom field is shown, but in practice, this depends on the display technology I would think.

The reason they're not displayed at the same time on an interlaced display is not because it would give some kind of artifact or whatever, it's because interlaced is designed this way, to increase the number of updates per second while still saving bandwidth (where each field is only half a frame, but it still gives the viewer 60 updates per second, to make the picture smoother than simply having 30 complete frames per second). It's the analog equivalent of video compression.

When interlaced content is displayed on a progressive screen (which do not display fields, only frames, so deinterlacing can be explained as finding the best way to reduce 60 fields to 30 complete frames), with the most basic weave interlacing, it's actually showing both fields at the same time. So if there's movement between fields, you get the weaving/comb effect, where the odd lines may be in one position, and the even lines are in another, like this:

[robertbiferi]

thanks but

Ok I thought first the Top Field was drawn then as it fades away the Bottom Field is drawn.

But you say NO am I right just Yes or No this will help me more before I go on I think I missunderstood you?

[robertbiferi]

sorry but

I am so so so sory I think I get you now.

One Frame is made up of two Fields a Top Field and a Bottom Field.
So when the Top Field is drawn this is just half the Frame.
Then when the Bottom Field is drawn this is also half the Frame.

This is why it takes two Fields to make up one Frame I get it.
Do I have this right just tell me Yes or No please just Yes or No?

[admin]

Yes, and by half, I mean like the way you already described in your post here, where the top field consists of odd lines from the frame interleaved with blank lines, and the bottom field starts with a blank line and is then interleaved with odd lines from the frame, but with slight time differential between the two fields, so that the picture shown by the lines in the bottom field actually occurs a 1/60th of a second after the lines in the top field.

Put both fields together, and you get one picture without any blank lines, but the way interlaced displays work, you only ever see one field at a time, but it's moving so quickly, your eye blends one field right into the next and it looks like a whole picture.

You might ask why doesn't progressive displays just do this too, turning the fields into frames and then double the framerate from 30 to 60, and actually there are many forms of deinterlacing that does exactly this, such as bob deinterlacing.

There are also more advanced methods that tries to fill in the blank lines based on fields/frames before and after the current one (interpolation), and Handbrake's "Slow" deinterlacing setting (Decomb only uses "Slow" when it detects combing).

[robertbiferi]

thanks but

Ok I do get that if I was able to see just the Top Field I would see all these Black Lines going from left to right through the Frame because this is were the Bottom Field will be drawn next.

But the two Fields go so fast they look like they are onscreen at the same time and they look like they are in the Frame at the same time but they are not.

I get all of this.

But if you are watching a car go by the screen tell me this?

The car would be moving none stop so would I be right the eatch Field will have very little movement of the car.

Just say Yes or No I may not have ask this right so just say Yes or No.

[admin]

Yes, each field represents the picture at 1/60th of a second, so between each field, there's movement of 1/60th of a second. So for fast moving objects, like a car, the bottom field would show the the car's position at 1/60th second later than than the car's position in the top field.

But this is intention, as this is where the bandwidth savings come from compared to having 60 complete frames per second to get the same smoothness in motion, which would mean double the amount of bandwidth (for analog and uncompressed content - for digital compression, it works by comparing differences between frames, and so the number of frames doesn't matter as much as the differences between frames).

Of course, converting interlaced back to progressive means that you have to fill in the details (the blank lines) that were missing from each field, and try to resolve the problem of the next field's position having changed. This is what deinterlacing is all about, trying to recreate the missing data and make the picture look better (while not sacrificing the smoothness of the motion too much).

[robertbiferi]

thanks but

Ok so say you see a Car going across the screen you know the Top Field will have movement in it and the Bottom Field will have movement in it.

And they Interlace them so you think you see both Fields at the same time.

Now if you De Interlace it you are turning it into Prograsive so I would gess it works like this?

Whatever De Interlace option you pick to De Interlace it with they all will try to take that one Frame that has two Fields in it.

And turn the Top Field into one Frame by its self and take the Bottom Field and turn this into one Frame by its self.

I think this is right tell me if I have this part I think I did understand you?

And one more thing
Standard TV is 525 Lines but only 480 are seen.
But standard TV DVD none widescreen is 720x480 but I thought if you convert Analog TV to Digital TV it is 640x480??

[admin]

Just to make sure, by movement I mean the following:

1. Let's say you have a scene with a fast moving car driving from left to right on the screen, and there's a lamp post in the background.

2. For the top field, which is a still picture that only has the odd lines (the even lines are blank links), the picture shows the car is still to the left of the lamp post

3. 1/60th of a second later, the bottom field is displayed (with even lines, and blank odd lines), but the front of the car has already made its way past the lamp post

On an interlaced display, this looks like normal motion because the screen is refreshing at 60 times a second.

