Some questions from a stills photographer.

D

dylansmith

Well-known member
Some questions below:

1. Coming from a stills background and having used Lightroom for years, I am finding it a bit hard to work on video using Premiere pro CC.
a. Instead of using the RGB corrector or Levels tool, where can I find a multitude of sliders (e.g. exposure, contrast, vibrance) for easier control?

b. I have heard about "masking" as being somewhat equivalent to the target adjustment tool in Lightroom. What is the easiest way to do targeted adjustments (e.g. brighten face in shadow, sharpen eyes) in video using PP CC?

c. How much dynamic range do I have in video mode on my 6D (understand there's RAW video using ML, but it's underdeveloped) compared to stills mode?

c2. Does ALL-I mode really offer more dynamic range than IPB? Or is it only about the NLE editing speed? If it is the latter, I feel absolutely no difference editing either of such videos.

d. What are some of the most popular color grading presets (yes, understand there are purists who prefer to adjust as per the situation) that you guys use for your videos? My videos are 99% about my kids - 1-2mins in duration and usually upbeat/cheery in nature. would love some recommendations please.

e. is h.265 profile fully supported in PP CC yet? Would love to export all my completed clips in h.265 for the best compression possible.

2. On the A7Rii, there is the Super35 crop mode vs FF 4k mode. If I use the former, does it mean my FL gets multiplied and DOF is increased?

3. Why is it that some cameras have a higher native video ISO compared to stills (800 vs 100) ?

4. When doing a cross-fade transition between 2 clips (10s clip each), PP CC says I don't have enough media duration to crossfade, and attempted to freeze my 2nd clip for a few seconds when I inserted the transition effect. How is 10s not long enough when the transition effect is only <1.5s? What can I do to avoid this situation?

5. The Warp stabilizer seems to produce erratic results on some of the short clips I have. Even on a clip that was lens-stabilized and just need a little post-processing help, it literally warps my clip (e.g. stretches faces of people and makes it look like a dreamy scene). Which plug-in do you guys recommend for that little extra help? Lock&Load / Mercialli or some other alternative?

6. What is the industry-standard method to grade clips from 2 different cameras? Is it done through 'eyeballing' or is there a button that can do a best guess to 'sync' the color in post? I currently own a 6D and Oly E-PL5 and the clips are vastly different (the latter's quality is very crappy).

Thanks!
 
Last edited:
Try the following
1) download Davinci resolve lite it's free & learn to use
2) It should
3) you're dealing in 24/25/30 fps this impacts sensor sensitivity hence higher ISO as opposed to longer shutter speed overcoming lower ISOs . ( IMO lower is more preferable than higher ..remember Kodak Kodachrome? )
 
Last edited:
As a long time still photo shooter ( since the mid 80's ), I would recommend downloading the free trial of Sony Vegas Pro and then take a look at the LEVELS, HSL Adjust, Brightness and Contrast, and the Color Balance controls. They are very similar to their still photo analogs, and should help you to get up to speed fairly quickly.

Sony Vegas Pro 13 web page

>>>e. is h.265 profile fully supported in PP CC yet? Would love to export all my completed clips in h.265 for the best compression possible.

Without hardware support h.265 is VERY processor intensive for both compression and playback, so the space savings may not be worth it.
 
Couple points that haven't been hit yet:
-Crop modes change your field of view to approximate a longer focal length, but do not affect depth of field--all other things equal. To clarify: If you compose a shot full-frame, then switch to crop mode and recompose to re-create the initial shot, you will zoom out and thus have deeper depth of field. But if all you do is switch to crop mode and make no other changes, you're simply grabbing a smaller portion of the same image.

-For kids videos you probably want a nostalgic feel, which I would say means tilt the colors toward warm (In RGB, try red +20 and blue -20.) and then desaturate. Add some jitter and film grain if you want to get extra cheesy.
 
