"Color Grading" - I don't get it - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 71 Old 10-02-2013, 08:37 AM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Many posters in forums (elsewhere) attracting enthusiasts and even pros (having in common only evident large equipment budgets) laud shooting flat (no color or contrast or sharpness) and then adding "to taste" color afterwards, partly in an obsessive quest for high dynamic range. But the cost in terms of color seems enormous.

I get that one might want a color "look" to convey a mood - post-apocalyptic world, film noir, dread, happy times (this is rare) - but these folks apply it to soccer games and strolls in the woods, making colors that look like faded Polaroids or out-takes from films with low-budget CGI (so color has to be sacrificed).

Here is an example of a color-graded woods walk video proudly posted to show the guy's skill and camera performance in a forum on the GH3:

[select 1080p]

And here is a video made by the same guy showing how he grades:


[select 1080p]

He takes an extremely dull video and makes it only very dull. Notice the same phony sound track - canned nature noise and synthesizer music (you can tell the nature sounds are not from the scenes - no running water sounds when you see running water, no change in sound character across clips).

I color grade too. At the scene I compare what my eyes see to what is on the lcd, then I alter camera settings (WB, Picture Style, exposure) to match them. Here is a video I took from a walk in the woods. I confess I was looking for color, to show that not all scenic videos need be dull.


[select 1080p]

It was a sunny contrasty bright day (and I guess I was in a good mood). No post processing at all - all pre-. and sound from the scene (mostly dominated by the waterfall).

I am not arguing against color grading or that my "color grading" technique is best, only that "shooting flat and post processing" rarely yields good results, so do it with caution. And there are alternatives.
markr041 is offline  
Sponsored Links
Advertisement
 
post #2 of 71 Old 10-02-2013, 09:08 AM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Here are some frame grabs from the woods video:







markr041 is offline  
post #3 of 71 Old 10-02-2013, 09:16 AM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Multiple questions in one thread. You don't get grading as a process? You don't get that shooting "flat" is good for grading? You don't get why grade an amateur family video? Considering that you also have answered some of these questions I am not sure what exactly is the problem.

One thing I know: with 8-bit per color component, it is better to shoot as close to your intended look as possible.
Ungermann is offline  
post #4 of 71 Old 10-02-2013, 09:37 AM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Quote:
Originally Posted by Ungermann View Post

Multiple questions in one thread. You don't get grading as a process? You don't get that shooting "flat" is good for grading? You don't get why grade an amateur family video? Considering that you also have answered some of these questions I am not sure what exactly is the problem.

One thing I know: with 8-bit per color component, it is better to shoot as close to your intended look as possible.

I agree with your last statement. I think you know I know what grading is and that I know that if you are going to grade shooting flat may be desirable (but note that if you shoot too flat you actually cannot retrieve the color (8-bit)).

The questions are why these guys do it as standard practice with 8-bit cameras (and recommend it to others) and appear to like their (to me) obviously unappealing results. Are they clueless about the results or am I?

The bottom line is you agree that this practice, with 8-bit cameras, of deliberately "correcting" or "coloring" in post and shooting in-camera dull, is not obviously desirable. Anyone care to defend it?

Do you think if I posted my op in the forum where this guy posted his videos I would be slammed as a know-nothing troll?

Oh, and here is a color-graded kid soccer-game video, complete with almost all action scenes out of focus, posted in the same forum about which there were only positive responses or questions about how flat he shot!

https://vimeo.com/75760934#at=0

Notice again there is no natural sound track, just some music added. I want to shout - get a camcorder! Learn audio! take an art class, or maybe not...
markr041 is offline  
post #5 of 71 Old 10-02-2013, 01:12 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Quote:
The bottom line is you agree that this practice, with 8-bit cameras, of deliberately "correcting" or "coloring" in post and shooting in-camera dull, is not obviously desirable.
I think that with equipment that we have in our disposal (low bitrate, low bith depth, low color) it is better to shoot as close to intended look as possible: color-wise and luma-wise. "Flat" usually refers to luma, when near-black and near-white regions have more bits to offer more gradations to shadows and highlights. This is more cinema-like profile, preserving highlighs even when you overexpose. But obviously in a world of limited bits, the mids suffer. Depending on scene, you may not have many mids, so this approach works. Some people like me like to see details in shadows and highlights, others do not, they prefer to crush them and crank up the contrast. For such a look shooting flat is pointless, IMO.

I actually liked the colors in the soccer video, but I totally agree with you that shooting is sloppy. Also, rolling shutter artefacts are very noticeable with handholding and fast panning.

Sometime in the last decade it almost became a norm to start with unfocused scene and then focus while shooting (and viewers watching) and while doing this even missing correct focus multiple times. You can see it on TV all the time now, it is en vogue. I hate it. I like a good focus pull as everyone else, but it must make sense. If the scene commands the shot to start unfocused because there is something unknown, uncertain, scary, etc then I am ok with that. But making every second shot like this only because your camera can do shallow focus is ridiculous.
Ungermann is offline  
post #6 of 71 Old 10-02-2013, 01:23 PM
AVS Special Member
 
slimoli's Avatar
 
Join Date: Nov 2001
Location: Miami area
Posts: 4,682
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 23 Post(s)
Liked: 30
I have nothing to add here, just wanna thank you both for posting very informative discussion, as usual. You are a real asset here.

