Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: A newbie question: manual exposure in video shooting  (Read 2150 times)
kwingyiu
Newbie
*
Offline Offline

Posts: 4


WWW
« on: September 05, 2013, 05:37:20 PM »
ReplyReply

Do you use manual control in video shooting? Does a lightmeter such as Sekonic L478D help?
Logged
Tony Jay
Sr. Member
****
Online Online

Posts: 2029


« Reply #1 on: September 05, 2013, 06:53:57 PM »
ReplyReply

Hi Welcome to LuLa.

Generally with video all settings should be manual.
One does not want the exposure to change mid-shot when one is doing a pan and the light changes for example.
There are exceptions to this rule but they are for specific situations.

Tony Jay
Logged
kwingyiu
Newbie
*
Offline Offline

Posts: 4


WWW
« Reply #2 on: September 05, 2013, 09:14:22 PM »
ReplyReply

Many thanks Tony.
I am not sure if this is good to talk about the brands here.
But would you recommend some affordable lightmeters in the market?
Logged
Peter McLennan
Sr. Member
****
Offline Offline

Posts: 1624


« Reply #3 on: September 05, 2013, 11:01:02 PM »
ReplyReply

You don't need a light meter if your'e shooting video unless you're using it to adjust the relative intensities of several lighting instruments in the set.
Your viewfinder is your light meter.  If your camera viewfinder has a "zebra" setting, it will indicate maximum exposure levels.
Logged
Tony Jay
Sr. Member
****
Online Online

Posts: 2029


« Reply #4 on: September 06, 2013, 03:23:03 AM »
ReplyReply

Peter is exactly on the money!

Tony Jay
Logged
kwingyiu
Newbie
*
Offline Offline

Posts: 4


WWW
« Reply #5 on: September 06, 2013, 04:16:58 AM »
ReplyReply

That helps.  Many thanks Peter.
Logged
Petrus
Sr. Member
****
Offline Offline

Posts: 458


« Reply #6 on: September 06, 2013, 05:57:20 AM »
ReplyReply

4 ways to determine exposure with video cameras:

1) viewfinder. If the image looks ok, it should be ok enough. Caveat: in bright conditions you tend to overexpose, as the viewfinder appears dark, in dark conditions the opposite. Thus this method is not exact. Also two identical cameras must have same brightness settings for the VF to give the same exposure, if determined by the VF. So this method is not very good, but often good enough. It is always possible also to switch to auto exposure to see how the camera evaluates the scene. If there is a big difference in a normally light scene, the camera might have it right, at least you should find out what causes the difference.

2) using zebra. Usually the zebra level is either 80% (caucasian skin highlights) or 100% (maximum highlights). Adjust exposure so that 80% zebra just starts to appear on facial highlights. Place a white card on the scene and adjust so that 100% zebra starts to appear on it.

3) using a videoscope, which is the best tool. Videoscope is either built-in in the camera (better prosumer models) or included in a modern external monitor. It is slightly related to histogram. You can see a live analysis of exposure values for each vertical row of pixels.

4) using a light meter. This requires some work with calibrating the meter to the camera at various gain settings, white balance conditions etc. Easy to make fatal mistakes (if the viewfinder image looks bad, you would trust that more anyway, so we are back at square 1...)
« Last Edit: September 07, 2013, 06:04:09 AM by Petrus » Logged
Morgan_Moore
Sr. Member
****
Offline Offline

Posts: 2196


WWW
« Reply #7 on: September 06, 2013, 07:38:03 AM »
ReplyReply

If you are shooting with a DLSR or other camera with a histogram I find that very good.

'Wave form' is better than histo but only on more expensive video cameras.

Some DSLRs dont have a live histo. In that case I often knock off a still (on my video settings) and examine the histo from the still

Some 'monitors' (small screens) have their own waveforms histograms and other exposure tools. (eg tvlogic, odessy 7)

S
Logged

Sam Morgan Moore Cornwall
www.sammorganmoore.com -photography
kwingyiu
Newbie
*
Offline Offline

Posts: 4


WWW
« Reply #8 on: September 07, 2013, 10:16:58 PM »
ReplyReply

Many thanks Morgan_Moore and Petrus.
Can I ask if there is equally the question colour temperature to be dealt with in shooting video?
Logged
Petrus
Sr. Member
****
Offline Offline

Posts: 458


« Reply #9 on: September 08, 2013, 01:34:29 AM »
ReplyReply

Can I ask if there is equally the question colour temperature to be dealt with in shooting video?

