John Sharaf
Forum Replies Created
-
The “benefit” of shooting an interview at 60p is that it has a “live tv” look, as opposed to the more “feature” quality that shooting at 30p (as 60 Minutes does) has.
Frame rate is a very subtle and subjective variable. In general 60p and 60i yield a live look and are used exclusively for sports and any live programming (like the six o’clock news). 30p is used for feature programing and especially for any material intended for internet or web use, because the repeated and fewer frames per second are most friendly to high compression rates that are used for this sort of distribution.
By having a progressive look rather than interlaced or “live” shows that are shot at 30p are different enough in temporal effect as to look special and even “film like” with less stuttering than actual 24fps, if not for the additional expense of film and processing film cameras would probably have standardized on 30fps when they transitioned from the silent era.
24p on the other hand exactly imitates the look of film (when transferred to video via telecine) and in addition, makes the material easy to translate to actual film for distribution in theaters and/or to 25fps for use in PAL countries, where 25p is broadcast by repeating frames as 50i and approximates the 30p look here in NTSC land.
JS
-
Ben,
Using gels “wastes” some of the output of the light. I prefer to use 2900K Kino Flo tubes and set the camera at 3200 preset. This equates to using a warming gel like 1/4CTO on the light without diminishing it’s output.
JS
-
Ben,
Not sure if you mean to use the “practical” fluorescents in the ceiling or “Kino Flo” (or equivalent) photo grade lights?
To use the overheads to illuminate a portrait, as in a typical interview shot, the lights are not in the ideal place relative to the face. What happens is that they top-light the nose and cheeks and not the eyes, creating what is often referred to as a “raccoon eye” effect. They can however be white balanced to create an almost accurate color. I say almost because they are a discontinuous light source with a green spike in particular. Fluorescent gels, such as they are, are meant to either eliminate the green spike (minus green), or add the green spike to tungsten balanced lights that is then white balanced out.
Photo grade fluorescents are tubes that are designed to minimize the green spike and be either 2900K, 3200K or 5600K in color. In addition they are often coated with a plastic cover to control the spilling of glass when handled oughly and broken. In addition, Kino Flo ballasts make the lights flicker, or go on-and-off at a higher frequency both to increase there output and to make them flicker-free at various shutter angles and frame rates.
It’s been mentioned many times here in this forum, but a rather elegent solution to shooting in a practically lit fluorescent interior, like a school or hospital, is to borrow some of the 4′ tubes that populate the overheads and use them on the floor in 4′ Kino Flo instruments; now every light is the same color and the floor lights can be used in a conventional portrait lighting location (into the eyes). Simply white balance and voila. The final touch would be to turn of or flag off the units directly overhead to eliminate the top lighting on the subjects nose.
Kino Flo units, especially the 4’Fourbank is a great portrait light, by virtue of it’s large size and soft quality.
JS
JS
-
John Sharaf
April 3, 2010 at 3:54 pm in reply to: Capturing a falling strand of hair in slow motion?Sorry, but effect you describe requires a specialized high speed camera like a Phantom HD which can shoot at 1000 fps. Cost with technician is about $2500/day. Abel Cine Tech is source, as is Panavision and other owner operators.
JS
-
Craig,
With mixed lighting and multiple cameras, which are looking in different directions with different mixed lighting, you really have to have remote CCU’s to “paint” them to match (also need to equalize video to match if analog). Your instinct to use 3200K preset was a good start, especially because you say you like a warm look, which the practical incandescents would have been (being closer to 2900K).
As far as the color temp readings in the camera, these are relative and cannot be taken as an accurate measure of the reading. They are derived from red and blue offsets from nominal and will not even match camera to camera. Being so far from normal though, the reading should have been a warning that something was wrong.
When white balancing under fluorescents it’s really about the green spike; the correction will add magenta to the overall look, making your tungsten instruments reddish in color. To go the route of minus green on them is to really chase your tail and never get it right. The only “good” solutions are to turn the overheads off, relamp with photo grade colored tubes or “poison” your additional lighting with the same lamps and white balance overall. This is the most elegant solution, however probably would not have worked for you here because of the need to “throw” spot lighting to action areas where the lead singer and others were.
In cases like this where you do not have the money, time nor resources to do the job right, you must make compromises, remembering that the most important thing is getting the recognizable skin tones right. If that meant gaining up and white balancing under a consistent light color, maybe that’s the best you could have done.
-
Ah, the magic of movies! You are correct that you need to shoot the foreground people walking as a green screen, either on the treadmill (green floor), or traveling on a green cyc and captured with a motion control camera to accurately repeat the movement (maybe use a click track to time steps). In addition you’ll need to shoot the background plate at exactly the same speed of movement to match.
If I got this assignment, I’d hire the best VFX Supervisor I could find, presumably one who has done something very similar before.
BTW, they better be willing to spend some $$$. My guess is about $100K, maybe more!
JS
-
The Panasonic Varicam, as its name implies, was the first “variable speed” video camera, whiuch while imitating the variable speed function of a film camera did so in a slightly different manner.
The ability to “undercrank”, essentially meaning to shoot at frames rates lower than normal, meant repeating frames, such that for example if you shoot at 10 frames per second each frame is repeated for six frames (as the camera always records 60 frames per second – called 60p).
At so called “normal speeds” of either 24 or 30 frames per second the camera duplicates frames either in a 2:3 cadence (like teleine) at 24p or by repeating every frame twice at 30fps. At all frame rates that require repeating frames the camera marks the first of each set of repeats with an “A’ frame in the user bits. This way if you want to recreate an undercranked effect, or a 24 or 30 fps timeline, your software can “pull” only the discreet frames out of the recording and not the duplicates.
Overcranking is made possible by playing pack the frame rates higher than “normal” (either 24 or 30) at the normal frame rate. For example if you shoot at 60p and play back at 30p the slomo effect will be 2x. If played back at 24p the slomo effect is higher at 2.5x. It is also possible to double print the 60p and have somethink like (but not exactly the same as) 120fps. The cameras have the capacity to make that many frames but are unable to record them.
Cameras since the Varicam have copied these techniques and now there are models that can write 1920×1080 at rates as high as 60p.
Hope this helps!
JS
-
Jumping in late here, but with multicam, ability to link time code would be a deal breaker for me. Does the 150 have this capability? If not then you should, as others have suggested, consider the EX3 very seriously, but there of course the SxS is a significant liability, solved by spending another $3000 on Nanoflash! Pro gear with pro-features cost money!
JS
-
Justin,
The “old Hollywood way” of covering an actor’s path with even light is to back the key light up and use single and double open end nets to “feather” the light output as the actor comes closer. Sometimes a grip will also “float” a net into the light to effectively “dim” the output as the actor approaches.
JS
-
Purity of white is a function in the camera of proper white balance, lens white shading, clip, knee, knee sat, gamma and then there are comparable settings in the recorder. To determine if the camera is truly at fault, you must bring it to the camera engineering lab and have a good look-see with proper charts, waveform/vectorscope and color monitor.
JS