Wednesday, November 25, 2009

Meters v. Histograms

This isn't going to be one of those "old photographers do it by the meter," updates. It's not a rant. I'm not going to infer that, if you're not using a light meter, you're not a "real" photographer. I just want to point out a few advantages of metering over relying on histograms.

Let's say you shoot weddings. I don't but, for the sake of argument, let's say I do. I'd probably be using a light meter for, at least, the formal portraits. Why? Well, because wedding photographers are dealing with lots of whites, as in wedding dresses. Could I use a histogram to dial in those whites? Sure, but possibly at the expense of skin tones. When there's a lot of white in an image, a histogram is going to look like you might be over-exposed, with a bunch of peaks on the right.

On the other hand, a meter allows you to precisely know what's going on with those whites while, at the same time, letting you know what's going on with the skin tones. By metering (and lighting) you can strike a great balance between the whites and the skin tones. If you're really clever, you'll be able to light, meter, and expose for both the whites as well as those black tuxedos, showing detail in both.

Let's say you shoot art nudes. I don't but, once again, for the sake of argument, let's say I do. Again, I'd be using a light meter. Why? Well, because in art nude photography, shooters are often dealing with shadows. Lots of shadows. Shadows often approaching black. Once again, by metering (and lighting) you can strike a more precise balance between the models' skin tones and the shadows. If you're relying on a histogram to give you this information, the histogram is going to look like you're under-exposed, with a bunch of peaks on the left.

Histograms are great. They're helpful and can lead you to proper exposure. But there are enough situations where the histogram is going to be misleading. Sure, possibly as much as 80% of the time you can get a good exposure using a histogram alone, especially if you're experienced reading them. But for that other 20%, a meter is what's going to dial you into proper exposure. I know some of you are thinking, "No problem. I shoot RAW. I can fix my exposure fuck-ups when I convert."

True enough. Leastwise, much of the time... but not always.

You blow those whites completely out and no amount of RAW converting is going to recover detail that simply isn't captured. Conversely, same holds true with shadows, albeit to a lesser extent, i.e., there's often detail in shadows even when they look very black. But blown-out highlights? Fuhgetaboutit!

BTW, while histograms will get you through most any exposure environments in a pinch, relying on the LCD screen alone is, well, is too iffy even for government work.

Ever watch those BTS vids of notable pros shooting? If the vid's content has much depth, you might have noticed those peeps are most always using a meter. Leastwise, an assistant is wielding one. If you aspire to be the next Annie L. or Greg Gorman or David LaChappelle or any number of top-notch shooters, you might want to consider getting and using a light meter, assuming you don't already own and use one.

The pretty girl at the top, trying either to push her way out of the picture or keep me at bay, you decide, is Dylan from a few months ago.

14 comments:

Anonymous said...

I use a light meter religiously, but if you don't calibrate your meter and profile your camera, it won't give you meaningful results.

Often the chip in your camera set at 100 ISO, doesn't deliver more than 50 ISO, so you get underexposed files if you haven't profiled your camera.

It is really easy to do this using the DNG profile program from Adobe, for FREE.

jimmyd said...

@Anon,

Good info! Thanks! Outa curiosity, since my 5D has an ISO 50 setting, what would 50 be if 100 is 50? Now I'm totally confused.

Lou said...

Very cool post, Jimmy. I was extremely fortunate to pick up what was marked as a $300 Polaris meter at Ritz when they were closing down stores for $40, although me being a relative youngin' in the practical photography biz, I don't find myself reaching for it enough, if at all--despite the fact that I know how epically useful they are.

Most of the time I don't have time to whip out the meter, as recently I haven't been doing any studio-ish work. That being said, is there any merit to using it in reflective mode at, say, a football game at mid-day?

And also, do you have any cool tips/pointers for those of us who are new to using light meters? I know the general theory and basic usage, but I know there's got to be more than "stick this in front of model's face, light, click button on meter, read display" lol...

jimmyd said...

@Lou,

...do you have any cool tips/pointers for those of us who are new to using light meters? I know the general theory and basic usage, but I know there's got to be more than "stick this in front of model's face, light, click button on meter, read display" lol...

When you figure that out let me know cuz, in a nutshell, that's the procedure! So simple, I gotta ask, "Why aren't you using it? :-)

Anonymous said...

Jimmy,

Each person's particular camera and lens combination, can be off, 1/3, 1/2, to 1 full stop off the dial setting.

You need to run a calibration test on both the meter and the camera profile. The meter you only run on 1 ISO, which is 100. In bright sun, at noon, on a clear day

(ALL of the above is important/mandatory)

you should get 1/500 @ f:8.0. Adjust the meter until you get that result. That way you will match everyone else's.


This assumes everyone else has done this too. If not have them set their meter to match yours.

On the camera profile, I would run it on 50(thought I never shoot there) 100, 400 and 800 under open shade, direct sun and electronic flash.