Standard Definition Causes Lung Cancer, Heart Disease, Emphysema, And May Complicate Pregnancy
slimoli is offline  
post #7 of 71 Old 10-02-2013, 02:06 PM
Advanced Member
 
thedest's Avatar
 
Join Date: Nov 2012
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 24
8-bit is not ideal for grading, but its not impossible unless you need to push the grading a lot

there is space in the planet for graded and not graded videos.

your videos, mark, are "documentary-like". you like real-world like images, with real colors, sound etc.

but imagine a big blockbuster with documentary look. sometimes you need to grade to transmit a sensation, a mood.

I did that video with an "8-bit" camera (NEX-5R) and I think that a graded video was a better choice to get the viewer in the mood.

BTW, the soccer video is not THAT bad. I think he wanted to create a nostalgic look, something like an old video, and thats nice sometimes. I bet it means a lot for him. Sometimes people shoot videos with EXTREMELY expensive cameras and destroy the dynamic range to create a "look", blowing out higlights, crushing blacks, creating "artificial" colors etc. Thats art. Its different from documentary.

BTW (2), the wood-walk video from the GH3 looks really boring. Your video is nice, as always
thedest is offline  
post #8 of 71 Old 10-02-2013, 02:10 PM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
This was a good discussion, and I agree with everything that was said (and thanks for any compliments) - which is why I post here. When we get those in-camera 14-bit videos, 4K an/d or manageable RAW, I may play with ex post grading and change my mood...
markr041 is offline  
post #9 of 71 Old 10-02-2013, 03:06 PM
AVS Special Member
 
bsprague's Avatar
 
Join Date: Dec 2008
Location: On the Road
Posts: 2,981
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 94 Post(s)
Liked: 97
I'll add a few things. Why not.

1. What forum (elsewhere) are the posters attracting enthusiasts and even pros (having in common only evident large equipment budgets) that laud shooting flat (no color or contrast or sharpness)? I might be entertained reading it.

2. When I started here, this forum seemed to be about equipment only. I enjoy discussions about the other parts of making video.

3. "Grading", and other words describing a shift from reality, have been around as long as people have had eyes wired to their brains. Favorite personal examples are works by Ansel Adams, Vincent van Gogh and Martin Scorsese. Adams' Yosemite work, van Gogh's Twelve Sunflowers and Scorsese's "Hugo" don't look real. The look better than real. They are imaginative alterations of reality that I like to look at and invoke emotions beyond the real image.

4. Shooting "real" video (and photos) is certainly a worthwhile goal. A friend won't show anyone a picture he doesn't think is "real". He has an expensive gadget to calibrate his monitor to his printer. My goal is to make the print evoke emotional response from the viewer.

5. There is no question that a lot of skill goes into making a HD TV screen look real when viewing video. It might be called a "National Geographic" approach. How close can the viewed image be to the real scene? But is exercising the technology to perfection always the primary goal? This is science. (I watch all of Mark's videos because he is so good at the National Geographic style!)

6. Art is not science. I like to think of grading, or any other part of editing, as potentially additive. The video or photo can get better when editing ("grading") is done right. Watch "Hugo" on a 3D TV. Scorsese apparantly does not see grading and editing it always as a negative manipulation of the primary footage. On the still photo side of graphic arts, there is raging debate over the HDR process. Some see it as an ethical violation of nature.

7. I would love to have consumer priced equipment that could make video grading as approachable as HDR photography. It's truly a lot of fun.

I think I'll put on the Sony glasses and watch "Hugo" again. Then maybe "Avatar".
brunerww likes this.
bsprague is offline  
post #10 of 71 Old 10-02-2013, 03:43 PM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Quote:
Originally Posted by bsprague View Post

I'll add a few things. Why not.

1. What forum (elsewhere) are the posters attracting enthusiasts and even pros (having in common only evident large equipment budgets) that laud shooting flat (no color or contrast or sharpness)? I might be entertained reading it.

2. When I started here, this forum seemed to be about equipment only. I enjoy discussions about the other parts of making video.

3. "Grading", and other words describing a shift from reality, have been around as long as people have had eyes wired to their brains. Favorite personal examples are works by Ansel Adams, Vincent van Gogh and Martin Scorsese. Adams' Yosemite work, van Gogh's Twelve Sunflowers and Scorsese's "Hugo" don't look real. The look better than real. They are imaginative alterations of reality that I like to look at and invoke emotions beyond the real image.

4. Shooting "real" video (and photos) is certainly a worthwhile goal. A friend won't show anyone a picture he doesn't think is "real". He has an expensive gadget to calibrate his monitor to his printer. My goal is to make the print evoke emotional response from the viewer.

5. There is no question that a lot of skill goes into making a HD TV screen look real when viewing video. It might be called a "National Geographic" approach. How close can the viewed image be to the real scene? But is exercising the technology to perfection always the primary goal? This is science. (I watch all of Mark's videos because he is so good at the National Geographic style!)

6. Art is not science. I like to think of grading, or any other part of editing, as potentially additive. The video or photo can get better when editing ("grading") is done right. Watch "Hugo" on a 3D TV. Scorsese apparantly does not see grading and editing it always as a negative manipulation of the primary footage. On the still photo side of graphic arts, there is raging debate over the HDR process. Some see it as an ethical violation of nature.

7. I would love to have consumer priced equipment that could make video grading as approachable as HDR photography. It's truly a lot of fun.

I think I'll put on the Sony glasses and watch "Hugo" again. Then maybe "Avatar".

