.comment-link {margin-left:.6em;}

High Definition Video for Independent Filmmakers
A How To Guide for Digital Filmmakers
Welcome all! This is my blog to share my latest research,
thoughts, etc. on utilizing HD for independent filmmaking.

YES, I am available for consulting
Contact me at mike@hdforindies.com

All content copyright 2004-2007 Mike Curtis.

Monday, May 16, 2005

H.264 - a little perspective from Mike 

I did a little bit more testing with H.264 over the weekend - 30 megabit doesn't seem to play reliably, even on a dual 2.5 GHz G5 with an ATI X800 card for 1080p24 content. Some 20 megabit 1080p24 didn't play back smoothly in certain sections. I don't know if it was because it was at the beginning of the clip, or if the footage (confetti thrown into an open sky, tons of tiny details) was more complex to play back. But there are definite limits to what H.264 can do. 20 megabit 720p60 footage of a guy running around (from my camera test) played back OK. More details later when I can do some real testing with Compressor 2.0. I hope to pick up my copy tomorrow.

Thinking some more, after having a few email conversations, H.264 is just starting. I'm sure we'll get better playback and better encoding as time goes by. Even if that encoding is done elsewhere, not as built-in OS X/QT 7 encoding. Just like there are sophisticated MPEG-2 based encoders that deliver the best quality for commercial DVDs, the same thing will occur for H.264.

While the quality of H.264 is miles beyond MPEG-2 (even comparing prosumer (QuickTime 7 built-in encoding vs high quality MPEG-2 encoding), the workflows will probably be quite similar - decent results with consumer/low end professional tools (and I'll consider Compressor a low end professional tool, as compared to what Universal/MGM/Disney will use to compress their titles), and great results with high end tools. High end tools will also offer realtime encoding, which will matter since we're now dealing with a much more calculation intensive encoding process, and dealing with up to 6 times as many pixels as standard definition.

The interesting question will be whether we can get "good enough" looking video within the data rate limits of the high def DVD specs using the low end professional tools. If your movie/video/doc/content looks good at Compressor capable data rates, and will play back reliably on the still unshipped HD DVD/Blu Ray players, and FITS on whatever media you're authoring on, you're golden.

As for professional tools, we'll need them - H.264 is VERY slow to encode. Miles slower than MPEG-2 it would seem. Tomorrow I'll start being able to do some real tests I would imagine, if Final Cut Studio ships as expected. I want to see how good the HD MPEG-2 encoding is, and how long it takes to do.

Don't forget that at the moment, H.264 is for progressive footage only from what I can tell. 1080i footage needs to use HDV/MPEG-2. Can we make 1080p24 H.264 discs that play back at 1080i? That's a very, very interesting question. I can probably find that in the online docs, but I need to look it up. Anybody know yet? Post in the comments field please if so.

-mike
Comments:
Mike, I don't know why you wouldn't be able to store 1080/24p content that would play back at 1080/30i. Isn't that just exactly how standard-definition DVDs work already? The content is stored as 480/24p and pulldown is inserted on playback. I'd be pretty surprised if that same feature were omitted from HD-DVD considering that practically every SD DVD uses it.
 
It could, it should, but does the spec definitively allow it? I can't imagine it wouldn't, I just don't know that it does for a fact. If the format allows it, will the tools we have handy (such as DVD Studio Pro 4, Compressor 2.0) allow authoring to that as well? Again, should, but don't have proof/verification that it does.

-mike
 
Post a Comment


Links to this post:

Create a Link

This page is powered by Blogger. Isn't yours?

Listed on BlogShares