It really doesn't take a lot of time to do this any more. You only have to do this once. After this, you periodically check it at one ISO to confirm it hasn't drifted.

There is some specific directions on reading with a light meter, but that is a very long post, so better read it in a book or send me a large check. :)

Unknown said...

There is a nice basic but comprehensive article on this here:

http://www.lighting-essentials.com/using-a-lightmeter-and-placing-the-photographic-exposure/

Unknown said...

Jimmy.

I'm a huge proponent of getting it right in the camera, and having to do as little post processing as possible.

I DO have a lightmeter, but have yet to figure out how to use it.
So I have to say Thanks to Robert for the link. I'm going to read it this weekend.

Thanks for the post.

Lou said...

@Jimmy

When you figure that out let me know cuz, in a nutshell, that's the procedure! So simple, I gotta ask, "Why aren't you using it? :-)

Haha... Well, damn, when you put it that way... :P

@Robert
Thanks for the link, looks like a pretty good article. I'll probably read that while enjoying a tryptophan coma later today. :D

James said...

Well, you're almost right about wedding photography ;-) When I first started I used a meter for the formals all the time. After a while you begin to just get certain things down. Like I just know that if you put a 100ws light on full, with just a standard reflector in the the third pew, high and to one side it'll be F5.6 @iso200, same for a 300ws light at 3/4 power into a large bounce umbrella. In most situations that'll hold the highlight details and you check the histogram just to make sure. I rarely have to tweak it more than F5.0-6.3. As long as you hold the highlight details you can dig out shadow (tux) details easily in Lightroom. That's more a problem of latitude than exposure.

The big thing that helped me ditch the meter was RGB histograms. When shooting people it's easy to just blow the red channel a little bit and have it not show up in the luminosity hist. At least in Canon cameras anyways. This will throw the skin tones off. When I up graded to the 40d which has histograms by color channel I was much more confident not using a light meter.

jimmyd said...

@James,

If truth be known, I often use both a meter AND the camera's histogram. For me, it's usually not an either/or situation, rather, it's this AND that. Thanks for the comment!

Ashley Karyl said...

I agree 110% Jimmy about using handheld lightmeters and shall continue to do so, even if some believe I may be old fashioned by doing so.

Although the inbuilt meters on most DSLR cameras are very good these days they are not perfect and relying on the histogram is very limiting because it is not nearly as accurate as most people believe. Depending on your camera, you can generally push the exposure quite a bit more than the histogram would indicate towards overexposure without burning the highlights when shooting Raw and coincidentally also benefit from better shadow transitions.

It's not all about Photoshop. Do it right in camera and you will find your post processing work far easier. A poorly exposed image shot at 100 ISO can still look pretty horrible and a well exposed image at 800 ISO can look surprisingly good.

Rovingrooster said...

I just bought my first light meter ever based on your article. I just couldn't see the value of a meter when there's a nice LCD display on the back of the camera. It's not film; digital = instant gratification.

In your comment to Lou, you said "... in a nutshell, that's the procedure! So simple, I gotta ask, "Why aren't you using it? :-)" That just made something suddenly click inside of me after years of hearing the importance of a meter. A lot of photography books and sites write pages of fluff designed to confuse and sell. You have a unique knack of keeping things in perspective. I like that you don't sell on your site; that it's not an "Either, Or", and there are other solutions to the same problem.

So a week with my shiny Sekonic L-358 meter, it takes 2-5 minutes to nail my exposure after setting up my flashes. Versus about a half hour of test shots and chimping when I was being cheap with no meter. More importantly, the numbers sure help in fixing my light setup when something just isn't working out. It's not always apparent from chimping what exactly is messing up the shot!

Here's a neat article that helped me calibrate my meter with Photoshop. There also info on light ratios:

http://super.nova.org/DPR/Ratios/

I think you'd make a great salesman. But, I'm glad you're not.

Thanks Jimmy,

John

jimmyd said...

@RovingRooster,

Congratz on taking the "meter plunge" and joining the photographic world of--as you indicated--making this stuff easier, more efficient, and more precise rather than time consuming and guess-work intense.

Thanks for link!

Lou said...

@rovingrooster I'm glad you've taken that step to buying and using a meter--I know I'm stoked about using mine now, as while I haven't done anything serious with it up until Monday night (which got cut short because I'm a dumbass and didn't sandbag a stand with an umbrella outside, which came crashing down and broke the hotshoe mount off of my new-to-me old Vivitar 285 before I could really even use it, damnit), but I had lighting up for the test shoots within two minutes. Although I suppose if I didn't use the meter, and was closer to the lights at that time instead of a 70-210mm lens away with a drunk buddy being my only savior near the light, I wouldn't have lost the damn flash lol.

I agree with you though, I'm glad that Jimmy's not a sales guy. It's nice to be sold on something by a non-sales guy.