Best technique for getting 24p footage (Just checking)

[DouglasSteinberg]

Hello,

I recently got a used hv20 (in good shape). I have been reading this forum tonight and wanted to check if this is the best technique for getting my footage into 24p (I actually haven't tried yet).

1. Capture clips with HDVSplit. Each clip a separate file.

2. import clips into After Effects and guess pulldown.

3. Drag those clips into separate comps.

4. Either open that After Effects file in Premiere if you have dynamic link, and edit (probably the preferred method, no?) or render to a file using Lagarith codec.

Is this still considered the best way, or have things changed and there is now a better way?

Just checking before I try it out.

Thanks!

[1]

More descriptive THREAD TITLES please!

[DouglasSteinberg]

Sorry, is there a way to change it?

[Lunchbox]

I changed it for you. If you want more people answering your question and involve in the discussion, you should come up a more descriptive title.

[DouglasSteinberg]

Thank you very much. I can understand what you mean. It is better to be as descriptive as possible. From now on, I will try to give more information about the thread in the title.

Regarding the thread itself, I am just wondering whether there has been any significant changes in workflow with 24p in the past year or so, since some of the descriptions seemed to be dated a while ago.

Thanks again, and sorry for the non-descript title.

[DouglasSteinberg]

I can only assume a few reasons why no one has answered this post:

1. They've been raptured.

2. Everybody's bought a new camera and no one uses these anymore.

3. This technique is still valid.

Hoping it's number three...

[wildside50]

Yes, this site is frankly best used for reference and shooting the breeze at this point. New questions regarding 24p workflows etc are probably best put up at dvinfo.net or something more general, since most consumer camcorders still use the 24p in 60i stream method so the Sony, Panasonic and most canon guys have all come together to answer this question.

But I'll give my stab anyway. Yes, that is still a very useable workflow (the one I use anyway). If you wanna buy NeoScene and convert it on the fly to 24p in the cineform codec, that's another popular method. But you're fine using AE if you've got it. You should get fine results as long as the clips you're using aren't really short, in which case AE is going to have trouble guessing the cadence.

So while many have moved on from these cameras, the need to remove pulldown in 24p is still alive and well, unfortunately.

[ejolson]

Just as there is no best kind of computer and no best video editor, there is also no best pull-down removal technique. Although the HV40 records native 24p, this forum continues to have active discussion about pull down. Even though dvinfo and dvxuser are named after video standards and camcorders which are even older than the hv20, these forums are also active and contain useful information. Of particular interest here is the HV20 forum-member developed utility

http://www.hv20.com/showthread.php?3...lldown-Remover

[DouglasSteinberg]

dvinfo.net

I will look into those other forums you mentioned.

I see that you use After Effects for your workflow (and HDVSplit?). Question: Do you render it to another codec or leave it and import the comps into Premiere?

If you wanna buy NeoScene and convert it on the fly to 24p in the cineform codec, that's another popular method.

I see. Is the Cineform codec superior to all other codecs? Would you recommend using it? Is there a loss in quality in transcoding it to Cineform?

Of particular interest here is the HV20 forum-member developed utility

Thanks!

[ejolson]

Rumor has it that quality is improved by transcoding to Cineform. This improvement may be related to good 4:2:2 color space interpolation as well as the naturalness of the digital filtering that happens automatically when using wavelets to compress an image. Never mind that. I have been happy using dnxhd at 145 mbps for my intermediate files instead of Cineform.

[DouglasSteinberg]

Thanks for the info. What is dnxhd? Where does it come from?

I haven't shot any footage yet, but when you capture it with HDV-Split, does it make it with an "HDV" codec?

What would happen if you re-encode it with HDV from within After Effects (if you can do that?)

[cgbier]

DNxHD

[drapeama]

I have been happy using dnxhd at 145 mbps for my intermediate files instead of Cineform.

Sounds great...how big are the files after?

I haven't shot any footage yet, but when you capture it with HDV-Split, does it make it with an "HDV" codec?
What would happen if you re-encode it with HDV from within After Effects (if you can do that?)

1) When captured with HDVsplit, you'll get a .m2t video file mpeg2 video and audio, all in high-definition with a 25mbps constant bitrate...HDV isn't a codec, it's a standard.
2) I've done it several times without seeing a difference: Capture using HDVsplit, color correct the videos in After Effects and render as Mpeg2 CBR 30mbps in a 1920x1080 resolution at 23.976fps (very similar to the HDV specs). The results are just as great as the original file and it doesn't waste much of your HDD space.
You should try it.

[ejolson]

HDV files expand by a factor of about 6 when using DNxHD at 145 mbps as an intermediate. It also makes sense to remove pull down, color correct, export back to HDV 24p native if you can smart render from your editor afterwords.

[wildside50]

To the OP:

Codecs my man -- welcome to a world of hurt. My honest suggestion if you haven't shot any footage yet? Just go shoot. Don't worry about things like intermediate codecs, rendering settings, native 24p vs pulldown removed 24p, workflows etc etc, or you'll quickly lose months to research instead of just shooting.

If you're hell bent on getting native 24p footage to edit, then yes, just go ahead and use dynamic link and AE. It'll do ya just fine I'm sure. HDV is pretty easy to work with natively (as compared to, say, 60p MPEG4) on most modern PCs, so until you really need to, don't worry about intermediate files that are 50-100 GB per hour. Just shoot my good man -- just shoot.

[drapeama]

[...] Don't worry about things like intermediate codecs [...] If you're hell bent on getting native 24p footage to edit [...] don't worry about intermediate files that are 50-100 GB per hour.

1) Totally agree. I did some tests last year and I came to the following conclusion: rendering from HDV to 1920x1080 mpeg2 CBR 30mbps is more than enough/needed. It's just a bit better than the original HDV files: more resolution and more bitrate. The final result is just as great as the original captured file. I tried this for my contest #19 entry (The Godfather) and I saved so much time rendering mpeg2 compared to lossless AVI (tried that too before and doesn't look better when compressed to bluray/AVCHD/DVD/web) and so much space.
2) About editing with native 24p: if you need to remove the pulldown just be sure to capture your footage with scene selection, otherwise After Effects won't be able to guess it properlly. Guess it, set your comp to 23.976fps and then render your corrected footage and you'll be able to edit in 24p in any NLE. Don't waste your time using the Dynamic Link between After Effects and Premiere (if using Premiere to edit): last time I tried it took all my available ressources on my PC and I was only color correcting some clips in AE. Best option, open in AE, color correct and render with the preset I suggested: then it's as smooth as editing HDV, and you're ready to export your final video out of your NLE. I use After Effects mainly to color correct my footage as it allow to do pretty much everything compared to other editing software.
3) Absolutely! It's just a waste of HDD space: your camcorder capture the image and record it on tape in a compressed format...what's the point of editing it in an uncompressed format? Your footage won't look better. Think about it: take a DVD (mpeg2 compression) and "capture/rip" it in an lossless format: will it look better when you'll render it after?

[ejolson]

Intermediate codecs are designed to preserve the original image quality over multiple decompress and compress generations. Multiple generations with mpeg2 will continually amplify macro blocking and other artifacts. Hopefully the discussion in this thread and elsewhere on how to remove pull down will help others avoid months of research. That's the idea, anyway.

[drapeama]

Multiple generations with mpeg2 will continually amplify macro blocking and other artifacts. Hopefully the discussion in this thread and elsewhere on how to remove pull down will help others avoid months of research. That's the idea, anyway.

1) Multiple, yeah. I'm talking about rendering one generation before compressing to h.264 or DVD, which probably don't make much difference.
2) Yeah, months of research!