Your opinion - Color grade IN camera or POST

C

carlone

Well-known member
Hey Guys,

We're about to start on a new project and are getting various opinions on whether we should be shooting "Flat" or adjusting "Color" for the scene at hand. We're shooting the project in HD (720PN).

My understanding is that the best color processing possible with the HVX happens at the time of acquisition. Anything in post (while it may be needed) is in actuality denegrading the image. Again, I know you've got to Color correct in post but should you at least be trying to get it in the ballpark on set?
 
Blaines right, your better off doing everything you can to get your recorded image as close to its final form as possible.

Some people like to wait until post to decided on a certain look or color scheme for certain scenes, but I think that most of that information should be decided on before actually shooting.
 
macintosh said:
Some people like to wait until post to decided on a certain look or color scheme for certain scenes, but I think that most of that information should be decided on before actually shooting.
Click to expand...

Absolutely right. If you don't know what you want your movie to look like, you're not ready to shoot it. You need to spend more time in pre-production.
 
Damn Blaine, sounds like you are really getting this producer thing down!!

Seriously, the cost to correct something goes up at each stage of the process. Better to solve it at the pre pro phase if you can, next best is at the production phase and the least effective and often most costly (in terms of time spent and actual monetary costs) is in post phase.

Any planning that you can do in advance which will make your shoot go smother will pay you considerable dividends as you won't be burning through your or your cast and crews time unnecessarily.
 
It depends. There's the manipulation of the captured image via production design and lighting, which you should always get as close as possible to the desired look. Then there's the manipulation of the camera settings to create variants. There's a lot to be said for shooting flat (in camera) and obtaining as much information as is possible for later post work. The likelihood of you living with whatever settings you had the camera set at without any desire to change that in the edit is very remote, so you will have to address the possible image degradation anyway (uncompressed, CineForm, whatever). Remember it is a lot easier to manipulate a flat image than an already heavily stylized image.
 
I'm actually surprised that this thread has gotten since most people seem to think otherwise.

I'm a big fan of doing it in camera. Back in the DVX100 days, you simply HAD to because with 4:1:1 compression, there isn't a whole lot of room fo error. You had to get as close as you could and then take the last few steps in post.

Now, DVCProHD really does let you push your footage a lot more than you once could. I can do things with my HVX200 stuff and change it up FAR more than I ever could with something shot DV or HDV.

That said, at the end of the day, I'd rather get it right in camera. Part of it is that I think the image settings are there for a reason. Another part of it is that I'm pretty damn particular about how I want something to look and have a good idea of it when I shoot (if I didn't, I wouldn't be doing my job). I guess another part of it is arrogaunce since I'm confident I hope I know what I'm doing and like to show off my skill with the settings (makes me feel more like a pro). Most importantly though, DVCProHD is still compressed and you are going to get a better image if you do your effects BEFORE the compression.

That said, there are some things I still always do in post. I am a firm believer of over-exposing a half stop with the HVX200 to reduce noise then bringing it back down in post.
 
DavidBeier said:
I am a firm believer of over-exposing a half stop with the HVX200 to reduce noise then bringing it back down in post.
Click to expand...

That's great info David. Do you over-expose by using the scene file auto iris setting to get you close or do you eye-ball it?
 
David - what exactly do you over expose and how do you do it? (ie shadows, mids, high lights or global)

Its not like you can take a grey scale reading with a light meter then open 1/2 stop on the iris so I am curious how you are increasing exposure by this precise degree.
 
ive actually been seriously toying around with this idea for a while

Something that ive noticed..

If i shoot with a higher master pedestal in camera, and keep the tone relatively flat, i am able to do alot in post, giving me options out the wazoo.

HOWEVER, im starting to see a difference between correcting in camera and out of camera

What do i mean? correct me if im wrong but:

Shooting with a higher master pedestal will flatten the image, but at the same time it alters the overall contrast of your image.

If i try to shoot flat and crush the blacks in post, i get crushed blacks with an overall darker image. No matter how much i play with the CC in post, it is never the same as if i Crush the blacks in camera,

i noticed the highlights are much brighter when u do things in camera, if you do it in post, it affects alot of the surrounding image.

So i dunno, but i find shooting flat does give you the most options, but if you have a specific look in mind, trying to get it in camera first, and you succeed, it is very difficult to replicate that look from a flat image.

ill try to show u guys an example.
 
