Shooting it perfectly vs. fixing it in post.

[Scott Wilkinson]

[Please excuse my typo in the subject line, which I couldn't fix---it should say "fixing it in post."]

A rhetorical question here, just for the sake of discussion. :-) (Not a troll---really. It's a worthwhile discussion in my opinion.)

In a convo today with one of my producers, he was obsessing over lighting and skin tones...going on and on about how the lighting must be just right to render the skin tones perfectly. I started to wonder...is this really necessary?

Now let me state for the record that I'm fully aware of GIGO (garbage in, garbage out)...and I'm also aware that if footage isn't acquired with some basic standards (it's in focus, not horribly under- or overexposed, etc.) there isn't much that can be done in post (though this continues to change).

Back to my wonderings...it seems to me that these days...if someone is VERY proficient with an NLE like Premiere Pro...then a LOT can be effectively fixed in post (again---assuming the footage isn't horribly hosed to start with).

I say this because (as I've suggested elsewhere) it is staggering to me what can be accomplished with some in-depth knowledge of color correction, blending modes, etc. And I'm beginning to think that the time producers spend obsessing over getting the shot perfect...might be better used learning more about color correction and the myriad other techniques in apps like Premiere Pro for fixing things.

To put it differently, few people these days question what can be done to a still image with Photoshop. Photoshopped images are fast on their way to becoming acceptable even in journalism (as long as they don't distort the facts), and people routinely correct poorly-shot still images with dramatic and beautiful results. Video is, of course, nothing but a bunch of rapidly moving still images---anything that can be done to one can be done to the other.

I'm not arguing that people shouldn't have a good understanding of lighting, color temperatures, etc...of course they should. What I'm arguing is a matter of degree...and more specifically, I'm suggesting there is a continuum between shooting how you want it and fixing it in post.

I'm also suggesting that (in some organizations, anyway) there are some old-school types who seem deeply entrenched in the mindset that if you don't get it perfect during the shoot, it's all over---you lost it. And I just don't see that as being true anymore (rather, I think it's more a case of "you can't teach an old dog new tricks").

Thoughts? :-)

Scott

[Seth Bloombaum]

There is no "perfect" to be found by spending seemingly endless time in setups and lighting. IMO it's all a compromise.

Now, where to compromise between lighting setups and post-correction, I think that's going to be different on different shoots, that have different needs.

I'm trying to write from the perspective of the producer, here. There is a very needed role, if you call it "producer" or something else, that person who is empowered to say "that's good enough", or "that's not good enough". Going overboard in the shoot... well, one can go overboard in the post color correction too. This is on the assumption that time and money do matter. If they don't matter, more power to you, have a good time, and your teams should have every resource and ability to make great product every time. Do you have any open positions?

My first law of being a professional: Someone who gets paid for their work.

My *new" second law of being a professional: Someone who knows how good is good enough.

Someone who doesn't know when to stop lighting and start shooting needs to pay more attention to the bottom line and/or take more instruction in lighting, so that they can get to their desired look more quickly.

Me, I've been shooting flat for several years now, and I color correct everything. That doesn't take the place of knowing what style and ratio of lighting I'm aiming at, and how to get there.

As an old-timer myself, I think it's wonderful that we have such ready access to incredible post tools. Yes, I used to pay hundreds an hour for "online" editing and color correction. No more. It is a new world. But that's no excuse for shoddy lighting & shooting, either.

[Josh Bass]

Ill take this further and say its not that stuff CANT be fixed in post, its that folks generally feel they shouldnt HAVE to in order to get a result that couldve been achieved in camera had mistakes not been made. Also, that kind of post work, unless it was planned in the first place, may add expenses to the project that werent in the budget so thats no good, and also, what if the client has a really quick tu rn around time with just barely enough time to edit as it is? They literally may not have time to carefully color correct stuff that wasnt acquired correctly. This is not to say people (e.g. me) dont make mistakes, but i would say its definitely the wrong attitude to go into something planning to do it less than perfectly with the idea that it can be fixed later.

[Bill Davis]

You know what actually matters here?

Money.

If you can get it right in production - you don't have to burn up your time or your money to fix it in post production.

But some things cost MORE to fix in the field than they do in post.

