I've done some online searching but haven't got a handle on this one yet.

How does 24/25p in a 60i/50i stream/wrapper work. Are the progressive frames halved into 2 fields and written/stored as 60i/50i (adding pulldown with the NTSC framerate)?

Is motion any smoother or does that not play into it as the 2 fields are recorded at the same point in time?

Sorry for the newbie question.

The wikipedia article for pulldown will get you most of the way:

Telecine - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/2:3_pulldown)

Thanks Chris.

I understand the concept of pulldown removal.

What I'm curious about is how 25p in a 50i stream works (I record in PAL framerates). It's not real 50i like the GH2 (with 25 frames divided into 2 fields, each representing a different moment in time). So are each of the frames halved into 2 fields? Does this affect how motion is seen on an interlaced TV (making it less studdery)?

Sorry about that. I should have been more specific about what I was recommending you check out. Part of the page there has info directly relevant to your questions. Since you stated you were a newbie I didn't want to assume too much.

The wiki page under the section "Frame Rate Differences" shows how a frame of film (progressive media) is divided into 2 fields and pull-down added so it can be embedded into an interlaced stream. Depending on whether you are shooting PAL or NTSC it shows the differences in pulldown for the different standards. That is the answer to one of your questions. Yes, progressive material is line divided and then treated (stream wise) as interlaced material for transmission. If everything works correctly the content flag identifying the material as progressive is set so this fields can be correctly recombined when the images are displayed.

As far as the question whether converting progressive to interlaced increases judder the answer that I can give is that I've not noticed an increase of judder on modern HDTVs.

BUT

Modern TVs have the ability to read the content flags embedded in the stream and know if the source material was derived from progressive and/or if pulldown needs to be removed. If so the display will correct any of this stuff so the material is displayed properly. I have noticed a good bit of judder on older tube type TVs that lack any signal processing for pulldown removal or progressive display.

Thanks Chris.

I understand the concept of pulldown removal.

What I'm curious about is how 25p in a 50i stream works (I record in PAL framerates). It's not real 50i like the GH2 (with 25 frames divided into 2 fields, each representing a different moment in time). So are each of the frames halved into 2 fields? Does this affect how motion is seen on an interlaced TV (making it less studdery)?

It should look identical to 25p played on a progressive monitor. Splitting the progressive frame into fields is invisible for a playing file. Any motion artifacts (judder) are the result of shooting in 25p which is essentially half the frame rate of 50 interlaced. This is why sports are usually shot with interlaced equipment. The motion is clearer although each frame is half the resolution of the signal.

Thanks guys for your help.

I have to admit though that I'm still a little confused.

If you're able to bear with me, what's the difference between "25p wrapped in 50i" footage captured by the GH1 and true "50i" footage captured by the GH2? If 2:2 pulldown is added to the GH1's 25p footage shouldn't the results be as smooth looking as the true 50i footage captured by the GH2 (considering both contain 25 full frames - and both involve shooting at 1/50 sec shutter speed).

You see i'd actually prefer to be shooting 50i on my next project as it will be more run and gun style and I don't want too much jerkiness. Plus, my delivery format is 50i (for HD broadcast).

Thanks for your patience!

I think I found an answer to my question on another forum:

"In normal interlaced PAL footage (50i) the camera shoots one field every 1/50 second. Basically the camera shoot field A followed by field B then another field A followed by B etc. Field A and B together makes the full frame (25 frames each second)

In progressive PAL footage (25P) the camera does exactly the same thing as in interlaced, the only difference being that it captures both A and B fields at the exact same time instead of first A then B. You could say it shoots 25 frames per second which then are divided into two fields (50 fields per second) in order to be compatible with the PAL video standard."

So since the the fields are captured sequentially in native 50i, whereas they're captured at the same time with 25p wrapped in 50i, the motion of true 50i should be a lot smoother. Hmmm ....

I'm not sure why you would not just shoot in 50i if that is your end product?

The jerkiness of 24p or 25p is not the result of the the footage being inherently jerky, it is the result of not using proper camera technique to fit the frame rate. If the footage is jerky it was not shot correctly. If it is smooth, the proper camera technique was used.

Yes it will still be jerky, of course whether it is is a wrapper or not, if you do not shoot properly.

When the FX1000 came out everyone was trying the 24p or 30p in a 60i wrapper, and then complaining how jerky it was. The wrapper's purpose, I don't believe is to smooth it out, but to make it editible for amateurs. I could be wrong on that, but that's my suspicion.

Anyway, for broadcast purposes your focus should be on minimizing issues,IMO. Just my 2 Cents, again.

What you are hoping for Dave is 25p footage that doesn't have the "negative" characteristics of 25p, or less of it, and that when it's in a 60i stream it should be smoother.

What I'm trying to say that the only way to "smooth" it out is to take away the very same characteristics that you need for the film effect. If you need smooth footage but are shooting documentary style, or run and gun style, then 25p is a poor choice regardless and it should be skipped. I think I'm on track here, but someone correct me wherever I'm mistaken.

