Capture PAL 50fps / 25fps

[Frankman]

Can I capture every single field (50 fps) from a PAL source with a frame grabber (e.g. Haupauge WinTV)? Or can I only capture the frames (each including 2 fileds) (25fps)?

Frank

[ropdoh]

In a way it is the same thing...

1 field appears every 50th of a second (but these have half the vertical resolution)
the capture card takes two of these and puts them together to create a frame every 25th of a second.

What you can do is capture interlaced video at 25fps, and then convert it to non-interlaced video at 50fps (BUT with half the vertical resolution).

You can then resize to full resolution if you want to. I would use AVISynth to do it. Below is a script:

DirectShowSource("input_video.avi")
SeparateFields() # Separate to 50 fps
LanczosResize(720,576) # Resize To Full Res

[Frankman]

OK, I understand:

this means a capture card can not direct grab every single field.

Frank

[non-linear]

this means a capture card can not direct grab every single field.

Huh???

I don't think you understand the difference between a field and a frame. Whether you're dealing with PAL or NTSC video signals, it's the same concept. I'll use PAL for example. The PAL system uses 50 fields per second. Each field comprises half of the picture used in a frame. The first field has the video for the odd lines, so lines 1, 3, 5..., then the second field comprises of the even lines, so lines 2, 4, 6... These two fields together make a frame. One single field alone does not contain the whole picture, but every second line of it.

With concern for capture cards,they do in fact capture each field. If you capture interlaced video, you will see the interlacing when you play back the video on your PC, as your monitor uses progressive scanning, as opposed to interlaced.

I'm sure more people here could give you much more detailed information about fields and interlaced scanning.

[Frankman]

Thank you for info.

I know the difference between fields and frames.

What I want to know is, if a capture card can capture separate fields or if it has to "wait" for a complete frame.

Your answer is that it is possible to grab every field.
That would be good for my purpose.

Frank

[edDV]

Thank you for info.

I know the difference between fields and frames.

What I want to know is, if a capture card can capture separate fields or if it has to "wait" for a complete frame.

Your answer is that it is possible to grab every field.
That would be good for my purpose.

Frank

Yes the cards are capable of catching every field.

The fields are captured in a odd then even then odd, etc sequence. The important concept to understand is that each field is sampled at a different slice of time 1/50 sec apart. If you attempt to combine two adjacent fields into a frame, there will be a 1/50 sec time displacement every other line. If there is no motion this doesn't get noticed. But if there is motion, the "deinterlaced" frame will have serious line tear. Different techniques are used to process this time displacement to simulate a frame exposed at a single point in time. Each technique results in loss of picture detail and quality. Some techniques take advantage of limitations in human eyesight to hide the damage.

Film and other natively progressive sources are sampled in a different way.