So in the field, somebody has to understand when spending more to perfect a particular shot either "is" or "is not" just doing stuff that you can more efficiently fix in post.

Here's a simple example. If you can fix a "mixed light" problem by simply switching OFF the overhead fluorescents - you're nuts not to do that because it's so quick and cheap.

But if the ONLY way you can correct the mixed light problem is to spend 4 hours gelling every square inch of an large offices many windows - then assessing whether you can fix it with a mask and color correct in post is perfectly reasonable.

As a director I'd have to look at the shot. Figure out issues like whether the shot is static and/or whether the talent is crossing the window. That will tell me if I should shoot a separate plate of the windows for post replacement - or need to take another approach.

Video is constantly conditional. Any "rules" are only useful when they're useful.

So figure out the money - and the rest typically falls into place.

FWIW.

[Brian Drysdale]

It also depends if you're going to have time to fix things in post. Often post production time is limited and I know of editors complaining about their ever tightening schedules.

Making good decisions during the shooting stage can minimize the amount of post production work. What you can do during the shoot will depend on your resources, but that can also apply to post. If you've got the time and resources to fix things fine, but there's also the chance it won't get fixed in post.

[Derek Heeps]

I agree that it is always desirable , and I'd venture usually less work overall , to get it right in the first place .

However , the example mentioned above "But if the ONLY way you can correct the mixed light problem is to spend 4 hours gelling every square inch of an large offices many windows - then assessing whether you can fix it with a mask and color correct in post is perfectly reasonable." does demonstrate that sometimes it is not always the best option .

Of course , there are always different ways of doing things , and in the above example , putting blue gels onto your lights to light the office to match the daylight coming in the windows might be the simplest option of all .

Every job is different and needs to be evaluated individually .

[Jon Fairhurst]

This is a matter of risk and cost management.

You might want a high-contrast output. In that case, do it in post, since a small error in exposure while capturing high-contrast in 8 bits will make things unrecoverable. On the other hand, if your end product has lower contrast, you might be able to nail the look in production with little risk of messing up. In that case, you only have to fine tune (or fix problems) in post.

It also depends on skill. On the 5D2, I've found that shooting B&W with glass color filters gives a wonderfully, clean result. But this isn't for everybody. With experience, color filters are very powerful. When used improperly, you can get things very, very wrong.

So, if you're good and fast in lighting, get it close to right in production - but don't paint yourself into a corner with overly high contrast. If your grip crew is small, inexperienced, and under-equipped, shoot fast with latitude and expect to spend more time in post.

[Paul R Johnson]

I'd suggest considering the problem from an audio perspective. You can't get the mic in exactly the right place, so you can tweak the eq (gently or savagely depending on what it sounds like). Maybe there's a hum or two, I can stick in a sharp filter and remove that. There's maybe clip or two, that I can repair. The auto tools can't quite do it, so I draw in the waveform by had to fix it and I can't hear it so it's fixed. Maybe the guitar was a bit out of tune, so I repair it re-recording the part in the studio and replace the guitar. This then means I can hear a few dodgy notes so while I'm at it I'll re-record the bass. The kick drum's a bit thin, so I'll set up a sample that can be triggered from the thin one and that will sort that.

By this point I've got a great sounding product. Trouble is, quite a few musicians who played on it are now replaced by me, and for the purist, some of the data isn't even music any more, it's just non-distorting data that hides where nasty clips were.


The product is much better to virtually everyone who listens to it, but the question remains - is it a truthful version of the original? No. But it's better than the original, isn't it?

[Don Palomaki]

The golden rule applies - the man with the gold makes the rules - so you deliver what the paying customer wants when he wants it and achieve it the best way you know how at the price he is willing to pay. If he is satisfied with what you did you get more work from him (or his referals).

Some things cannot be fixed in post, and in any case to the extent you are post processing you are moving the inmge up or down the exposure curves introducing (hopefully invisible) artifacts to the image. (Your digital image is limited to a fixed number of bits range.) So other things equal, getting it right in the shoot is best.

As noted by others, it is the art and science of striking the sweet spot balance between cost, schedule, and performance that separates the professionals from Joe and Jane Six-pack.

