h a l f b a k e r y"More like a cross between an onion, a golf ball, and a roman multi-tiered arched aquaduct."
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
When a 24fps movie is shown on a conventional 60 field/second (30 frames/second) television set, every fifth frame must be shown twice to keep the presentation at the proper speed. When shown at 60fps, half the frames need to be shown twice and the other half thrice. This can introduce slight, but
noticeable, jerkiness on some movies.
I would propose that a 72fps video standard would allow movies to be shown without any additional jerkiness. If the number of scan lines per frame were reduced to remove the top and bottom "blank" portions when playing anamorphic DVD's, the horizonal scan rate would even be about the same (or could be made the same) as with the existing standards.
Additionally, if a 24frame/sec 3:1-interlaced (72 field/sec) standard were introduced, most conventional television sets would be able to handle it except that (1) the vertical sync circuitry might need to be tweaked, since they'll generally accept out-of-tolerance video signals, but not that far out; (2) Closed-Captioning would likely not be usable in the set, so if desired it would have to be added to the DVD player [DVD's generally have their own captions, but some only use line-21 Closed Captions].
The same conclusion
http://www.goodwins...end.com/library.htm
"All other things being equal, video that was transferred from film will tend to look better at 72 fps than at 60 fps." [bristolz, Oct 07 2002, last modified Oct 05 2004]
3D Fusion DVD Player
http://www.cyberthe..._Fusion/3d_dvd.html
Examines 72 fps playback from this device, in detail. [bristolz, Oct 07 2002, last modified Oct 05 2004]
Sony 24P system
http://www.whites.c...ort/document_1.html
24 fps progressive video. [bristolz, Oct 11 2002, last modified Oct 05 2004]
TI's excellent DLP info site.
http://www.dlp.com/
Bring money. [bristolz, Oct 11 2002, last modified Oct 05 2004]
A 24p TV for your home.
http://www.dvinside...ews/news.asp?ID=440
Reasonably priced, even. [bristolz, Oct 11 2002, last modified Oct 05 2004]
Maxivision 48
http://www.geocitie...303/maxivision.html
48 fps film format [ConsultingDetective, Oct 04 2004, last modified Oct 05 2004]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
Baked. Put a DVD drive in a PC and feed the video signal to a projector at 72 FPS. (Granted, a PC costs significantly more than a DVD player, but if you're posting here, you probably already have one...)
Croissant, though, because it would be nice to see a dvd-size/price box pre-made to do this. |
|
| |
Does anyone know of any home-theater-targetted displays and DVD players that can operate at a vertical scan rate of 72Hz? |
|
| |
Define "home theatre." For those with a lot of money and can afford very expensive p-scan Barco projectors, you can use a media PC with 3DFusion DVD and Mpact 2 to drive 24 fps material at a 72 fps display rate. There are people I work with who have such systems at home. It is not uncommon for these theatres to represent a fat 6 figure investment (however, I can think of much better ways to spend money). |
|
| |
Media-outfitted PCs are finding there way into the home theatre more and more these days. |
|
| |
72 fps was kicked around a lot as a ATV/HD standard frame rate during the HD standards wars a few years back. It certainly isn't a new thought. A couple of groups of people really liked the idea: film geeks, for the pulldown reasons you mention and video editors who very much liked the idea of having 72 edit points per frame for creative reasons. |
|
| |
Film geeks and preservationists further liked the idea because they feel that the standard should favor film as culturally more valuable than TV. |
|
| |
bristolz: Any idea whether any consideration was given to the notion of having the 72Hz vertical scan used with the existing horizontal scan rates? |
|
| |
Also, while a 72fps VGA may be in some ways better than a 60Hz one, unless the "72Hz" display is exactly 72Hz and genlocked to the video source (usually not the case with computer displays) there would still be some rate-conversion artifacts. For example, if the frame rate on the computer is really 72.2Hz, every five seconds there will be 119 frames shown 3 times each and one shown 4 times. Probably not too noticeable in most cases, but in e.g. the Beatles' _Yellow Submarine_ song "All together now", the some of flashing effects may have a visual "glitch" every 5 seconds. BTW, a 60Hz field rate is better than would be e.g. a 75Hz rate since former, if used with a 3:2 pulldown chain, would render a 12Hz flash as an asymetric but "even" 12Hz flash, while a 75Hz display would "beat" visibly at 3Hz. |
|
| |
Well, if you want to be picky, 60 fields per second, NTSC, is really 59.94. |
|
| |
Don't confuse this otherwise excellent idea with details.. |
|
| |
[bristolz], your anno inspires me to re-investigate the possibilities of such adaptations for our family's "canned" entertainment.
|
|
| |
<aside>
I don't think I've come across such referenced technical information in your anno~s before even though I've caught vague mention to your association for such from HB veterans. -- The "a la cart factor" of the HB design?
Having seen this now (I think) for the first time, you appear entirely different in my collective impressions of, "Who is 'bristolz' ?" |
|
| |
Smart girls like you *should be electrical or mechanical engineers!" ;-) |
|
| |
This new awareness elucidates every 'bi-way' interpretation made.
Belatedly but respectfully, I humbly offer your due credit .
</aside> |
|
| |
Just goes to show, we really should have had 6 fingers. |
|
| |
<spews forth cauldron of laughter...> General, sir, I "sense" you're saying something important here?.. |
|
| |
// Well, if you want to be picky, 60 fields per second, NTSC, is really 59.94.// |
|
| |
Ah yes. One field per 59718.75 cycles of the 3579575Hz chroma clock. I never did understand why they chose the chroma clock so as to yield that odd frame rate. My understanding is that 60Hz black and white video originally had a horizontal scan rate of 15,750.0Hz and a vertical scan of 60.00Hz and was synchronized with the AC line (I've seen some old sets where the picture geometry changes slightly depending upon the relative phase of the AC line and the incoming video; when used with things like computers whose "60Hz" isn't, the beat is quite noticeable). Why they changed I have no idea. |
|
| |
Anyway, you do raise an interesting point: are movies generally telecined using 'drop-frame' video, or are they generally telecined 0.1% slow? |
|
| |
Drop-frame is a timecode format, not a video format. As for the exact speed of the telecine for NTSC transfers . . . I don't know. I do know in the PAL format it is common for them to just eat the difference and transfer at a true 25 with audio pitch harmonization. In NTSC, sometimes film intended only for broadcast is shot at crystal 30 (or 29.97 fps) and transferred at 30 to get around 3:2 altogether. |
|
| |
I know that Europe sometimes does the 4% time compression. I'm not aware of anyone shooting actual film at 30fps in the U.S. (I've thought it would be logical to shoot cartoons at 30fps to give the animators a choice of animating their characters in twos (15fps) or threes (10fps). An animator told me that shooting at 30fps would be too much extra work even if they did a lot of the animation in threes, since in many scenes it's necessary to reposition the cells on each frame even if the cells themselves are being used twice.) |
|
| |
Still, while I can understand filmmakers being upset at a TV station time-compressing their work by 4% (which is rather a lot), I don't think they'd really mind or even notice that their two hour movie came out eight seconds short. |
|
| |
// their two hour movie came out eight seconds short // |
|
| |
<pedant> actually 288 seconds, or 4 minutes 48 seconds to be precise, Captain. </pedant> |
|
| |
I recall (I could be wrong) that the 3.5 Mhz Chroma subcarrier clock was to do more with available crystals (cheapness) and carrier bandwidth than aything else; it has to slot in below the sound carrier. Likewise the 4.43 subcarrier on I-PAL is slotted in below the 6Mhz sound carrier. I'm sure ot was something to do with keeping the cost of the crystals down. But I'm prepared to be corrected on that. |
|
| |
Just get rid of the freaking 60Hz migraines... it's not so bad with a TV set, but CRT monitors have no business flickering like that. Maybe it's just me, but I can still see flicker at 72Hz. Anyone else notice this? (I managed to fanagle a sweet LCD panel from my company because I had so many migraines from the cheap ass monitor they forced me to use) |
|
| |
//actually 288 seconds, or 4 minutes 48 seconds to be precise, Captain.// |
|
| |
If a movie is shown on PAL, showing one 25fps (40ms) video frame for each 24fps (41.7ms) frame of film would cause a two hour movie to come out 4'48 short. This level of time compression--4%--can be noticeable. |
|
| |