The issue is simple: you better be a very good artist if you are going to grade. You need just be a technician to produce a realistic video - there is an objective, not artistic, standard. What is photographed, how it is framed and sequenced is where the artistry is in that case. Watching technical people insist on grading, while mostly producing awful-looking boring videos (e.g., the woods video) may make some uneasy about what they should be trying to do. Shooting flat and then grading, with 8-bit equipment, is not the standard. Grading and HDR are artistic tools that when well done can be great. They are not a standard.

Btw, I thought Ansel Adams' "previsualization" philosophy was precisely to get it to look the way you want in the field - no cropping, no nothing. He is not an example of an ex post grader. Reality can evoke an emotional response - there is still framing and subject matter.

The two examples you gave are 3D. I would love it if everyone would watch 3D videos. I tried that, but no one will watch. 3D can add more realism, or provide another dimension for artistry.

Here is a link to one forum: http://www.personal-view.com/talks/discussion/7091/gh3-best-video-settings/p8
brunerww likes this.
markr041 is offline  
post #11 of 71 Old 10-02-2013, 04:15 PM
AVS Special Member
 
bsprague's Avatar
 
Join Date: Dec 2008
Location: On the Road
Posts: 2,981
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 94 Post(s)
Liked: 97
Quote:
Originally Posted by markr041 View Post

.....Btw, I thought Ansel Adams' "previsualization" philosophy was precisely to get it to look the way you want in the field - no cropping, no nothing. He is not an example of an ex post grader. Reality can evoke an emotional response - there is still framing and subject matter....
My personal experience viewing gallery prints of Adams' work was that they were surreal. As I understand it, there was plenty of "post processing" manipulation in his work. In a darkroom, we called it burning and dodging. Developing times for both film and print paper were altered as needed as well. I've forgotton a lot of the details, but depending on the measured range of light intensity, develop times were adjusted.

I tried for awhile to use his "Zone System" in the darkroom and struggled. Now the room lights are on and I struggle in Lightroom!

Of interest to this discussion of post processing (perhaps) is that I have a 40 year old print I did with my best efforts at darkroom skill. I used every bit of "grading" skill I had. I dug out the negative from my files and scanned it on a $150 Epson. I ran it through Lightroom, did some B&W computer "grading" and printed it on a new Canon printer. I had thousands invested in my 70's darkroom. I have only a few hundred invested in my scanner, printer and software. The new print is twice as good as the old print.
bsprague is offline  
post #12 of 71 Old 10-02-2013, 04:24 PM
Advanced Member
 
thedest's Avatar
 
Join Date: Nov 2012
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 24
Quote:
Originally Posted by bsprague View Post

Of interest to this discussion of post processing (perhaps) is that I have a 40 year old print I did with my best efforts at darkroom skill. I used every bit of "grading" skill I had. I dug out the negative from my files and scanned it on a $150 Epson. I ran it through Lightroom, did some B&W computer "grading" and printed it on a new Canon printer. I had thousands invested in my 70's darkroom. I have only a few hundred invested in my scanner, printer and software. The new print is twice as good as the old print.

LMAO
thedest is offline  
post #13 of 71 Old 10-02-2013, 04:27 PM
AVS Special Member
 
bsprague's Avatar
 
Join Date: Dec 2008
Location: On the Road
Posts: 2,981
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 94 Post(s)
Liked: 97
Quote:
Originally Posted by thedest View Post

LMAO
I didn't think it was that funny. It was about "grading" a scanned image!
bsprague is offline  
post #14 of 71 Old 10-02-2013, 04:35 PM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,807
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 104
Quote:
Originally Posted by markr041 View Post


Here is an example of a color-graded woods walk video proudly posted to show the guy's skill and camera performance in a forum on the GH3:

He takes an extremely dull video and makes it only very dull. Notice the same phony sound track - canned nature noise and synthesizer music (you can tell the nature sounds are not from the scenes - no running water sounds when you see running water, no change in sound character across clips).
I think you start the thread n the wrong premises by using an example videos of someone doesn't do much changes to the color and doesn't show that he is particularly good at color grading.

Your piece shows that you colors for that kind of subject better than him.
Quote:
Originally Posted by markr041 View Post

I agree with your last statement. I think you know I know what grading is and that I know that if you are going to grade shooting flat may be desirable (but note that if you shoot too flat you actually cannot retrieve the color (8-bit)).
It is not possible to shoot any camera "too flat" or "so flat you can't retrieve the colors" in any camera using proper editing/grading tools.
Quote:
The questions are why these guys do it as standard practice with 8-bit cameras (and recommend it to others) and appear to like their (to me) obviously unappealing results. Are they clueless about the results or am I?
They are just bad at color grading. You don't like their grading.
To shoot flat is wisest, just to avoid letting "some dude in Japan" post process a look in camera and bake in settings you can't undo in your material.
Better to leave that to you and your computer.
Quote:
The bottom line is you agree that this practice, with 8-bit cameras, of deliberately "correcting" or "coloring" in post and shooting in-camera dull, is not obviously desirable. Anyone care to defend it?
Any video or photo amateur that have shooting images as a hobby like to post process the results and create "their personal art". That they are not great artist and maybe never will be isn't something one should criticize them for.
The most talented of them learn a lot by this trail and error and might become the next most sought after color grader for Hollywood block busters. It is really the only way to learn, you can't go to school and learn color grading.

Quote:
Do you think if I posted my op in the forum where this guy posted his videos I would be slammed as a know-nothing troll?
Why bring the guy down by pointing out he his neither talented in color grading nor in film making?
If you go there to give him tips on how to do it better, I am sure you might be welcome, but that depends of the type of forum.
He might only need to calibrate his monitor and maybe buy a better one.

