Home · Register · Software · Software · Join Upload & Sell


  Previous versions of cgardner's message #10095590 « Do I really need a Light Meter? »


Upload & Sell: Off
Re: Do I really need a Light Meter?

First you need to understand the goal of the exposure exercise and what defines success. The broader goal beyond the technical stuff is to make a photo look real and create the same emotional response in the mind of the view as seeing the same content in person. In other works making it seem real. With respect to exposure making a photo seem as real as possible requires recording detail everywhere the eye sees it in person. That makes the exposure both optimal technically and perceptually \"normal\".

That goal can be met quite simply with studio lighting with just the camera and two towels; one white, one black.

1) Start in a dark room, set camera to the desired f/stop. Place a light stand at 8ft from camera and drape the towels over it.

2) Place one light (the fill source) just over or under the lens of the camera. Raise its power as you make test exposures until you see the fuzzy texture of the black towel visible in the playback, pegging fill intensity to the darkest shadow detail you will encounter on your portrait subject standing in the same place a few minutes later.

3) Place the second light off axis. For a full face view a good starting baseline is to either place it directly above the camera so it hits the eye line of the subject at a 45 downward angle (butterfly) or is placed so it hits the nose at a 45 angle from the side and a 45 downward angle. For oblique and profile views maintain the same 45/45 angle relative to the nose (with key light winding up 90 and 135 from the camera axis respectively).

4) Turn on the key light and raise its power, while making test exposures, until the white towel target on the stand begins to visibly clip in the playback warning show by the image blacking out and sometimes blinking in areas of blown out detail. Adjust power until the warning is just barely visible in the brightest parts of the towel.

What you will have done is meet the technical and perceptual goals of exposure with just your camera with very little effort. With a bit of practice and experience knowing the power of your lighting gear it will take you about 30 seconds, a minute at most, far less time than metering each light with a hand incident meter.

The lighting ratio numerically, is not important in meeting the goal of matching scene to sensor range. In fact the ratio numerically needed to do that varies from camera-to-camera with digital because the range of the sensor varies due to physical design of the light capturing sites. The sites are like buckets and those with the largest overall volume have the longest ranges. Full frame cameras with larger sensor sites will usually have longer ranges than those of cropped sensors. Within a die size lower MP sensors generally have longer MP ranges that higher MP sensors. My 1.6 crop 8MP 20D has sensor sites the same size and range as a FF 5D. My 15MP 1.6 crop 50D has smaller sensor sites than the 20D but similar overall volume due to clever design changes and a range that is slightly less than my 20D. Optimal exposure in the technical sense is fitting scene detail to whatever range your sensor has and the best way to determine that is photographic a dark and light object matching the extreme of any scene you will encounter.

Once you obtain optimal \"scene-matches-sensor\" exposure by adjusting fill and key power visually based on shadow and highlight detail your photo will look similar to this:

It looks \"normal\", exactly how the guy looks in person in average lighting. Not thoughtful and moody or over the top light and carefree, just normal. That\'s what optimal exposure does, create \"normal\" seen by eye rendering because that is how it is engineered to work.

That is of course not how you will want to light all your portraits. For the wife and kids you\'ll want lighter shadows and for your wise old 80 year-old uncle you\'ll want darker shadows. The highlights in both will wind up looking the same in both. What changes the reaction of the viewer is a reaction to the tone of the shadows, controlled with fill.

That creates a bit of a dilemma exposure-wise. Above the shot has perfect exposure with detail in the darkest parts of his clothing if fill is increased any more to lighten the shadows on the face it will make the shadows on the clothing lighter than \"normal\" giving the photo a washed out look. Conversely if fill is reduced to make the shadows on the face darker for your wise old uncle Fred you will no longer see the detail in is black suit.

The solution to that dilemma? Learning to \"feather\" the key and fill light sources and control their footprints with modifier size, placement and direction of fall-off.

This shot taken with a butterfly pattern with both lights back near the camera record the full range of the scene (i.e. perfect exposure technically) and produces a perceptually \"normal\" result.

This oblique shot with the key light placed 45/45 from his nose on the left also has a full range of tone, but changing the background and making the facial shadows a bit darker changes the mood a bit.

For this profile shot I turned him and used the small 16\" x 22\" SB with a circle mask and grid used as hair light in the other shots as my key light because I wanted attention focused on the face

This is what the out of camera RAW file looks like for that shot:

The profile shot has the same overall tonal range as the others, but I changed the direction of the fill source, moving it directly in front of him instead of over the camera so it would fall off front>back relative to his face and clothing not front > back relative to the camera as in the other views.

I didn\'t use a meter for any of those shots, I just did the basic set-up of the lights visually using a target on a stand for each set-up while he amused himself watching TV in the next room. When he came back under the lights I didn\'t need to fiddle with them while he waited.

What was the ratio for those shots? Don\'t know, don\'t really care. I don\'t build my lighting strategies from numerical blueprints I build them based on goals, primarily the reaction I want to evoke in the mind of the viewer. In shooting with progressively darker shadows I wanted to create a progressively more serious vibe conveying he fact it was his confirmation and passage out of boyhood to a become a mature young adult.

I could if needed tell you the lighting ratio very easily using the white towel and the camera clipping warning. I would turn on just the fill and adjust my camera aperture until the fill was clipping the white towel. Then I\'d turn off the fill turn on the key light and again adjust aperture until the white towel was below clipping. The difference between the two would tell me exactly what the ratio between incident strength was in f/stop. I could measure the hair light and the background the same way by simply putting a white towel where they hit. I buy them by the bag of 50 at Costco (shop towels).

When lighting a group before the group arrives I\'ll put a light stand or chair on the four corners of where the group will standing and one in the middle. I then adjust my lighting until I can get all the towels clipping in the playback at the same time, telling me the lighting pattern is even. There is a trick to doing that with one or two centered lights; arrange the group in an arc or chevron pattern in front of the light(s).

Once the lights are even all I need to do is close the aperture 1/3 stop and I have even lighting and perfect exposure. SInce I know from experience how to arrange a group for even lighting the process doesn\'t take very long and the results are far better than the film days trying to do the same thing with only meter readings.

I use a meter: the clipping warning of the camera which makes every pixel in the camera playback a spot metering zone. That\'s million zone metering! Hard to beat that.

Digital camera added three new forms of \"metering\": instant playback, the histogram depicting the sensor range, and the highlight clipping warning.

Just by looking at the playback it is possible to tell if the exposure is in the \"ballpark\" of correct or not. By including a white textured and black textured object a photograph as a test object the playback, in combination with your brain and eyes, can tell you whether exposure is correct more accurately, faster than a meter.

The full width of the histogram depicts the range of the sensor. It represents the scene as a bar graph of 256 tonal values from no light (left) to the point of max. saturation of the sensor \"clipping\" on the right. It will show whether or not the scene fits the sensor, but only after you first \"peg\" the exposure to the highlights, which is most easily done with the clipping warning.

The clipping warning, when enabled in the playback shows when and where exposure is maxing out the sensor and obliterating detail.

With those three methods of metering available do you really need to carry another meter on a lanyard around your neck? Unless you have an unlimited budget the money would be better spent elsewhere.

Nov 18, 2011 at 01:28 PM

  Previous versions of cgardner's message #10095590 « Do I really need a Light Meter? »