Richard Herd
Forum Replies Created
-
That raises a fine question all around:
Film is shutter angle * frame rate = shutter
180/360 * 1/24 = 1/48.
Piece of cake and accounting for my film speed, I can get an exposure.Video is what?
“Synchro” rate * sampling frequency = Huh?I have no idea what the equivalent equation is, and it’d be great if someone could cut through the murk.
-
LOL! You can’t use a cute kid to test skin. I need the results for a hung over actor who smokes too much!
-
There’s two parts. First part is how to fix it. Second part is what I believe is going wrong.
1. You’ll have to use Apple Compressor to transcode your footage into the SAME format. That is, the frame size, Vid Rate, Compressor, Aud Rate will all need to be the same. Since most of your footage is DVCPRO HD 1080p30 @ 23.98 fps, then I suggest transcoding the 29.97 fps footage into that vid rate, simply because it appears it will take less time. Make sure your sequence corresponds to this new “stuff” (frame size, vid rate, etc.)
2. I believe that the error you’re seeing is caused by placing mismatched footage into a sequence. I assume that also means you’re having to render in the sequence. A good rule of thumb is if you bring footage into a sequence and you have to render, something is wrong, so stop and figure it out. FCP Studio 2 is cool because it will “auto pick” your sequence presets. (Why they didn’t do this with 4.0–when I started FCP–I’ll never know!) But that won’t solve your problem because your vid rates differ.
Vid Rate is a trippy thing. Imagine the first frame as being 0, followed by the second frame 1, then 2, 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 then where does it go? Well, if you’re in a 23.98 sequence, then it goes to the next second, but if you’re in a 29.97 sequence, then it goes to 24 25 26 27 28 29, but since the footage itself is 23.98 and the sequence is 29.97 (or vice versa), then the computer has to recount the old frames. But it doesn’t “transcode” the vid rate. Instead, it creates a render file which tells the sequence what the new vid rate is. It’s terribly complicated.
2:3:3:2 is way different. It’s about video FIELDS not frames, lots more on that but I’ll skip it. Anyway, when we edit in FCP we do not make our edits on fields, but on frames, so the fancy math creates a frame out of fields. But the two fields exactly next to each other will not match up, so the fields must be interleaved to create a frame. If you assume it takes two fields to make one frame, then you’ll notices 2:3:3:2 means something like “the first frame has two fields, the second frame has three fields, the third frame has three fields, the fourth frame as two fields.” Add the number of fields up 2+3+3+2=10. That means 10 fields are making 4 frames. Normally, we’d expect 2 fields to make 1 frame, so 10 fields should make 5 frames. But that extra frame has been “pulled down.” Here’s a bit of video blasphemy: this “pull down” all by itself won’t make your video look like film 🙂
Well, that’s my best guess on what’s going wrong, and I hope someone checks this post for accuracy!
-
The footage is what it is, and in my experience FCP is very good about reporting the metadata facts.
Make a new setting that corresponds to those facts; rather than creating the sequence something the footage isn’t. (Sorry if that’s awkward.)
Please include a screen capture of your footage bin.
Additionally, deinterlaced video differs from progressive video, and it’s a weird topic. There’s a great explanation at https://www.lurkertech.com
-
That’s the spirit!
-
Thanks! I have some silks I can use to eliminate that double shadow. It hasn’t been a problem, though, because talking heads just don’t move and my aim has gotten pretty good.
I’ve used exterior light for the key when the sun gets low enough. Tahoe buildings are designed in such a way that the snow will slide off the roofs, so there’s always an overhang, shade type of thing.
-
These fireside chats, are they one take, well-rehearsed type of thing? I love windows. I live in Tahoe and am always having to expose INT. DAY for EXT. DAY because the view of Lake Tahoe, of course, is central to the shot.
My very simple technique (or is it a trick?) is to expose for “the view,” and frame accordingly. I light the talking head with two 650 open face Arris with 1/4 CTB (’cause that’s what I have) — placed to the far right of my camera, trying to get a bit of shadow on the far side of the face because in my opinion chiaroscuro looks cool. Just to make the picture a little more interesting, I clip a foam core to a C-stand and bounce as much daylight back into the subject. I experiment with placing the bounce card until I get the rim light I’m after.
A couple other things: the whole light, scrimmed and shapped to taste, is placed, I don’t know, 10 to 15 feet away from the subject, and I keep my camera 10 to 15 feet from the talking head.
For a mic, I use a shotgun mic or lavalier, plugged into the camera. I wear headphones while shooting, which, by the way, makes it easy by mistake to yell at the talent. (So there’s a little cautionary tale.)
The reason this works in editing is because it’s one take. It would be pretty hard to edit if the background light didn’t match from cut to cut, like in winter when a snow storm moves in over a few hours.
(This set up also entail that you pick the correct time of day to shoot, morning/evening look the nicest, but you can always gel the window with ND gels if you have to shoot at other times.)
The room without the windows: nifty trick here is practical lamps. These won’t actually add much luminance but they’ll “trick the eye.” A floor lamp and or a table lamp shown on camera and set in convenient 3-point lighting pattern will mean you can light the talking head in a very typical standard beauty lighting. There’s tons of places to learn that. Be sure to put 15-watt bulbs in the practicals and expose so they are your brightest point in the shot. On the other hand, if you have to shoot a whole board room with folks sitting all around: use China balls above the actors, center of the table (maybe a couple of them), and a few between the back of the chairs and the walls to remove nasty shadows. Frame the china lanterns out of the shot, but keep the 15-watt practicals in the shot. If the room has built in fluorescent lights in the ceiling, take those out, but turn the light switch on 😉
-
I think that’s a pretty good rule of thumb, but the sampling frequency and codec also play important roles.
-
I hope my post isn’t too oblique.
The thing to be sure of is (that) the monitor on which you plan to display your picture will resolve the signal being fed to it (through the input you plan to use).
For example, the back of most monitors have several inputs to choose from, so you need to make sure the signal being output from your camera (or DVR) matches the input of your monitor. You’d think this wouldn’t be an issue, since the cables match up, but it is an issue. Often, the monitor and the camera have menu settings where the signal can be manipulated, so it all matches properly. I’ll state the obvious, so please don’t think I’m being rude: reading the manual for both the camera and the monitor goes a long way to explaining some very important details.
Having said that I recommend https://lurkertech.com
From a more creative point of view, the appropriateness between 720p, 1080i, 1080p, 2k, 4k (and 6k so they say!) is very hard to quantify and turns into cost-benefit analysis. I’m the first to admit I have a lot to learn, so, from personal experience, I am not using high end system because that also means I risk capturing more noise than signal–and that’s too expensive. As my skills improve, okay, maybe I’ll move upwards in capture devices, but it’s not like you can say, your movie will be better simply because you shot with “X gadget.”
Here’s a quick question to get you started thinking about the whole topic: Where are you going to store the data?
-
The autofocus lens and servomotor tighten/widen button on modern dv cameras (like the panny ag-hvx200) make this camera move super easy to execute.