But if you want to convert the fields back into progressive frames, you can see where the problem lies, since if you simply combine the top and bottom fields (adding the even lines from the bottom field to replace the blank lines in the top field), the position of the car won't match up, and you get what's called the weave artifact. This simplistic method simply shows both fields at the same time on the screen (turning 60 fields per second, into only 30 frames), and is one way to deinterlace.

There are many many other methods of deinterlacing, including turning each field into a frame, and filling in the blank lines by guessing what should be there. There are smart ways to do this (like checking the field before and after, to see what should be in the blank lines), or simple ways (simply blend the blank lines with their surrounding pixels).

[dr_ml422]

Just throw the video in the standalone/w/e and let the TV/HDTV do it's thing. You'll either get 720P/i or 1080P/i depending on your set and player and of course the source DVD. In this case if it's deinterlaced it'll be converted to w/e your set can handle. I think I got this right. Let me know if I'm wrong.

Better yet get a PC and use one of the many other and way better sw available for free for encoding etc... Handbrake is apple/Mac land I believe no?

[robertbiferi]

thanks but

Thanks but just som things?

1.
I know there is a lot of other Video Converts that are free but I don't know of any that let you De Interlace if you want to?

2.
Am I right that 640x480 will be standard TV 4:3 I ask because in a program I am trying it has a list and it say this
320x240 standard
640x480 VGA
720x480 NTSC

Now I have a Digatle camara to take photos and it take video at 640x480 and if I conect it to my TV it plays good and is 4:3 so am I missing something?

Is 720x480 just for DVDs and 640x480 is for Compters?

[admin]

Most video converters handle deinterlace in one way or another, do you mean a video converter that lets you keep interlaced content?

For resolutions, I recommend you read this guide (where would we be without the Handbrake docs):

https://trac.handbrake.fr/wiki/AnamorphicGuide

Also this page:

http://lurkertech.com/lg/pixelaspect/

[dr_ml422]

Yes I do believe 4:3 is 640x480 on the set. 720 you hitting Hi-def land and if the source is 720P and your set can upscale, it will bring it to 1080i or 1080P which is true Hi-def. The i being interlaced I believe, and the P all together.

This should clear things up a bit on that end: http://www.digital-digest.com/highdefdvd/faq.html

and this should help w/the minor yet still differences between i and P and what's better according to the individual:

http://www.zimbio.com/High-Definition+Television/articles/151/1080p+720p+mean

http://www.avsforum.com/avs-vb/showthread.php?t=698612

http://www.hometheater.com/geoffreymorrison/0807061080iv1080p/

It's good to know the basics then take it from there. I still am a lil green myself though a bit advanced in the encoding w/certain software, mostly free. 1 I'm already familiar w/is MeGUI. The newest and simplest is VidCoder.

Both can be found on the Main site: http://www.digital-digest.com/ under software. You'll need to log-in separately and I think register as well, for any feedback/suggestions, though welcome to browse any time. If you can't find something here that suits you, which i seriously doubt, just post. I'm betting though you will.

[robertbiferi]

thanks but

I wanted to thank you for all the help and I do need help with one thing and I hope you can help?

I have Movie Studio 8.0 and it is by Sony.

When I start the program and go into File then Project Properties all the settings in here is for when I import a video into the program.

Video Size 720x480 NTSC
Frame Rate 29.97 FPS
Field Order Bottom Field First
De Interlacing type Blend Fields

Now I did go to there Forums and asked about the settings in here.
They just tell me that I sould just set everything to the same specs as the video I will be importing.

And I get Frame Size 720x480 NTSC is for the USA this I get.
And Frame Rate 29.97 FPS for the USA I get this.

But Field Order Bottom Field first I don't get sould I not set it to Top Field First?

When video is broadcast on a CRT screen it draws the Top Field first then as this Fild fades the Bottom Field is drawn.

So is I have to match my Project Properties to the video I import sould I not set it to Top Field first?

What am I not getting?

[admin]

Interlaced digital video can be either way, top field first, or bottom field first, and the advice you received is correct, you should match the encoding to the properties of the video (so if the source is BFF, then you should encode this way).

Field order/dominance is really only an issue with digital files and the players/editors that work with these files. On analog/CRT displays, field order doesn't matter, it's just a series of fields displayed one after the other. I've only used top field being displayed first as an example, I mean when you turn on the TV, how can anyone tell which was the first field being displayed, and nor does it matter. If you converted to DVD, for example, your DVD player will read the field dominance flag and output an analog signal to your CRT TV accordingly (the analog signal won't care about field dominance, neither will the CRT TV, but the DVD player needs to know in order to output the right analog signal).