Shooter said:
David - what exactly do you over expose and how do you do it? (ie shadows, mids, high lights or global)

Its not like you can take a grey scale reading with a light meter then open 1/2 stop on the iris so I am curious how you are increasing exposure by this precise degree.
Click to expand...

carlone said:
That's great info David. Do you over-expose by using the scene file auto iris setting to get you close or do you eye-ball it?
Click to expand...

In short, I eyeball. In general it just means I go ahead and try to get the midtones and sometimes shadows as bright as possible WITHOUT letting the highlights blow out. I initially just tried getting it a half a stop brighter than I normally would but I've found that this isn't the best strategy all the time and now just relie on my eye. The real balancing act comes with Cinelike D. It's going to give you the most lattitude but also has the most noise. This is the one where you really want to go ahead and expose a bit more to get rid of some of that noise but, if by chance you blow out, it's the most difficult to work with since you have no knee and trying to take it down in post is going to be hard. It's taken A LOT of practice to come up with the right amount that looks good and I still have bad days when I screw things up. At the end of the day though, I'm finding that if I can stop from blowing out it really is a great way to reduce noise and get some great looking images.
 
k, here is the example

This was done in camera, Crushed master ped, i think around - 11, colorGamma down to -5 and color temp to +6


duckba5.png


And here is a clean image with a master ped of +3, regular Cinelike for both.

ducks2qv1.png
 
PaPa said:
k, here is the example

This was done in camera, Crushed master ped, i think around - 11, colorGamma down to -5 and color temp to +6


duckba5.png


And here is a clean image with a master ped of +3, regular Cinelike for both.

ducks2qv1.png
Click to expand...

Papa, great looks... You could have psoted this in the now defunct scene files thread! :)
 
It's not a totally negative thing to correct after post, either. Though it'd be awesome to work on a FinalTouch bay, which is why I'm hoping that Apple releases FinalTouch in the next full Final Cut Studio/Suite. That'll raise the bar for post-processing images.

Check this thread for great post processing on footage/images from the HVX: http://www.dvxuser.com/V6/showthread.php?t=71299

And this to see what the guy did to achieve it:

http://www.dvxuser.com/V6/showpost.php?p=690891&postcount=24

I say he did a damned good job with getting his look, after the fact.
 
cool, lol but i had a question!

haha

i find it hard to achieve the first look if i just shoot flatter and try to mimic it with some Post CC.

i want to know if there is any truth to saying that when you correct in camera, not shooting flatter, you get more control of the overall pictures contrast with the pedestal.

i find if i shoot very flatt, the highlights mesh a bit with the mids, and the mids mesh a bit with the lows, making the picture flatter, so when i crush the blacks in post, the entire image darkens, without retaining the highlights and those kinds of areas.

comments or opinions on this idea?
 
PaPa--

You are pretty-much in the ballpark when it comes to the idea that the overall image becomes dark and gloomy when you shoot flat and then crush blacks in post.

I think that's where David has a really good point about getting the look in camera: even with the DVX you seem to get an image that's a bit more pleasing when you've achieved a lot of the look in camera as opposed to post work.

That's where great lighting comes in, IMO. Because, you get it good in camera and then tweak in post.
 
Its all subjective, but anything you do in post has its costs in terms of the limited dynamic range available.

Don't count on post to save your rear, rather use it as sparingly as possible.

Test your setups (lighting and camera) to achieve as close as possible to the look you want, then if necessary, adjust in post to fine tune the look that you already have.
 
I tend to shoot clean and favor grading in post, aside from color-correction for lights, ND and polarizers on the camera.

I do generally "paint" the HVX with a fair amount of pedestal play and a minimal color saturation bump. Bringing in, and more importantly, matching a look scene-to-scene, is much easier in post when you can side-by-side your footage on a calibrated monitor under optimal viewing conditions.

It always pays to light it properly at the right color temperatures, but beyond that I vote for scene-by-scene during the edit.

e
 
khmuse said:
Its all subjective, but anything you do in post has its costs in terms of the limited dynamic range available.

Don't count on post to save your rear, rather use it as sparingly as possible.

Test your setups (lighting and camera) to achieve as close as possible to the look you want, then if necessary, adjust in post to fine tune the look that you already have.
Click to expand...

Hell yeah to that. I'm beginning to LOATHE the idea of fix-it-in-post. I've had to clean up so many horrible images at work that were supposed to go on television, and they could've been avoided if they took an ounce of time to pre-plan what the hell they were going to do.

I just don't like it.
 
You must log in or register to reply here.
Back
Top