Quote:
Oh, and here is a color-graded kid soccer-game video, complete with almost all action scenes out of focus, posted in the same forum about which there were only positive responses or questions about how flat he shot!

https://vimeo.com/75760934#at=0

Notice again there is no natural sound track, just some music added. I want to shout - get a camcorder! Learn audio! take an art class, or maybe not...
Maybe it is you that need a class in cinematic arts and not the guy experimenting and training his skills at a football game, because he seems quite aware of what he is doing and why.

I think he has done wonders with putting atmosphere into a simple football game for kids (his nephew) and his focus pulls are quite impressive.
The focus pulls are done very intentionally and he hits focus at the right moment nearly every time, pulling the viewer into the film.
Study it and see if you can see it as something else than missed focus.



As for 8-bit color and grading from DSLR, this should be a simple example that show that shooting flat and recover the color and contrast should do wonders.

Professional Grading; Summer Blockbuster Colour Grading Tutorial.
http://vimeo.com/65617394#at=0



Jake Blackstone - MOD Color is a very good professional color grader showing a wide range of styles, this is his latest demo reel.
http://vimeo.com/72235893

I don't know why these Vimeo links don't load.
coolscan is offline  
post #15 of 71 Old 10-02-2013, 05:57 PM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Thanks for the comments, most of which are useful. I disagree strongly with this statement: "To shoot flat is wisest, just to avoid letting "some dude in Japan" post process a look in camera and bake in settings you can't undo in your material.
Better to leave that to you and your computer."

You have many, many controls in camera to customize the look as well as custom looks you can download into the camera made by other than "some dude in Japan,", and you have the reality right there to compare. You also then do not suffer from re-compression artifacts and the limits of 8-bit color in making changes ex post in the computer. It's a trade-off. And you are simply wrong that if you shoot super flat and super soft you can always recover all color (8-bit again) and resolution in post. You cannot.

It is precisely the notion that shooting flat is "wisest" that I object to. Wisest for whom? - not your average user of cameras, not evidently for most of the more-intense users who post to video forums judging by the results they proudly display. As I said, there are some good examples of coloring (you have posted some examples - I like the Blackstone stuff), there are good examples without it, and it is often way overdone.

And, what is the excuse for the lack of audio - laziness? incompetence? or is this art too? Or the "wisest" practice: just add some canned music in post and some sound effects.

It is fun to disagree on the soccer video; looks to me like he can't focus on what he wants; looks to you like creative focus pulls. Given the camera used, I bet on the first. And I truly do not get why a "nostalgia" coloring (one of the many pre-set "look's" plug-ins you can buy) is particularly relevant to the mood of a kids' soccer game. So I see use of the wrong camera and an inappropriate plug-in and no audio; you see art. But you are not wrong; it's just your tastes..
markr041 is offline  
post #16 of 71 Old 10-03-2013, 03:38 AM
AVS Special Member
 
brunerww's Avatar
 
Join Date: May 2011
Posts: 1,019
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 31
I agree with slimoli. Lots of good ideas to think about here. Thanks for the thoughtful discussion, everyone.

Personally, I shoot saturated because my 8 bit camera gives me sufficient control over color in the camera, but the new 10-bit camera is supposed to arrive in the mail today, so I guess I'll have to figure out this whole "colorist" thing. Looking forward to it smile.gif
brunerww is offline  
post #17 of 71 Old 10-03-2013, 09:02 AM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Quote:
Originally Posted by markr041 View Post

It is fun to disagree on the soccer video; looks to me like he can't focus on what he wants; looks to you like creative focus pulls. Given the camera used, I bet on the first. And I truly do not get why a "nostalgia" coloring (one of the many pre-set "look's" plug-ins you can buy) is particularly relevant to the mood of a kids' soccer game. So I see use of the wrong camera and an inappropriate plug-in and no audio; you see art. But you are not wrong; it's just your tastes..
I believe you have seen my Sedona video. Starting from 2:09 I used "1960-ies 16mm film" preset or something similar to reduce contrast and to emphasize the sandy feel. I also used overlay music. What do you think about it?

I could have kept the original sound and look but I wanted to convey a feeling of vast space and redrock mountains and add some nostalgia, yep. The whole thing has been shot on a 1/6-inch camcorder.

Ungermann is offline  
post #18 of 71 Old 10-03-2013, 09:36 AM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,807
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 104
Quote:
Originally Posted by markr041 View Post

Thanks for the comments, most of which are useful. I disagree strongly with this statement: "To shoot flat is wisest, just to avoid letting "some dude in Japan" post process a look in camera and bake in settings you can't undo in your material.
Better to leave that to you and your computer."

You have many, many controls in camera to customize the look as well as custom looks you can download into the camera made by other than "some dude in Japan,", and you have the reality right there to compare. You also then do not suffer from re-compression artifacts and the limits of 8-bit color in making changes ex post in the computer. It's a trade-off. And you are simply wrong that if you shoot super flat and super soft you can always recover all color (8-bit again) and resolution in post. You cannot.
First; Lets different between super-flat and super-soft, that is two different things.