[Garrett Low]

This is a perfect example of why planning through your entire production is important. There is no "one right way." It will depend on what you're end goals are and what your resources are. If you are going to have to turn around with a final production in a very short amount of time, you obviously would want to get the look you are going for while you shoot. This would apply to a lot of interview situations or if you are doing SDE's.

However, I work with a lot of productions that go through heavy color grading in post. These are usually narrative films and music videos. For these, the conventional thought is to light and shoot for maximum latitude for post manipulation. As translated from a colorist I work a lot with, "shoot it flat and keep things from being blown out or crushed." Skin tones are usually slightly underexposed instead of over, and lighting is set up so that the desired shadows are captured.

The key, as has already been stated, is to make sure you get what you need during the shoot. And, you will always have to compromise. You could spend hours trying to get the perfect lighting setup. However, you have to remember that unlike still photography, motion pictures are just that, capture of motion. Your subjects will be perfectly lit at one mark but not at others. The trick is to get is so that they are correctly lit where they need to be and be able to live with the marks that they are not.

[Michael Wisniewski]

Somewhere along the line, these 3 phrases got confused with each other:

• Fix It In Post - for amateurs (even if they're working on a professional set)
• Finish It In Post - for people who understand how to shoot & edit properly
• Fix It In Pre-Production for people with the humility, time & money to say "we don't know yet"

[Scott Wilkinson]

Great responses---thanks all!

One point to consider: I don't work in Hollywood circles or any corner of the industry where I'm able to use top-flight colorists, etc. But in my decades of experience, one thing I know for certain: the vast majority of all video production people who use a given piece of software, typically use no more than maybe 50% (at the high end) of that software's potential (I refer mainly to the big creative production apps like Photoshop, Premiere Pro, and top color grading apps.

It's not that people aren't skilled...but rather that I believe we're at the point where one could argue the depth and breadth of a program like After Effects is as vast as the depth and breadth of using a camera. You have a team of dozens of people working for years to create a single application that is so packed with tools and features and possible workflows that it's VERY rare that one person learns them all. Typically, we learn a very small subset of those tools and features that does what we need it to, and disregard the rest.

So part of the spirit of my initial post was to suggest that while there are plenty of people out there who have used a camera for decades and know it like their own hand...I believe there are far fewer people (even today) who have a similarly deep knowledge of one of these powerhouse post-production apps. (Again, I'm talking about across the range of the industry---I realize that there are probably people who work for Pixar who know some of these apps inside-out!)

For example, does anyone here know anyone who can recite---from memory---every single menu item in After Effects and explain exactly what it does? I realize some might say "Why would you want to do that? What's the point?" You wouldn't...my point simply being that there ARE people who could recite every control on multiple cameras and explain exactly what they do...but we don't have that level of expertise with the software (not yet, anyway).

So I simply wonder if we might reach a point in the future where it seems just as natural for someone to "live" with these apps day in and day out...and reach a level of knowledge and capability (in post-production) that we can barely imagine now?

Scott

[Warren Kawamoto]

An ad agency says they'll do a shoot for an estimated $10K. They'll hire a good crew with a Red, then spends 8 hours setting up to do 20 takes. The more time they spend on the shoot, the more they can charge the client.

Then there's a freelancer who gets a $1K budget upfront to do the same shoot. He'll set up his own lights, spends 1 hour setting up, does 3 takes, then tweaks things in post. The longer he shoots, the more money he loses.

In the end, when both play their shoots on the web, the shots look identical. See the difference?

[Claude Mangold]

IMO, it yes, and yes.
Do the best work you can out on the take, and do the best in post.
When I shoot a docu or fiction thing, part of the art process is to figure out what characteristics I want in my images. Then I sit down with the DP and we discuss my vision in practical terms, what's realistic in the camera&lens&filters&lighting, what's realistic in post. .
Agreed with Garrett , most of my work goes through extensive grading as well, not to do work not done in the field, but to match, enhance moods and narration. Check out Sokurov's films for expressionist grading.
Also, an eye for detail will be useful on the shoot. There's some stuff that's very difficult to get corrected in post, like green hues on pale faces that you've not re-lit even though you're shooting under trees, etc.
As to your last point, you don't need to be in Hollywood to find professionals who know their tools inside out. Don't bother with those who don't.