1a. Use 3-way Color Correction and RGB Curves.
1b. You have 2 layers: 1 with unchanged image and 2nd with whatever you want to do to the face. Then you feather the layer so the changes you do are not apparent.
1c. Yeah, dynamic range is not what you get in RAW. Basically what you see is what you will get later on exposure wise. You shoudl go with RAW capable cameras if you want to fuss around with RAW later on.
1c2. No. It's how the footage is compressed - meaning how far you can push it in the post before it starts falling apart with over pixelation.
1d. There really isn't set of presets unless you want it look like "that things everyone else is doing".
1e. Don't know about implementation, but name of the game is compatibility of your codec with as many system as possible. Because you have to ask the questions means that the codec will not be very widely supported. Stick to h264.
 
charliemcdanger said:
Couple points that haven't been hit yet:
-Crop modes change your field of view to approximate a longer focal length, but do not affect depth of field--all other things equal. To clarify: If you compose a shot full-frame, then switch to crop mode and recompose to re-create the initial shot, you will zoom out and thus have deeper depth of field. But if all you do is switch to crop mode and make no other changes, you're simply grabbing a smaller portion of the same image.

-For kids videos you probably want a nostalgic feel, which I would say means tilt the colors toward warm (In RGB, try red +20 and blue -20.) and then desaturate. Add some jitter and film grain if you want to get extra cheesy.
Click to expand...

talking about kids' videos, i find it hard to get examples of professionally done clips. I am trying to get a feel of what the gold standard is like, for well-known videographers who capture their/others' kids everyday lives and milestones (e.g. start to crawl/sit up/eat their first solids). All of the clips on youtube/Vimeo so far are impromptu/amateur clips.

Where can I find great examples in this genre? Anything else in particular to take note for this genre?



Patryk_Rebisz said:
1a. Use 3-way Color Correction and RGB Curves.
1b. You have 2 layers: 1 with unchanged image and 2nd with whatever you want to do to the face. Then you feather the layer so the changes you do are not apparent.
1c. Yeah, dynamic range is not what you get in RAW. Basically what you see is what you will get later on exposure wise. You shoudl go with RAW capable cameras if you want to fuss around with RAW later on.
1c2. No. It's how the footage is compressed - meaning how far you can push it in the post before it starts falling apart with over pixelation.
1d. There really isn't set of presets unless you want it look like "that things everyone else is doing".
1e. Don't know about implementation, but name of the game is compatibility of your codec with as many system as possible. Because you have to ask the questions means that the codec will not be very widely supported. Stick to h264.
Click to expand...

1b. How is that achieved in PP CC? At least I need a keyword to find help in their manual. thanks!
1e. My videos are currently for personal consumption only, hence compatibility doesn't worry me 1 bit. Is there any way I can export to h265 then (assuming I don't have to export twice and degrade the footage).

Surf it Mike said:
Try the following
1) download Davinci resolve lite it's free & learn to use
2) It should
3) you're dealing in 24/25/30 fps this impacts sensor sensitivity hence higher ISO as opposed to longer shutter speed overcoming lower ISOs . ( IMO lower is more preferable than higher ..remember Kodak Kodachrome? )
Click to expand...

1) DaVinci Resolve looks more complicated than the default tool. What grading tool do beginners use for such a purpose?

TheDingo said:
As a long time still photo shooter ( since the mid 80's ), I would recommend downloading the free trial of Sony Vegas Pro and then take a look at the LEVELS, HSL Adjust, Brightness and Contrast, and the Color Balance controls. They are very similar to their still photo analogs, and should help you to get up to speed fairly quickly.

Sony Vegas Pro 13 web page

>>>e. is h.265 profile fully supported in PP CC yet? Would love to export all my completed clips in h.265 for the best compression possible.

Without hardware support h.265 is VERY processor intensive for both compression and playback, so the space savings may not be worth it.
Click to expand...

I'm already invested in PP CC, hence it'd be good to stick within the ecosystem.
My videos are currently for personal consumption only, hence compatibility doesn't worry me 1 bit. Is there any way I can export to h265 then (assuming I don't have to export twice and degrade the footage).