To shoot super-flat (depending on the camera) is just meta data without boost of in-camera colors and contrast setting. I have never heard anybody claiming that colors are not fully recoverable, and that goes against the principles of digital capture of data from a CMOS sensor.
You have to come up with some links to facts that prove you are right about this, and such facts would surprise everybody in the motion capture community.
If your claim was right then Technicolor should retract its Cine Style picture style settings for Canon DSLRs, as that is made to get most DR out of the capture files.
Quote:
It is precisely the notion that shooting flat is "wisest" that I object to. Wisest for whom? - not your average user of cameras, not evidently for most of the more-intense users who post to video forums judging by the results they proudly display.
By shooting Flat it is easier to control that you don't blow highlights or underexpose shadows because you see what the camera is capable of in the viewfinder/EVF.
The problem is that the file you get is not RAW and has a baked in look where there is less possibilities to recover problems.
And if you are going to use one of the many pre-made plug-in looks, it is easier to make them work if the footage doesn't have a baked in look from the camera.

But you are probably right that many people should rather stick to the settings and look that "some dude in Japan" has set for them, just because they are bad at color grading.

If one could afford more expensive cameras with RAW capture, one would always have Flat files, and can in addition set a pre-set look for the shoot in the viewfinder like they do on those cameras.
Quote:
And, what is the excuse for the lack of audio - laziness? incompetence? or is this art too? Or the "wisest" practice: just add some canned music in post and some sound effects.
It is just a choice. I am a little surprised you mention that, because I see pro-videos almost weekly that does this, everything from surfer videos to drama type of videos.
Quote:
It is fun to disagree on the soccer video; looks to me like he can't focus on what he wants; looks to you like creative focus pulls. Given the camera used, I bet on the first.
He is shooting WFO which gives a shallow DOF to separate the actions focus point from everything around. He does it very consciously by starting (mostly) out-of-focus and reaching focus when the action he want to focus on happens.
He also does it with a good sense of rhythm and feel of the tempo of the action.
So I will disagree with you that it is a result of accidents or lack of ability to hit focus because of the camera.

The shallow DOF style, often overdone, comes from feature filming and the possibility for shallow DOF was very much the reason people got so excited about Canon 5D II video and started the DSLR video industry.
Very much thanks to Vincent Laforet's little short REVERIE.

Before that people had been using special adapters on video cameras to mimic the cinematic shallow DOF, as 35mm film (and some only for rent expensive digital cameras) was the only and more expensive alternative.
The big sensor in the 5DII did that, and you get acquire a similar look in the micro-four-thirds cameras, but not on the very small sensors of the video cameras.

You seem more of the deep-DOF shooter in the style more often used in Natural History documentaries, but you surely must be aware that the Shallow-DOF style is something probably a majority of photographers use and like for separation and dramatic effect in all kinds of projects.
Maybe you should be more daring and try out all of these techniques, that you dislike so much, in your own videos and see if some of them benefit the quality and artistic style of you videos.
Often a specific taste is something one has to acquire over time after trying them out repeatedly.


But at the same time I will do a shout-out to THEDEST for the video he posted in this thread, where the Look he use in that video, which is probably quite far from the Natural Look of the location, makes the video much more watchable (I believe without seeing the raw footage) even if it is a documentary.
Quote:
And I truly do not get why a "nostalgia" coloring (one of the many pre-set "look's" plug-ins you can buy) is particularly relevant to the mood of a kids' soccer game. So I see use of the wrong camera and an inappropriate plug-in and no audio; you see art. But you are not wrong; it's just your tastes..
Call the plug-in look nostalgic or maybe sun-drenched, but yes it is a taste, but if you where shooting a similar scenario you would prefer more natural colors of a documentary style presentation with natural audio which all would tell how life was that day.

Lets revisit Vincent Laforet's Reverie which started the DSLR video industry and changed Vincent Laforet's life (and many other photographers and film-makers) forever.
The funny thing is that Canon's engineers had no idea what they where releasing, and hadn't it been for the fact that Laforet got hold of a preproduction 5DII and hurriedly made this little "kitschy perfume add spoof" it might have taken people much longer to discover the possibilities. The 5DII was not he first DSLR with video capabilities.

It even opens with a shot that goes out of focus. smile.gif

http://vimeo.com/7151244
coolscan is offline  
post #19 of 71 Old 10-03-2013, 10:37 AM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Thanks for the info. There are still a few things I think you are getting not quite right.

"By shooting Flat it is easier to control that you don't blow highlights or underexpose shadows because you see what the camera is capable of in the viewfinder/EVF."

Nonsense. That's why we have aids in the camera - you have heard of zebra stripes, false color and vectorscopes, right? You do not need to shoot flat to get exposure right (meaning set to what you want, given the limited dynamic range); that's preposterous.

"The problem is that the file you get is not RAW and has a baked in look where there is less possibilities to recover problems."

The problem is precisely that the file you get is not a RAW file - so you have to deal with a highly compressed 8-bit file that will degrade with a lot of mucking around in post. The customizable in-camera modifications work with the RAW data. The quality out of the camera will always be better, for the same "look."

"And if you are going to use one of the many pre-made plug-in looks, it is easier to make them work if the footage doesn't have a baked in look from the camera."

I agree. The issue is whether it is "wisest" to shoot that way with 8-bit, highly-compressed video. And it is not, though it is certainly a viable alternative; just not necessarily better.

"But you are probably right that many people should rather stick to the settings and look that "some dude in Japan" has set for them, just because they are bad at color grading."

No, you still fail to recognize that you can customize the settings in the camera to get the look you want in camera. People could be equally bad at getting good results in the editor or in the camera. Moreover, I agree that the entire team at Canon, who have been working in the field for decades, can probably provide better templates for video looks than the average shooter.

"If one could afford more expensive cameras with RAW capture, one would always have Flat files, and can in addition set a pre-set look for the shoot in the viewfinder like they do on those cameras."

