|
|||||||||
|
Thread Tools | Search this Thread |
January 22nd, 2007, 04:19 PM | #16 |
RED Problem Solver
Join Date: Sep 2003
Location: Ottawa, Canada
Posts: 1,365
|
Thanks for spotting that Gary. It was certainly my intention to soften the moving edges to eliminate some of the judder.
Graeme
__________________
www.nattress.com - filters for FCP |
January 22nd, 2007, 06:05 PM | #17 |
Wrangler
Join Date: May 2003
Location: Eagle River, AK
Posts: 4,100
|
FWIW, I did a quick read of the amo.net article and cannot recommend it for more than recreational reading. It definitely isn't a peer-reviewed scientific paper. It has a quite varied mixture of isolated correct facts, incorrect statements, fallacies of logic, and invalid interpretations. Well intentioned, no doubt, and written by someone who is obviously well educated, but in areas other than the human visual system.
That doesn't take away from use of tried-and-true shutter speeds for given frame rates to produce motion blur, or that different types of displays are perceived differently by the human eye and brain. The article just doesn't get you there.
__________________
Pete Bauer The most beautiful thing we can experience is the mysterious. It is the source of all true art and science. Albert Einstein Trying to solve a DV mystery? You may find the answer behind the SEARCH function ... or be able to join a discussion already in progress! |
January 23rd, 2007, 07:10 AM | #18 |
Regular Crew
Join Date: Feb 2006
Location: Los Angeles
Posts: 42
|
I read that article. He's got some of his facts and conclusions seriously wrong.
I have always thought that alot of 24p video motion looks less smooth than your typical feature film. As I mentioned before, many Varicam trainers are telling DP's to set their shutters to 200º. Most shooters just leave electronic shutters in the default setting unless they are going for some effect. They are assuming that 180º on an electronic camera looks the same as film. I think the combination of electronic edge enhancement as Graeme mentioned and inaccurate modeling of mechanical shutters is at the root of this. |
January 23rd, 2007, 08:52 AM | #19 |
RED Problem Solver
Join Date: Sep 2003
Location: Ottawa, Canada
Posts: 1,365
|
http://www.bbc.co.uk/rd/pubs/whp/whp...les/WHP034.pdf
Explains why video judders more than film. Graeme
__________________
www.nattress.com - filters for FCP |
January 23rd, 2007, 11:37 AM | #20 |
Major Player
Join Date: Nov 2004
Location: Canada
Posts: 547
|
Nice article Graeme.
Technical details aside, I can't help to question the accuracy of people's "such and such looks fake" statements. It's wrong on so many levels. First off, the world doesn't operate at 24 frames a second. So yeah - 24p is fake. It's fake on film and fake on video. In fact, 60i, 60p, 120p etc. etc. are all fake by that same standard. The objective of a film is never to look real - it's to present information. Secondly, unless all people's viewing is done in a fully calibrated movie theatre with a mechanical projector - they aren't seeing accurate time presentation anyway. In Europe everything is slowed down 4%, and in NTSC land, most of the time they see 2 frames than 3 frames, 2 frames then 3 frames if they have a good TV! If they have a lousy TV there are interlacing artifcats, superimposed images, improperly decompressed images, etc. not to mention colour calibration. The next gen of state of the art TVs might operate at 72 Hz or 120 Hz to eliminate this problem, but even then they can't be expected to work 100% of the time, thanks to incorrectly encoded flags and such. Next up we have the whole competence issue. My personal feeling is that with the advent of 24p camcorders, there is so much badly shot material out there that people are giving 24p video a bad name. Not only is it badly shot, it's badly handled in post, re-compressed multiple times, and then displayed on TVs and DVD players that add edge enhancement, scaling, and different colour spaces. People who are shooting on celluloid cameras know how to pan, frame, and hold an image at the low frame rates that 24p provides - they've been doing it for decades, and to be honest, they can't afford to do a lousy job due to the cost of the materials. A lot of celluloid flicks are carefully handled in post to maximize image quality, and if they're lucky, the quality exceeds the quality of the delivery medium. That said, I've seen a lot of lousy prints in theatres too. Watching 1080p trailers from Apple makes me question the point of a theatre, when the image quality is fuzzier than the 720p flicks make it out to be. It seems to me a lot of people have begun to associate blurriness and grain with the "film look". Not only that, but "edge enhancement" or "sharpening" is giving actual "sharpness" a bad name. One has ringing ugliness, the other has valid image detail. The other "film look" people talk about is colour representation, which I also find to be a joke. Firstly, a lot of films are colour corrected in the digital domain and then re-printed to film. So digital isn't the problem there. Secondly, basically any "look" can be produced by film, and equally by video, if the data is handled properly. There's no one film "look" there is an infinite set of possibilities. You want to talk about 24p looking fake, the discussion better be entirely about a shutters. If you want digital to look like film - make the Red camera, adjust the curves appropriately, selectively add grain, blur the image and throw some dust and scratches on it. -Steve |
January 23rd, 2007, 12:00 PM | #21 |
Major Player
Join Date: Nov 2006
Location: Victoria BC
Posts: 400
|
Just dropped in on this conversation, but I'm pretty sure it's over - Steve White just capped it all off - well said.
