Difference in using Camera Gain vs. Post brightening methods?
 

As we all know ''gain" is electronically brightening a dark picture image in camera. So i was wondering, Is there any difference in using in Camera gain vs. post production methods of electronically doing the same thing in the NLE?

One difference that comes to mind is that if you shoot with too little gain (in the dark) underexposing the footage there is no detail in the shadows (dark part of the video). So it will be impossible to make the shadow area lighter and get any detail where there was none to begin with.

It is certainly possible to lighten midrange a bit in the post, but somebody should do a controlled test to see which works better.

That's a really good question; shooting documentaries I have no choice but to gain up to +3 or +6dB and I hate to do it, but my personal theory (and it's only a theory) is that if I need gain, I'm better off gaining up in-camera, before HDV or DV compression, than in post where I'm trying to make something out of nothing. I haven't put this through any scientific tests other than shooting and looking at a lot of footage.

Typically, the gain in the camera is performed before the signal is digitized and compressed. Therefore it would be better to use gain in the camcorder rather than post.

The gain applied in post is a lot more susceptible to grain than upping the gain at capture by the camera.

I'm also in support of using camera gain.
Even at 0db gain, the dark parts of your
image will still have lots of grain.

I am going to sort of disagree with the traditional wisdom. Mostly I am saying its more complicated in reality.

What everyone has said is true- gain applied in camera before compression is better than the same thing attempted after compression. The more compressed the more true.

The thing is that you have to consider the whole system. What if I am capturing dual link HD SDI to a 4:4:4 uncompressed video? How about regular HD SDI to a 4:2:2 compressed mastering format like ProRes, DNxHD or Cineform? Clearly that's a bit different than DV or HDV. The post colorist can definitely wiggle the pixels better in those cases than the camera.

We also have to consider the amount of control and precision applied by the colorist.

Using Apple Color as my example, I can process the image at 32 bits per pixel. That is significantly better than any camera's internal processing.

The camera can't tell the difference between a dark hallway and the leads flowing raven black hair, the colorist can. I can mask regions and lift shadows in one area, deepen them in another. I can do multiple iterations of this using masks, keys, rotoshapes and tracking to refine the areas I am controlling.

Given that sort of post coloring, a lot of times its better to leave those decisions to the colorist in post.

So... the net result is that the answer is unclear.

The only advice is to retain as much information in your image as practical.

Control your highlights and shadows and get data. More important than getting information in the shadows however is to make sure your highlights clip as little as possible. In my experience highlight details are more noticeable to most viewers.

On a film set, I would never use gain, but rather would use grip and electric to control my shadows and highlights. Unnatural lighting to yield natural results and all that.

In documentaries or news you may have no choice but to use gain to control exposure in some instances. Do so. Sometimes its easier and just plain faster to just flick the gain switch, but I try to think about it before I flip away.

Use the least gain that will get you the details you need. Sometimes your lowest +gain setting will put information in the shadows but you can't see it.

A lot of cameras record data in the 0 to 7.5 IRE range, but that information isn't displayed in many environments until you lift it over 7.5 IRE. A waveform or histogram should reveal it. (That's why I recommend monitors that include a waveform monitor.) That's enough, leave it alone. Raise it further in post if you must.

If you have worries about a specific part of the image- its time to go and adjust gamma, knee and shoulder settings. To some extent this is like applying gain selectively to your image.