I am all for RAW video, which would solve the problem - no trade-off whatsoever to making adjustments out of camera. In the meantime almost all video cameras are giving us 8-bit video that is highly compressed. So, there is no obvious one way of shooting video with them.

Finally, there are endless debates on forums on whether it is best to shoot flat using 'Natural" or "standard" (pulling sharpness, contrast, and color, all the way down), for example, based on the fact that they cannot get reasonable skin colors in post no matter what they do with some flat settings. And they debate whether Flaat or technicolor or whatever give best results. These "flat" settings are evidently not the same as RAW video, not even close. And, btw, I have shot (and posted) RAW video and enjoyed setting colors in an editor to my taste. It is fun, and no worry about degradation. I am waiting for a reasonable workflow to continue. The EOS M shoots RAW; you do not need an expensive camera to at least try it out.
markr041 is offline  
post #20 of 71 Old 10-03-2013, 10:56 AM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Quote:
Originally Posted by coolscan View Post

To shoot super-flat (depending on the camera) is just meta data without boost of in-camera colors and contrast setting. I have never heard anybody claiming that colors are not fully recoverable, and that goes against the principles of digital capture of data from a CMOS sensor.
Could you elaborate on that? I mean, on the principles and stuff.
Quote:
Originally Posted by coolscan View Post

By shooting Flat it is easier to control that you don't blow highlights or underexpose shadows because you see what the camera is capable of in the viewfinder/EVF. The problem is that the file you get is not RAW and has a baked in look where there is less possibilities to recover problems. And if you are going to use one of the many pre-made plug-in looks, it is easier to make them work if the footage doesn't have a baked in look from the camera.
How "super-flat" is not just another "baked-in" look? You get more in shadows and highlighs, but you lose in mids. It is a tradeoff, just like any other profile.
Quote:
Originally Posted by coolscan View Post

He is shooting WFO which gives a shallow DOF to separate the actions focus point from everything around. He does it very consciously by starting (mostly) out-of-focus and reaching focus when the action he want to focus on happens.
Nah, IMO he starts out of focus simply because he sees action, presses Shoot button, and his focus is not there.
Ungermann is offline  
post #21 of 71 Old 10-03-2013, 04:04 PM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,807
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 104
Quote:
Originally Posted by markr041 View Post

Thanks for the info. There are still a few things I think you are getting not quite right.

"By shooting Flat it is easier to control that you don't blow highlights or underexpose shadows because you see what the camera is capable of in the viewfinder/EVF."

Nonsense. That's why we have aids in the camera - you have heard of zebra stripes, false color and vectorscopes, right? You do not need to shoot flat to get exposure right (meaning set to what you want, given the limited dynamic range); that's preposterous.
Well, not all cameras has that, or you need to use a hack like the magic Lantern one for Canon DSLR.
But then you are going into at least as advanced use as setting a flat look in the camera, so why argue this at all?
Quote:
"The problem is that the file you get is not RAW and has a baked in look where there is less possibilities to recover problems."

The problem is precisely that the file you get is not a RAW file - so you have to deal with a highly compressed 8-bit file that will degrade with a lot of mucking around in post. The customizable in-camera modifications work with the RAW data. The quality out of the camera will always be better, for the same "look."
No reason that the file should be degraded by mucking around in post if you know what you are doing.
The in-camera look is set there with elevated sliders for contrast and saturation and sometimes sharpening.
Why is it worse to do that in a computer than let the camera do this?

A computer is more sophisticated than a camera, and the right software has more and more detailed sliders to control the results.
In the end, if you set all image control sliders in the camera to Zero, you get closer to what the camera really capture, the sliders are just gain controls.

To say that the cameras "computer" is doing a better job than a PC is quite contradictory.
Quote:
"But you are probably right that many people should rather stick to the settings and look that "some dude in Japan" has set for them, just because they are bad at color grading."

No, you still fail to recognize that you can customize the settings in the camera to get the look you want in camera. People could be equally bad at getting good results in the editor or in the camera. Moreover, I agree that the entire team at Canon, who have been working in the field for decades, can probably provide better templates for video looks than the average shooter.
Depends on the look you want, if you have set the desired look before shooting or if the look you want is possible to set in the camera, which in most (or all) cameras are not possible.

I have three preset looks + the standard look in my 7D which I can choose from depending on the subject and light conditions. The three presets are variations of flat looks.

The standard look is so baked in and is quite impossible to rework in post because of the elevated contrast easily causing blown highlights.
I also set white balance to auto and sunlight after finding that such setting gives me an more even colored white balance if the light is shifting due to passing clouds.

This just to illustrate that a baked in look can be used to ones advantage as to ones disadvantage.

There are so many options, choices and choices to be made every time one are out shooting.
And that is very much what photographic and filming techniques are all about, make the right choice of combinations for every new motive.
Quote:
Finally, there are endless debates on forums on whether it is best to shoot flat using 'Natural" or "standard" (pulling sharpness, contrast, and color, all the way down), for example, based on the fact that they cannot get reasonable skin colors in post no matter what they do with some flat settings. And they debate whether Flaat or technicolor or whatever give best results. These "flat" settings are evidently not the same as RAW video, not even close. And, btw, I have shot (and posted) RAW video and enjoyed setting colors in an editor to my taste. It is fun, and no worry about degradation. I am waiting for a reasonable workflow to continue. The EOS M shoots RAW; you do not need an expensive camera to at least try it out.
It is great that RAW for video is coming along, but surprisingly slow.
When Magic Lantern can get RAW video by hacking Canon DSLR, one can just imagine what Canon engineers could have done with the cameras had they wanted to, and speculate why they haven't done this long ago.