Question however - what do you mean by this? "Watching 1080p trailers from Apple makes me question the point of a theatre, when the image quality is fuzzier than the 720p flicks make it out to be."
__________________
Mac + Canon HV20 |
January 23rd, 2007, 12:16 PM | #22 |
Major Player
Join Date: Nov 2004
Location: Canada
Posts: 547
|
I'm just saying that when I go to the movie theatre and they're projecting film, I can see less detail in the projected film than I can see in the 1080p trailers. This is due to the quality of the film prints, as well as the projector and projectionist.
In going from a digital or a physical original stock to the film print that you see in your local theatre, a print has to be made. Due to the nature of film, unless a higher quality stock than the original, there is going to be a loss in quality in this process. This degree of quality loss may or may not be relevant, however you can bet that the version used to compress the digital version is high on the chain of prints made - if not the original digital version. Sometimes the print at a theatre is just dirty, other times it's a low quality stock. It could be old, scratched, or simply out of focus. Sometimes the Xenon bulb is aged and its colour spectrum suffers. All these serve to degrade the image you see in a theatre. My experience has been that a properly compressed 1080p video on the Apple site very often exceeds the quality of the print I see at the cinema. -Steve |
January 23rd, 2007, 12:46 PM | #23 |
Major Player
Join Date: Nov 2006
Location: Victoria BC
Posts: 400
|
I see, got it. I've seen that too.. watch the 1080p trailer just before I go, arrive there and point out to my friend/guest that "whoa.. man that looks fuzzy.."
__________________
Mac + Canon HV20 |
January 23rd, 2007, 01:14 PM | #24 |
Trustee
Join Date: Dec 2006
Location: Little Rock
Posts: 1,383
|
This thread reminds me of a time many moons ago when I purchased my first 16mm camera, an Eclair NPR.
All the 35mm guys gave me crap about shooting low res 16mm format. ..."The same format the original posted touted as his format of choice." |
January 26th, 2007, 11:41 AM | #25 |
Major Player
Join Date: Apr 2004
Location: san miguel allende , gto , mexico
Posts: 644
|
The difference is that the 24 frames of film are each projected twice with a standard 3 blade shutter on projectors. This smooths the motion of film in contrast to the 24p projected material of video. Maybe we need 48p to actually see a more true film motion effect.
|
January 26th, 2007, 04:15 PM | #26 |
New Boot
Join Date: Mar 2004
Posts: 10
|
48p - or better yet...
Yes, Kurth's last post makes sense, but since the film projector just holds the same frame in place for two blinks of the shutter, it wouldn't take 48p and extra bandwidth to replicate the effect. Couldn't someone like Graeme just write a plugin for FCP or Quicktime that would create a projector shutter effect out of 24p material?
|
January 26th, 2007, 05:30 PM | #27 |
Trustee
Join Date: Dec 2006
Location: Little Rock
Posts: 1,383
|
My old telecine projectors used a 5 blade shutter.
|
January 27th, 2007, 10:47 AM | #28 |
Major Player
Join Date: Apr 2004
Location: san miguel allende , gto , mexico
Posts: 644
|
Yes , a telecine projector uses a 5 blade for ntsc material but a normal projector as used in any theater uses a 3 blade.
|
January 27th, 2007, 10:59 AM | #29 |
RED Problem Solver
Join Date: Sep 2003
Location: Ottawa, Canada
Posts: 1,365
|
If the multi-blade shutter is the key to 24p smoothness in the cinema, why does 24p from film look smooth when it's projected in the home on a DLP projector?
Graeme
__________________
www.nattress.com - filters for FCP |
January 27th, 2007, 03:12 PM | #30 | |
New Boot
Join Date: Jan 2007
Location: USA
Posts: 9
|
Quote:
|
|
| ||||||
|
|