Yeah Jeff you've got it nailed. So far I've mainly used the GH13 locked off (except when I'm using it to film family and friends). I've used both native 25p and 25p in 50i wrapper, and to be honest couldn't tell the difference looking on my 21" LCD.

I kinda assumed 25p in a 50i wrapper consisted of 2 fields recorded in the same instant, and so the same rules of filming in 25p applied to 25p in a 50i wrapper. I just wanted to double check.

Like you advised, I would certainly use 50i to film run and gun stuff but of course the GH13 doesn't give me that option. And I would have bought a GH2 but I need the ability to switch from NTSC to PAL, so it doesn't work. Let's see what Vitally comes up with eh?

While I appreciate that many people use the GH1/2 etc for the film look, 25p motion judder is not the biggest turn on for me. What i do appreciate about the GH1 for the work I do is the tiny size, the sharpness, the exposure latitude, the higher resolution of progressive frames, and of course the shallow dof.

Bring on 1080p50!!

I'm using 720 60p, which doesn't have the resolution of 1080, of course, but like you I have different cams with different abilities, so I'm kind of stuck with 720p. Not that I'm complaining, I love 720p, it is SO easy to work with, and it converts to SD widescreen perfectly from Vegas' timeline, where 1080 I need to resize and it just gets more complicated. So 720 60p is truly perfect for me anyway, really. And of course for the web 720 is optimal, as Vimeo only does 720 anyway for HD, so I am fortunate is a way to be forced into 720p.

While I appreciate that many people use the GH1/2 etc for the film look, 25p motion judder is not the biggest turn on for me. What i do appreciate about the GH1 for the work I do is the tiny size, the sharpness, the exposure latitude, the higher resolution of progressive frames, and of course the shallow dof.

That's the reason I finally decided to not shoot 24p. It's so jittery, especially when I watch it alongside 30/60p. And I really don't see it sticking around much longer, as it by no means gives the instant "film look," and seems to only be around so that people can say that they shoot in 24p. I was only considering it because of the GH1's inability to do 1080/30p. I was even ready to go through pulldown removal on my unhackable GH1.

So 720 60p is truly perfect for me anyway, really. And of course for the web 720 is optimal, as Vimeo only does 720 anyway for HD, so I am fortunate is a way to be forced into 720p.

I love 720p, and am grateful that you helped me come to terms with it :)

Now that I've been working with it for a while, it has really grown on me. The only thing that suffers are screen grabs for the packaging art. But I manage to work with them enough that it's not really an issue.

I've never tried 720p50 as I deliver in 1080 for HD broadcast. I need to check out if the sacrifice in resolution is worth it for the smoothness.

What patch do you use for 720p60 Jeff?

Is the final prouct broadcast in 1080? Here some networks broadcast in 720p.

Yup. 1080 50i.

Oh well, what can you do....

Corey, your post from a few days ago just showed up in my email box, some kind of strange delay. Anyway, I'm glad you have found 720p acceptable for your work.

We would certainly all prefer 1080p, but the option is just not there with my gear right now.

Now in defense of 24p, the guys that know how to use it can make it sing. Think Pacific Pictures, etc. The best known Cincinnati videographer uses it and his stuff looks perfect, no issues, and actually a lot of guys are using it across the country. The problem is the vast majority of us don't take the time to learn how to shoot for it, or they are using it for the wrong applications, and then it looks bad and they think 24p must be crap.

I'm just tickled to be shooting in HD in a format that is relatively easy to work with and looks great.

1080 50i can be thought as 540 50f (yes, I know it effectively isn't but that is the resolution of a single field) so actually 720p has a higher single frame resolution as it's a progressive format. 720p up-rezes to 1080i very well. Many high end cameras can now record 1080p (and above). Interlacing will become a thing of the past. It has it's roots in 1940's broadcast television.

Thanks William. I need to do a test upresing 720p50 to see how I like it. What patch do you use to film 720p?

I'm using 720 60p, which doesn't have the resolution of 1080, of course, but like you I have different cams with different abilities, so I'm kind of stuck with 720p. Not that I'm complaining, I love 720p, it is SO easy to work with, and it converts to SD widescreen perfectly from Vegas' timeline, where 1080 I need to resize and it just gets more complicated. So 720 60p is truly perfect for me anyway, really. And of course for the web 720 is optimal, as Vimeo only does 720 anyway for HD, so I am fortunate is a way to be forced into 720p.

Jeff, can you share how do you bring it to DVDA? Do you render your 60p timeline to MainConcept MPEG-2? What setting?

Alex, it depends, are you talking about internet, or bluray or DVD?

I've got it down for Bluray and internet now, but for DVD I don't have the answers as I thought I did, as you can see from another thread I brought up. My current DVD project I've encoded to interlaced and haven't seen how it plays out yet, as I'm still edtiting.

It is for a DVD. So far I tried a standard NTSC template (29.970i), with bad artifacts. Tried 24P- looks clean, but too much motion jerkiness here and there...

I made my last one by turning the mpeg 2 DVDA Widescreen template for rendering into progressive, and it worked beautifully, but I have been told this is not the right way to go about it. I am giving the whole subject a rest until I'm done edtiting. Check out my other thread for SD DVD in this forum.