RAW has been the stable for DSLR stills for more than decade, but when RED released a camera that shot RAW video they got a lot of back-clash from the movie industry for years. People didn't even know what RAW really was (obviously never owned a DSLR shooting stills) and criticized RED for not providing baked in files like ProRes.
DPs and Editors didn't know what to do with RAW files and blamed RED cameras for their own incompetence.
First now the year or so that RAW has become the desirable format for video shooting too.

But it is still more enthusiastically received by the amateur hobby shooters than in the movie industry.
coolscan is offline  
post #22 of 71 Old 10-03-2013, 04:31 PM
Advanced Member
 
thedest's Avatar
 
Join Date: Nov 2012
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 24
Quote:
Originally Posted by coolscan View Post


No reason that the file should be degraded by mucking around in post if you know what you are doing.
The in-camera look is set there with elevated sliders for contrast and saturation and sometimes sharpening.
Why is it worse to do that in a computer than let the camera do this?

A computer is more sophisticated than a camera, and the right software has more and more detailed sliders to control the results.
In the end, if you set all image control sliders in the camera to Zero, you get closer to what the camera really capture, the sliders are just gain controls.

To say that the cameras "computer" is doing a better job than a PC is quite contradictory.

Actually thats a wrong observation.

The BIG difference is that the camera is baking the RAW video. When you do that in your PC you are baking an AVCHD file. THATS the big difference, not the processing power or the algorithms
thedest is offline  
post #23 of 71 Old 10-03-2013, 04:34 PM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,807
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 104
Quote:
Originally Posted by Ungermann View Post

Could you elaborate on that? I mean, on the principles and stuff.
It is not much more to say than what I said.
The camera takes the RAW data off the sensor and convert i to .Mov or some such format. If you don't boost the settings in the camera you get closets to the RAW data the camera records.
Then it is just a matter for you to choose f-stop and ISO depending of it is most important to protect highlights or avoid too dark shadows.
Quote:
How "super-flat" is not just another "baked-in" look? You get more in shadows and highlighs, but you lose in mids. It is a tradeoff, just like any other profile.
"Super-flat" is as wide dynamic range the camera can record and is more of a "anti baked in look" than a baked in look.
You can't loose any mid levels. If you have max highlights and max shadows, the mids also have to be there if they are there in the motives.
The type of cameras we talk about doesn't have such a wide dynamic range that the mid levels can disappear between the highs and the lows.
In the end it is all converted meta data from the sensor, and if it is not fiddled with in the conversion process, it is just what it is.
Quote:
Nah, IMO he starts out of focus simply because he sees action, presses Shoot button, and his focus is not there.

Study it some more. He probably shot 90 minutes of footage and made a short film of the shots where he succeed, so he probably missed a lot of focus.
To me it is apparent that he is very conscious of what he is doing.

->->->
Apropos skin color; Came to think of this tutorial that might be interesting for people that have problems with skin colors, which I guess everybody have at some point.
This is about using Vectorscope in Premiere Pro. This is video #49, there are more about use of Vectorscope in color grading, specific to skin color in this one.
The whole s tutorial series I believe have 60 video tutorial, so I am sure there are more interesting tips one can find.

.#49 and maybe some before or between the next posted.

#53
coolscan is offline  
post #24 of 71 Old 10-03-2013, 04:40 PM
AVS Special Member
 
coolscan's Avatar
 
Join Date: Jan 2006
Posts: 1,807
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 104
Quote:
Originally Posted by thedest View Post

Actually thats a wrong observation.

The BIG difference is that the camera is baking the RAW video. When you do that in your PC you are baking an AVCHD file. THATS the big difference, not the processing power or the algorithms

So what do you mean is best?

Let the camera bake in a strong look that you can not undo in your PC if you have regrets, ......... or let the camera do a minimum of baking, that you easier can add to on our PC?
coolscan is offline  
post #25 of 71 Old 10-03-2013, 04:57 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Quote:
Originally Posted by coolscan View Post

RAW has been the stable for DSLR stills for more than decade, but when RED released a camera that shot RAW video they got a lot of back-clash from the movie industry for years. People didn't even know what RAW really was (obviously never owned a DSLR shooting stills) and criticized RED for not providing baked in files like ProRes. DPs and Editors didn't know what to do with RAW files and blamed RED cameras for their own incompetence. First now the year or so that RAW has become the desirable format for video shooting too.
REDCODE has bit depth to allow for recording without gamma correction. 8-bit cameras that we use do not have this luxury, so they use this or that gamma profile, so the picture is still "baked-in" using your terminology. And again because of 8-bit depth, any extra sensor latitude will either be lost, or will cause banding, because luminance changes larger than 1% is seen to a human eye as a jump, banding, not as a smooth change. Any tricks that claim to offer to preserve or even to recover more information from the video in 8-bit space ultimately rob it in some other place. If you save more shadows and highlights you lose more mids. If you extend your dynamic range you get banding. The well-received Sony's S-LOG is just another gamma profile. Any professional camera has several profiles to choose from, and an option to create your own. Shooting "super-flat" on a 8-bit camera simply means using a gamma profile with a cinema-like knee.
Quote:
You can't loose any mid levels. If you have max highlights and max shadows, the mids also have to be there if they are there in the motives.
You have only 255... actually only 255 - 16 - 20 ~ 220 levels from black to white. This is not a lot. With all the gamma-correction tricks you can get at most 8 stops from this range. Any talk about 10 or 12 stops on a camera that records into 8-bit format is nonsense. OR, you will get 12 stops, but you will get banding because jumps are larger than 1%. The only solution is more bits, which RED has in its format. Panasonic has 10 bits too, and I believe that S-AVC is also 10-bit format. I don't object your choice to preserve shadows and highlights, I do this myself whenever I can, but just don't delude yourself that you get this for free simply because you are shooting "flat".
Ungermann is offline  
post #26 of 71 Old 10-03-2013, 05:17 PM - Thread Starter
AVS Special Member
 
markr041's Avatar
 
Join Date: Jun 2005
Posts: 3,228
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 136
Quote:
Originally Posted by coolscan View Post

So what do you mean is best?