**********************************************

On another note, I am also facing another 2 issues:

4. When doing a cross-fade transition between 2 clips (10s clip each), PP CC says I don't have enough media duration to crossfade, and attempted to freeze my 2nd clip for a few seconds when I inserted the transition effect. How is 10s not long enough when the transition effect is only <1.5s? What can I do to avoid this situation?

5. The Warp stabilizer seems to produce erratic results on some of the short clips I have. Even on a clip that was lens-stabilized and just need a little post-processing help, it literally warps my clip (e.g. stretches faces of people and makes it look like a dreamy scene). Which plug-in do you guys recommend for that little extra help? Lock&Load / Mercialli or some other alternative?

6. What is the industry-standard method to grade clips from 2 different cameras? Is it done through 'eyeballing' or is there a button that can do a best guess to 'sync' the color in post? I currently own a 6D and Oly E-PL5 and the clips are vastly different (the latter's quality is very crappy).
 
Last edited:
Where can I find great examples in this genre? Anything else in particular to take note for this genre?
Click to expand...
The classic, which is obviously exaggerated to capture the era, is the Wonder Years opening, and a couple guys who copied it perfectly are Tripp and Tyler. Those have the warm colors, the jitter and film grain, plus the cropping to replicate 8mm projection. You'd probably want to tone all these down a bit, but the vibe is one that everybody will feel fuzzy about.

A couple points for shooting: 1) Don't shoot down on kids--get down to their level; and 2) Get their faces. Kids aren't straining and posing and phony like adults are, and people adore that innocence. Don't be afraid to use a long lens. Here's a pre-wedding video I did for some friends; you can see how I incorporate the kids and whether you like it:

https://vimeo.com/70661061

Re: crossfade. You cannot dissolve into the beginning of a clip or out of the end of a clip, if there's no unused portion of it available for the dissolve. For example if you have the very end of one clip leading into the very beginning of another with no removed portion, a 1-second dissolve would require you to chop off 1 second of either one of the clips to allow material for the dissolve. And which clip you chop will dictate where the dissolve can be placed. Now if you're working with shortened pieces of your clips, you'll never have this issue.

Re: mixing different cameras. If they're vastly different as you say, don't try to match them. It will look s*%tty and distracting. Pick an effect like black and white or the exaggerated Wonder Years look, or whatever you want, and apply that to the footage from the second camera. It will add some nice variety to your edit. And then stop using that camera.
 
Last edited:
About e2, crop mode effect: Yes it does affect field of view and depth of field. When you switch from FF mode to s35, you have an APS-C camera now, with the same crop factor on factor on focal length and the same depth of field (Shallower dof on FF)

There are technically three elements that affect how shallow/deep your depth of field is,

-Focal length (the longer, the shallower)
-Aperture (The larger, the shallower)
-Distance from subject (The closer, the shallower),

So sensor size isn't technically one of them, if you put the same lens of a FF camera and a Tiny 1" sensor camera, the area in focus is the same depth, but, the reason why smaller sensor sizes have deeper depth of field is because they affect these three elements in real world shooting, hence indirectly affecting the depth of field.

For example, using a smaller sensor will make you use a wider lens to get the same field of view, therefore making depth of field larger, also a smaller sensor will make you go farther from your subjects, hence making depth of field larger.

These are the reasons why a smaller sensor (S35, s16, 1", APS-C) have a deeper depth if field than larger sensor cameras in real world use. FF cameras make you use longer lenses to get the same shot and get closer, therefore blowing the background of-focus more. What some call the FF aesthetic.

About easy editing, the easiest video editing software that will allow anyone to get great results easily whilst also allowing for highest-end work, is Sony Vegas Pro. Try the Trial version. When I was choosing an NLE I tried all and Vegas was a godsent, no other software has the timeline that's as tactile and visual as vegas, you just put clips, want to cross fade, over lap them, want to add an effect, add a Colour levels adjustment plug-in on the clip, simple sliders or curves are available , simple and effective. The plug-ins also include a very effective stabilization one, sharpening, colour correction, grading, practically everything is there just easy. Also the audio capabilities is better than any other NLE (as Vegas started as an Audio software), the only area where vegas lags behind is masking and targeting specific areas in video frames, you have to do it manually, the best software that does this job is the free Da Vinci Resolve, you just pick an area and it tracks it automatically. That's why I have resolve just for masking jobs, for everything else Vegas, Resolve is VERY pro level and complicated.