I think changing to 24p out of Vegas is definitely NOT the way to go with 720p footage, however.

Now in defense of 24p, the guys that know how to use it can make it sing. Think Pacific Pictures, etc. The best known Cincinnati videographer uses it and his stuff looks perfect, no issues, and actually a lot of guys are using it across the country. The problem is the vast majority of us don't take the time to learn how to shoot for it, or they are using it for the wrong applications, and then it looks bad and they think 24p must be crap.

For sure. I've shot in 24p for things and they've turned out great. It's just definitely not for the weddings I shoot. And once I get my mitts on some good 1080/60p equipment, I'll never even think of 24p again for any application (unless it's some sort of delivery requirement).

Alex, essentially, this is what I've been told to do by David Newman of Cineform, if I understand him correctly.

Using Cineform, I re-encode the footage to 16:9 SD progressive, then edit on a 30p timeline in Vegas, and render when done using DVDA mpeg 2 Widescreen template, but change it from interlaced to progressive. He says progressive DVDs play fine, and that he's made hundreds of them.

Alex, this is the latest, changed to fit you if you are not converting footage.

1. Start a new project using 16:9 DVD type settings but change interlaced setting to progressive. In other words, field order should be "none", not upper or lower field first.

2. Bring footage into edting program, and if you want to, then crop the footage to fill the screen. When you first bring it in it will have very small black bars on the sides, cropping will fix that. This is optional and up to you. If you use Vegas and need to learn how to crop your footage, let me us know, there is a simple script that will do it with basically on click.

3. Edit footage

4. Render out to 16:9 for DVD, using correct template in your editing program. Make sure to render to 30p, to match your project settings.

Done.

Jeff, I have been doing this last few hours, and was about to post a question to your previous post, why to use Cineform? Seems to be an extra step. Direct encoding to MPEG-2 30p does the trick as the interlacing artifacts are all gone, and video looks pretty smooth, better than the same at 24P. But just about when I was getting ready to post, I run into a puzzling problem. The MPEG-2 file that was rendered from the timeline came in at just over 3GB, with plenty of room under 4.7GB DVD limit. The same file when brought into DVDA 5.2 shows as 5.7GB (!) way over the red line. Tried the same with my older DVDA V3 and got the same results. There was something in 30P file that confused DVDA and caused this problem. I decided to prepare the DVD anyway thinking it was just a bug in reporting the file size. Unfortunately DVDA spent about 40min doing something (may be duplicating the frames) and ended up with about 5.7GB DVD folder.
Then I re rendered the same timeline to much higher compression setting, got a small ~2GB MPEG-2 file which DVDA had no problem using and reporting correct size. So my last test which I just finished, was splitting the MPEG-2 file into 2, bringing them to DVDA as to separate files and preparing the DVD again. This time the visual quality was good as the bit rate was kept around 4.2 in average, and the total size fit the DVD at about 4.4GB.
I wish somebody could explain what exactly is happening when using 30p for DVD. I suspect that DVDA treats each frame as a field and duplicates the same frame 2 times.

Alex, I've been going for almost 24 hours, so I need to keep this short, just got back from a wedding, long day.

However, regarding Cineform, I need an intermediate codec because I am edting 4 cameras at a time, so the raw footage is too difficutl for me to get smooth playback when editing. And since the current projects I'm working on are going to DVD and not bluray, Cineform is just what the doctor ordered.

Regarding your issues with your DVD,be sure to use the DVDA template as shown for rendering, with the only change to it being progressive. If your project is over 90 minutes or so it will not fit on DVD unless you render after adjusting the bit rate in Vegas.

You did not say how long your project was. And if DVDA is recompressing, you definitely (I think) did something wrong in your render, at least that is my thought. If your file is not DVDA compliant, then DVDA will recompress it, which is not good. That' why you must use the correct DVDA template in Vegas for rendering, to keep it compliant and to avoid recompression.

I have not used this workflow all the way through as you have, so I suppose I could run into the same problem but I hope not.

Jeff, I am not doing it that often, my current licensed Vegas installation is V6/DVDA V3 :-) (I am using trial downloads for this project) so I might be way behind the times, but I never heard of a 1.5 hours rule. This project in fact is 1 hour 31 minutes long. I believe the only thing I changed in the template was the frame rate to P as you said. When DVDA showed almost 6GB size and started processing it, it did not just render I assume, as it would be several hours to render this file. It took about 40 minutes, seems to be it was changing something in the file without doing heavy duty work of recompressing.
After all that fiasco, I sliced the same MPEG-2 file in two halfs, and the size problem went away, the program fit into the 4.7gb limit and looks fantastic - clean and no judder.

Alex, the 90 minute thing isn't a rule so much, that is just "about" what will fit for a DVD, depending also on the size of the audio file. You can fit much more on a DVD, but the bit rate must be lowered accordingly.

I highly doubt progressive files are larger, but I don't know. My last project was progressive, 90 minutes and fit perfectly from Vegas, without any recompression etc by DVDA, so I'm not sure why your files are so large. Something happening, can't imagine what, wish I had an idea for you.