There certainly is, and it is more problematic than with stills: WB might change during a take (with manual WB you can not change it smoothly on the run, with auto WB the color of the take might change when you do not want it to). The other problem is that there is much less latitude to correct the colors in post than there is with stills, as the material has less color space to begin with and it is tightly compressed losing much information. 75% of the color information is thrown away for starters (4:2:0 and 4:1:1 compressions).

Normal procedure is to aim the camera at white surface (paper etc) and press a button on the camera to set the white balance.

How about consulting the camera manual?
Logged
bcooter
Sr. Member
****
Offline Offline

Posts: 989


Bang The Drum All Day


WWW
« Reply #10 on: September 08, 2013, 09:34:31 AM »
ReplyReply

Yes, manually set white balance.  Some cameras will go smoothly with wb changes like the gh3, some drastic like the omd, or the sony's fs100., some don't seem to change at all.

A drastic change is a mess because the wb can move up and down on the same scene, same angle.

Regardless it doesn't stop there with motion, because motion has to have continuity in look, especially if your cutting from different settings and back.

You shoot outside in that standard Kodachrome everything is colorful look and then cut to an inside session where it's north light and muted and the difference is hard to accept in color and tone.

There needs to be a commonality in the look and color is probably the first thing a viewer notices and digital more than film, (I know nobody is talking about film), tends to be ambient color sensitive.

We just did an interview/lifestyle/dialog piece in one house, one subject.  We painstakingly made sure we used the same key lights, (kinos), the same fill (white bounce cards) the same cameras and settings (R1's), the same manual wb settings, the same tone/curve and gamma settings.  

Even with that going scene to scene it looks like different cameras, different days.  It has to then be worked in post and the RED's at 444 give you more information to work with at the start, still motion is a fragile file.

If your shooting, inside to out, car shots, to beach, different times of day, then this becomes problematic.

This is one of the reasons that so many dps shoot a flat log style with their footage, to give room to move the footage, though that can deceive you, as a flat, undersaturated file will give the impression that all scenes match, until you start bumping the contrast and saturation, then the differences appear.  

I believe that is why so many camera testers that shoot a new camera out of the box opt for black and white.

Everybody approaches this differently.  On the cable show Longmire, which is shot on REDS I believe, they give all the footage a warm global look.   That fits because it's a western themed show and it makes it a lot easier to hold the common look from inside to out, bright sun to overcast.  

Some shows, even very good shows just miss it and you can see the jump.  Continuous, fill light helps.

One thing we've done is to take footage, drop it into the NLE, or CineX while we're working and do a screen shot of the viewer.  Then we take those screen shots into a photoshop file and just keep a poster board up on the screen that shows the look from scene to scene.  This gives us some idea of where we are going, though most of the time, you can't do a whole lot on set to correct it, unless you have gobs of time and nobody has gobs of time on set anymore, so you just hope and pray you can pull it back in post.

You should make note that if your camera plays a movie in quicktime, quicktime X (usually a default player on most new apple systems) reads a different gamma than quicktime 7, which can read a different gamma and look than an NLE like final cut pro and this can be deceiving as sometimes what goes into all three with match sometimes they'll look 40% different.

Even making galleries for client review can change it.  We use Media Pro which use to be I-view and it reads something way different when a small quicktime file is placed into it.  It usually goes about 1 stop darker, which can be confusing.

This is one of the reasons a raw camera works well for fast production.

IMO

BC
« Last Edit: September 08, 2013, 09:53:01 AM by bcooter » Logged

Morgan_Moore
Sr. Member
****
Offline Offline

Posts: 2196


WWW
« Reply #11 on: September 09, 2013, 05:00:39 AM »
ReplyReply

Trying to give a simple response.

With a vid cam like the FS100 I point at a grey card and press the WB button.

With a DSLR I look at the RGB histogram and dial the manual WB until the R G and B peaks seem to be about the same height (when filming/photographing a grey card white reflector etc)

S
Logged

Sam Morgan Moore Cornwall
www.sammorganmoore.com -photography
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad