Panasonic to Preview AVC-Ultra (at NAB 2012)

[Amr Rahmy]

making uncompressed and compressed file formats is not difficult. it's quite easy actually.

but why would you not use a flavor of prores or dnxhd or avcintra.

for one thing, prores is easy to encode, easy to work with on older machines, they made an acceleration engine for it in fcp, it's easy to scrub through, fast forward/rewind, slow playback, go frame by frame.

 

[Garion]

AVC-Intra 100 is close. It meets 2, 3, and 4, but not 1 (commodity media). Surprisingly, it meets 2 (1 GB per minute) on P2 cards, even though its bit rate is not 133, because of some kind of redundancy to aid reliability.

The workflow I picture is: take card out of camera, copy to computer for editing, put card on shelf. It's hard to afford to do this with media that is $650 per hour. You also basically have to mail order it. You can't just pick up some at the grocery store. Is P2 a wonderful format? Of course. But not quite as sweet.

Actually, it turns out that ProRes 422 at 24 FPS is 133. So, maybe that would be good enough for me. You could get higher quality at that same bit rate, though, by using newer compression, like wavelets or MPEG-4.

I do something similar, i ingest my P2 footage into Edius and dump onto a raid array on pc. For backup i have an external sata caddy base where i can plug in a bare hard drive, dump files to hard drive and put drive in static bag on shelf.

Theres no reason AVCintra couldn't have been stored on SDHC etc as its only approx 10 MegaBytes/sec but iirc the P2 cards are actually raided internally to help prevent data loss.. but i agree it would have been nice to have had a cheaper form of internal storage though.

 

[Garion]

making uncompressed and compressed file formats is not difficult. it's quite easy actually.

but why would you not use a flavor of prores or dnxhd or avcintra.

for one thing, prores is easy to encode, easy to work with on older machines, they made an acceleration engine for it in fcp, it's easy to scrub through, fast forward/rewind, slow playback, go frame by frame.

I think they didnt go prores for two reasons, licensing and the fact that we dont all use apple macs. I can easily scrub around and play with multiple AVCintra files in edius no problems.

It was said earlier, all the big companies love their own codecs and make money through selling licenses to others, pain for us though.

 

[Amr Rahmy]

i'm not talking about panasonic, why is it a pain to you? choose your nle and work with what that nle can accept or is recommending.

 

[Garion]

Sorry Amr i slightly mis-read your earlier post thinking you were advocating pro-res. I meant a pain only in the sense that some nle's dont get on as well with some codecs over others.Personally i only deal with AVCintra so i dont care but i have friends who have to accept a myriad of differant formats and they would like some conformity.

I hope Panasonics announcement at NAB is worth the wait, they are remarkably quiet compared to their compeditors.

 

[Everts]

I hope Panasonics announcement at NAB is worth the wait, they are remarkably quiet compared to their compeditors.

Yes awfully quiet...I have been checking this thread everyday but nothing yet....I wanted pana mojo doing its thing with a 35mm chip !

 

[combatentropy]

But this fits only your workflow. Shoot on sd card and keep the card.

Some shoots could go hours and hours. And for others, its basicaly the same in commodity media with p2 where you record and dump. Only difference with it is p2 is much pricier but much secure.

I agree with your points And i would love to shoot avc-i on a cheap media.
t your concern is tailored to your workflow htough Which is not everybody's

You have to understand I am in perfectionist mode, just taking the chance to dream a little.

Second, I don't believe "what's good for me may not be good for you." I think that's true to some extent, but not nearly as often as it is said. Don't get me wrong, I'm not going to try to force you to change.

But I offer that my workflow is best for everyone. It's safest to use a card just once. The copy-to-hard-drive-then-erase-original-P2-card introduces another chance for a mistake. For someone in a hectic job, like news, the risk is higher. If the risk were necessary, I would just accept it.

P2 cards were introduced a long time ago, when solid state recording was just catching on. The rugged casing and RAID recording are nifty, especially if you're reusing cards. But SD and CF cards are twenty times more reliable than tape. P2 may be more reliable than they are, but is more reliability needed, and is it worth the extra cost? Suppose I introduced a format called P3, which was twice as reliable as P2 but also twice the price. Would it be better to use that? At what point would paying twice as much for two times more reliability become a waste of money? And the ratio is actually worse than that. You're paying 5-10 times as much for perhaps 1-2% better reliability.

If you think the question is academic, it's not. P2 may guard against write errors, but don't neglect other factors that may cause you to miss a shot. For example, not carrying enough cards could cause you to miss a shot. And you could lose a chance to reuse archival footage if there was an error copying the P2 card to a hard drive. "But the chances of those happening are rare," someone might say. And so are write errors to SD or CF cards.

The best design is the one that best balances all needs. To overdo one aspect often causes you to underdo two or three others. When some people dream, they dream of unlimited resolution, unlimited dynamic range, unlimited etc. When you talk about how everything is actually a trade-off, it makes them a little glum. But I think it's actually more exciting to work with real-world constraints. With a little time and thoughtfulness, you can find a design that gives you a little bit more than another, even though they both deal with the same physical realities. And I think such a design is actually more beautiful than the purely imaginary one that refuses to deal with present-day constraints.