Let the camera bake in a strong look that you can not undo in your PC if you have regrets, ......... or let the camera do a minimum of baking, that you easier can add to on our PC?

You just gave it away - get it right in the camera (manipulated RAW video), and you have the best (compared to manipulating 8-bit, highly compressed video no matter how minimally baked). Look for cameras that have a lot of customizing tools and exposure aids so you have no regrets.

So, the lesson is to work on your camera skills for the best results (so you have no regrets), until we get RAW video out of the camera, when all those computer software tools can come into play with no damage. Or maybe, lightly compressed, 4:2:2 video from the camera might be ok too (Ninja anyone?). Minimize regrets that's the ticket! Now, many cannot obtain the best, so then you can choose your poison. But it is clear what is best.

Oh, and don't shoot sports with a DSLR; no way it can autofocus on fast-moving players or you can use manual focus to follow them either. But you can always pretend your oof clips are art, or just show us the ones you by luck happened to get in focus (short video). Get real; I bet you never shot video at a sports event. It is incredibly challenging to do well, and you cannot correct focus in post, so you have to get it right in the field. Of course, you can obscure oof clips with hazy, strange colors and extra soft video...
markr041 is offline  
post #27 of 71 Old 10-03-2013, 05:38 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
One can see shallow DOF beauty shots of sports events all the time, in slow motion to boot: those pro cameras have huge zooms and the cameramen are super-pros, keeping the subject perfectly in focus while following it. But they have multiple cameras and a whole separate control center to choose a particular feed, and at least one camera or more shoots wide all the time. And they don't use cameras with rolling shutter for what I know.

I saw a very nice shallow-DOF go-kart video on Vimeo. Show with the Canons if I am not mistaken. It was staged, so it looked quite good. Cannot find it anymore, Vimeo sucks in search. It has probably the worst search engine in the whole Universe.
Ungermann is offline  
post #28 of 71 Old 10-03-2013, 05:41 PM
Advanced Member
 
thedest's Avatar
 
Join Date: Nov 2012
Posts: 618
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 24
Quote:
Originally Posted by coolscan View Post

So what do you mean is best?

Let the camera bake in a strong look that you can not undo in your PC if you have regrets, ......... or let the camera do a minimum of baking, that you easier can add to on our PC?

Actually there is no "BEST". You guys are too extreme. You sound like old grannies.

The answer is: do what you think that looks BEST for YOUR video. Some guys like documentary-look videos and some like alternative coloring.

8-bit 4:2:0 sucks baked in the camera and baked in the PC - there is no way out!! Even those videos posted here on the forum by cameras like the EOS M look really ugly to my eyes now that I learned to see the difference between 4:4:4, 4:2:2, 4:2:0 and latitude. Even the in-camera processed videos hurt my eyes with the lack of DR and colors that look TOTALLY artificial, digital and sometimes they even look fluorescent.

Example of a PC-baked video that looks nice. Shot with a NEX-7

.

Another one. Shot with a NEX-5

.

.

Both were shot in a flat profile and baked in the PC

But the real answer is: the 8-bit 4:2:0 era is OVER!!!!!!!!!!! STOP BUYING THOSE CAMERAS!!!!
thedest is offline  
post #29 of 71 Old 10-03-2013, 05:46 PM
AVS Special Member
 
xfws's Avatar
 
Join Date: Jun 2008
Posts: 1,459
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 26
I see the points on both sides of this... but what's being left out is that the final delivery with consumer cams will typically be to the internet. The content is relevant.

If for a narrative short, when adding a preset color or manually grading; it can set a certain tone. It's better to have those options in post even if sacrificing some quality that will not be discernible to the average internet viewer. Not many will be pixel-peeping, except video enthusiasts on these boards.

If it were pro, and the expectations that come along with that - there would be better cameras/software involved with perhaps delivery to a big screen or TV.

I can see keeping it ungraded for the walk-around videos posted to the internet where the focus is solely on the visuals, especially the nature shots where leaves on trees tend to get mucked-up.

It wouldn't be practical with narrative content to lack a flat consistency from shot-to-shot and not be able to change things in post as a separate process.
xfws is offline  
post #30 of 71 Old 10-03-2013, 05:57 PM
AVS Special Member
 
Ungermann's Avatar
 
Join Date: Aug 2006
Location: San Diego, California
Posts: 3,869
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 82 Post(s)
Liked: 29
Quote:
Originally Posted by thedest View Post

But the real answer is: the 8-bit 4:2:0 era is OVER!!!!!!!!!!! STOP BUYING THOSE CAMERAS!!!!
Although I am not a big fan of Star Wars, Lucas shot the Episode 2 on 8-bit HDCAM. I don't think that anyone on this forum is going to shoot something grander anytime soon. It is not the equipment.
Ungermann is offline  
Reply Camcorders

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off