Using a 5:6 pulldown chain to convert a 24fps movie to 29.97fps video would result in time expansion of 0.1%. This level of time expansion would make a video run 3.6 seconds per hour longer than it should. I don't think many people would notice that a 2:00'00"00 movie ran 2:00'07"06. I guess the ad-revenue people might object, but I can't see any other objection. |
|
| |
As for the ubiquity of 3,579,545Mhz crystals, was that frequency ever used for anything before the advent of the color subcarrier? Within certain bounds, it's no more or less expensive to make one crystal frequency than another, except that producing a whole lot of crystals of one frequency is cheaper than producing that mayn crystals at several different frequencies. |
|
| |
Shooting film at crystal 30 (29.97) is very common in the U.S. especially for commercials. Some folks even shoot _and_ transfer at 60 (1 frame -> 1 field) for a "hyper real" look. When combined with shutter synchronized HMI or strobe lighting the effect is pretty dazzling. |
|
| |
Personally, I would prefer to have a system that mimics digital cinema systems and allows me to play digital movie files at a true 24 just as intended when photographed. |
|
| |
//Personally, I would prefer to have a system that mimics digital cinema systems and allows me to play digital movie files at a true 24 just as intended when photographed.// |
|
| |
You mean something that shows 48 flashes/second showing each frame twice? I'd tend to think that a TV like that would flicker objectionably (when in England I found the 50Hz PAL annoying) which is why I suggested a 72Hz rate. |
|
| |
No, I'm talking about a true 24 fps video system as defined by the DTV 24p (progressive) standards. This includes a display system capable of locking to 24fps. They exist and they are heading to peoples homes, especially as DLP projectors. |
|
| |
24p and 24i (interlaced) exist as production systems as well. It's what Star Wars Episode II was shot with (Sony HDW F900 camera). |
|
| |
If a screen is flashing 24 times/second, the flashing will be noticeable and objectionable. Cinematic 24fps movie projectors are designed to flash each frame twice to avoid this flicker (this in turn can produce aliasing effects in some cases, but occasional aliasing is generally less objectionable than constant flicker). Projectors for 8mm (which runs 18fps) usually show each frame three times, since twice would only net 36 flashes/second, which would still be visible flicker. |
|
| |
Note also that if one were using a flicker-free display technology (e.g. active-matrix LCD) having a display scanned top-to-bottom 24 times/second would introduce considerable artifacting on objects which are moving across the screen. If the screen is scanned 72 times/second, this artifacting will be considerably reduced. |
|
| |
um . . . okay. I think I've forgotten what the point was. |
|
| |
<later> And, no, I don't think anyone was trying to shoehorn 72fps into contemporary TV sets/displays. All the 72 fps stuff I ever encountered, as proposals, called for a 72fps progressive scan display. Pretty rarefied when the sizes get big . . . but not for long. |
|
| |
//And, no, I don't think anyone was trying to shoehorn 72fps into contemporary TV sets/displays.// |
|
| |
No, but minimizing the level of modification that's necessary would be good. I would posit that keeping the horizontal scan rate the same and increasing the vertical scan rate by 20% would be the easiest approach, but for wide-screen progressive-scan sets and for 3:4-aspect interlaced-scan sets (the latter being used to show 24 frames/sec with 3:1 interlacing). |
|
| |
Seems like we're missing the obvious solution: a film format that displays at 60 fps. This would greatly enhance the appearance of theater presentations (I often notice flicker in fast pans in a theater; never on TV) and translate easily to video. There is already a proposal for a 48 fps format that would go a long way toward this. [link] |
|
| |
Douglas Trumball's 70mm Showscan is (or, er, was) exactly that: a film format shot and shown at 60fps. |
|
| |