Re-interlacing footage at the broadcasters request: normal or not?

[Ben Holmes]

Whether or not the resolution is reduced by playback on an interlaced display is immaterial – An interlaced master has the same field resolution. The only difference for playback is that the interlaced rushes have smoother more ‘video’ motion, with different video on each field – but they don’t have full resolution fields – that’s what interlacing is for as I’m sure you’re aware.

The real difference is how each format deals with motion – not overall resolution, and the effect is a subjective one, when you are dealing with the difference between 576p25 and 576i50. Yes, the progressive image has less motion information, but that’s not really the point here, and there are many advantages to mastering and shooting progressive (such as DVD mastering and the increased perceived vertical resolution when viewed on a progressive display – like all those LCD TVs we have now).

I take your point, but I don’t think that’s relevant to the station tech’s comments, which just sound ill-informed. He wasn’t complaining about the resolution, he just didn’t like the motion.

Ben

Editec Broadcast Editing Ltd

EVS & FCP specialists for live broadcast.

OB Server 1 HD – Mobile FCP editing done right.
https://www.editecuk.com/OBServer2.html

[Larry Asbell]

Lot’s and lot’s of misinformation so far in this thread.

First of all on the part of the broadcast engineer in Paulo’s original post. But then:

Arc nevada says one frame of 640 x 480 interlaced is made up of 2 fields of 320 x 240. A field has one half the lines of a frame but they are the same size lines! One frame of 640 x 480 interlaced is made up of 2 fields of 640 x 240.

Then the venerable Shane Ross I believe misspoke and didn’t catch the fallacy in the Engineer’s saying the progressive frame has less information than a interlaced frame. Shane says “progressive footage means that BOTH fields contain information…the SAME information.” – No, No, No. (Still Love You, Shane!)

Every line in a video frame that’s shot in progressive of a moving scene contains unique information. In PAL, each 1/25 of a second all the scanlines capture the scene from the top down, in order. In interlaced the odd ones capture the scene, skipping where the even ones would be, then the even ones capture the scene, skipping where the even ones would be. In each type of scanning, moving objects are captured in different positions in adjacent lines which is why there is no true changing of progressive footage to interlaced or vice versa.

Progressive contains every bit as much information about the scene as interlaced. Interlaced samples the scene twice as often but with half as many pixels. Interlaced has more temporal information but less spatial information.

Hope this helps.

– Larry Asbell

[Shane Ross]

[Larry Asbell] “Then the venerable Shane Ross I believe misspoke and didn’t catch the fallacy in the Engineer’s saying the progressive frame has less information than a interlaced frame. Shane says “progressive footage means that BOTH fields contain information…the SAME information.” – No, No, No. (Still Love You, Shane!)”

Hey, I’m a creative editor, not an engineer. That’s why I stayed out of it after my initial post.

Shane

Littlefrog Post

GETTING ORGANIZED WITH FINAL CUT PRO DVD now for sale!
http://www.lfhd.net

[Arc Nevada]

Matthew

Canopus and other companies have filters that will make the fields look like frames. It does work. I think they interpolate the fields. FCP may have a decent film filter but I can not say for sure. That would be the better way to achieve the film look. The filters will manipulate the video to sync with the NTSC and PAL fields.

[Jack Joyce]

The precise meaning of progressive, and more importantly its implications, have never been clear to me.

Reading this thread, I see I am not alone.

I would also like to know precisely what pseudo-progressive means.

Does anyone know of an authoritative, plain-English explanation on the Web of these issues? Failing that, is there an engineer reading this who can recommend a book?

[Gary Adcock]

thank you larry,

you took the words out of my mouth….

ga

gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows
Inside look at the IoHD

[Arc Nevada]

Larry asbel,

That is what I was saying but I did not think it was important to get to technical about the scan line method. In the end the motion data is not there when converting because the interlace method relies on two fields that alternate as opposed to one single frame that is quick. They both have scan lines with the same amont of resolution data but when converting some of the data might get lost. Some programs will only use half the data of the full progressive frame for interlacing progressive video although there are many ways to do it. Others can interpolate the missing data to give a better image (Telecide). In the end I think we can all agree motion data is absent and quality could be lost when converting. The technician at the station was right. I never said interlaced video had more data or better resolution than progressive but when converting data can be lossed. I think everyone gets the idea now. Until my first post everyone was totaly lost.

[Jeremy Garchow]

oooooooo we have a hot topic in this thread.

[Jack Joyce] “Does anyone know of an authoritative, plain-English explanation on the Web of these issues?”

Larry’s post is about as concise as you can get.

Paulo, it sounds as if the engineer at the station has not been introduced to progressive video. You have to assure him that there’s just as much resolution and not a problem as progressive video is still stored in two fields, it’s just that the fields are created or scanned from the same moment in time instead of two different moments in time.

Jeremy

[Jeremy Garchow]

[arc nevada] “The technician at the station was right.”

I think he was proved wrong, actually.

[Mark Raudonis]

[Jeremy Garchow] “Jack Joyce] “Does anyone know of an authoritative, plain-English explanation on the Web of these issues?” “

Try googling “24p”. There’ ton’s of info out there. wikipedia is also a good place to start.

The misunderstanding of the “progressive frame” concept is only exceeded by the misunderstanding of the difference between “Drop and Non-Drop” Time Code.

Don’t get me started on trying to explain “1080 Progressive segmented frame”…

Mark