You have less dynamic range in video mode vs raw stills on the 6D, about two stops less. Just use it and know the limitations. Very few cinema cameras have raw DR potential that comes with stills.

ALL-I vs IP-B is just two different compression methods, All-I compresses each frame indiviualy, giving less motion artefacts but larger files, IP-B compresses files in groups, giving smaller files but more motion artefacts that appear only in heavy situations. ALL-I is easier for the computer to handle as it's less compressed, the PC doesn't have to uncompress the files again on the fly as it needs with a more compressed codec like IP-B. however if your computer can handle IP-B that will not be a difference you'll see. No change in DR, just a slight change in compression, try both and see what you want, smaller files with more compression that you don't find objectionable or larger files with less compression to get the most out of it.

H.265 is not supported anywhere practically, use H.264 with a high bit rate, it's really high quality, or use ProRes or DNxHD, many great codecs out there H-265 is not one to go with as of yet.
 
Since you're shooting on a 6D make sure you turn contrast all the way down to get the highest DR, Turn sharpness to +1 max no more. This is extremely important for Canon DSLRs. Saturation is best reduced as much as possible, as it's possible to increase it (even if shot at -4, in post you can get it to +4 just as if it's been done in-camera), but reducing it in post is way more difficult since you'll have clipped colour channels at many occasions that can't be fixed by removing saturation. So keep it down as much as you feel the base you can shoot with without being annoyed with a gray image. from 0 to -2 is good.

Also something I find extremely important for shooting with these cameras, noise reduction software, Having the 100$ Neat Video is a magic ticket to no noise and MUCH higher quality footage. I can't stress enough how important it is especially with Canon Cameras that have ugly shadow noise, this completely fixes these codec issues and gives a clean lovely image to grade and look at, without removing any detail. Try the demo at least.
 
If you are from raw stills I would..

-Get Resove to do your colour (and edit?)

-Getting a camera that has at least 10bits of recorded data will help.

8bit is 256 levels of colour

10 bit is 1024

14 bit raw is 16000 tones

8bit is comically thin compared to a decent still.

S
 
I am just as much into stills as I am video. I also use premiere pro. Here are the answers to a few of your questions:

In Premiere Pro CC 2015 you best friend is the Lumetri effect for color grading. It's exactly what I've been wanting for years and is very similar to lightroom/camera raw. You have sliders for white balance, exposure, whites, highlights, shadows, blacks, contrast, saturation, vibrance, sharpness and plenty more. It's also a very lightweight effect.

the 6D will be like the equivalent of shooting JPEG in a DSLR where the color profile is burnt into the image. You can download flat profiles that give you more room to grade but nothing like a LOG profile. In contrast, cameras that shoot LOG (Sony FS7, canon C cameras, etc.) provide you with all the stops of dynamic range that the camera can handle. With RAW stills, you are always seeing your image with some sort of color profile "preview" applied, whether it's in camera or once you bring it into lightroom. By contrast, with a camera that shoots log, you are seeing all stops of dynamic range squeezed into an 8bit output which makes everything look desaturated and flat. You have to apply color correction or a LUT (look up table) to this in order to see it as intended. A LUT is essentially a color profile.

Warp stablizer only works well on certain types of movements and nothing too crazy. You can try lowering the stabalization amount (down to 5%-10%) to see better and more natural results.

You can do local adjustments in premiere but due to the nature of video it's not as simple as with lightroom. Apply an effect (such as sharpness) and then click the creat mask button. THen you have to keyframe or motion track the mask across the entire video clip.
 
You must log in or register to reply here.
Back
Top