Exposing for the shadows - when you want dark shadows

[mike_chalmers]

<p>I have finally got a medium format camera again after selling my Hasselblad to make ends meet - and while I've become much more experienced in photography, it's mostly digital and any film I've shot has relied on in-camera metering (with mostly good but often underexposed results).<br /> I'd like to learn how to meter for the results I want. As I'm likely to be mostly shooting portraits and people, I'm concerned with getting the skin perfect. <br /> So I've read and understand the 'expose for the shadows' rule but what if it's direct sunlight and I want the face to be exposed with rich tones and allow the shadows to fall to black? <br /> Or, if I want consistency in skin exposure and always aim to keep it with as much detail and deepest/richest colours possible, should I be metering directly in front of the face?<br /> I'm going to do some tests tomorrow but would appreciate any advice as my film supply is limited.<br /> I'm gonna shoot Portra and set iso to 1/2 box speed btw. Although maybe I will try some at box speed as well, to see if I prefer the results.</p>

<p>Here's a shoot that shows good skin, rich colours and dark shadows - not a perfect example of my goals but should illustrate my point: http://wearesodroee.com/2015/08/06/no-debutante/</p>

<p>Thanks in advance for any input, I really appreciate the people of these forums.</p>

<p>Mike</p>

[SCL]

<p>The highlights are blown in the shot you referenced...just applying increased contrast would do the same job on most well exposed photos. To your question of perfect skin...the answer is different depending on if you are shooting color, negative or positive films, or black & white, as well as actual skin color.</p>

<p>Assuming you are metering with a reflective meter:<br>

The best easy answer I can offer is to meter a gray card and in the same light meter the palm of your hand and notice the different readings. If you have light skin, the difference represents the increased exposure adjustment you would make for shooting somebody with the same skin tone you have. If you are shooting a dark skin tone, you would decrease your exposure. This general guideline applies for negative films shot near the mfr's suggested ISO.</p>

<p>If shooting positive films (basically mostly slides these days) you would typically meter off the highlights. I've followed these general guidelines for many years with very good success requiring little post processing.</p>

<p>If all this is too complex to keep in mind, switch to incident metering, and basically it is all taken care of for you.</p>

<p>One last thought. though. Unless you do your own processing and can control the variables involved, it will behoove you to run a test roll bracketing shots in approx. 1/3 stop increments from 2x the mfr recommended ISO to 0.5 recommended ISO so you can determine the best ISO to measure exposure at for each film you plan to use.</p>

[wblynch]

My old rule of thumb was to meter the face and if they were Caucasian I opened up one stop from metered and if they

were dark skinned I closed down one stop. For suntanned or browner toned skin I left it as metered. It usually worked well

for me.

[mike_chalmers]

<p>Hi Stephen, </p>

<p>Thank you for responding. As mentioned, I'm going to be shooting Portra - i.e. colour negative. <br>

And no, nothing to complex to keep in mind. I am very familiar with metering and the things you are suggesting and, as mentioned, I do plan to bracket and run different tests. <br>

What I'm really looking for are photographers experiences shooting at box speed, and perhaps over and under - as well as metering different areas of shadow or light on skin or in the shadows/grey card to achieve the richest, deepest colours on the skin, with the most detail.<br>

To me, the link I sent does not have blown highlights as I can still see detail in practically all highlights. Maybe you mean the very small areas on the clothing but overall I think they're underexposed compared to what a normal "correct" exposure would look like. It also looks like you might just be referring to the first image on the page, so here's another one that might serve as a better example: http://wearesodroee.files.wordpress.com/2015/08/anna-ewers-by-alasdair-mclellan-for-vogue-uk-september-2015-9.jpg?w=1480&h=1968</p>

[mike_chalmers]

<p>That's interesting Bill! Thanks for your input - I might try that during my tests.</p>

<p>I'm basically just looking to plan out the shots I'm going to take during the tests and document it all and then get the processed film back and work out what works. Doing 2x studio rolls, 2x daylight (before and during golden hour), probably split that between direct, side and back-lighting too. <br>

I've only got 10 shots per roll so want to plan properly. I've got an idea what will work best, but would love to hear others experiences / examples in case I'm wrong! </p>

<p>Cheers</p>

[mike_chalmers]

<p>Oh Stephen I just realised you assumed I was using the camera/reflective meter, but I'm talking about incident (handheld digital) meter.</p>

[SCL]

<p>Hi Mike - since incident metering measures light falling on the subject, rather than light reflected off it, I've generally found that no correction is necessary, again assuming you have determined the proper ISO for setting the metering. I did find, even with incident metering, that occasionally with VERY dark skins, I sometimes got better results by reducing my exposure by about 1/3-1/2 stop.</p>

[James G. Dainis]

If you are most interested in portraits and getting the skin tones correct then you don't meter for the shadows. You meter the skin, open up one stop if Caucasian, and let the shadows fall where they may. If you were shooting black and white film then you could expose for the shadows and develop for the highlights or skin tones.

[Uhooru]

<p>I think if you want to meter for the skin tones, you would use a domed incident meter, metering one side of the face, and then the other to get your lighting ratio. Remember when you are metering, point the meter back towards the camera, not facing up towards the sky. You can get pretty accurate readings that way.</p>

[glen_h]

<p>C41 films have a gamma about half the usual for black and white films. Subtract a little for the orange mask, and you still have a big dynamic range.</p>

<p>On the other hand, that makes printing harder. RA4 papers and C41 scanners have a compensating gamma. You mostly don't need to worry about the film exposure, but twice as much when printing or scanning.</p>

<p>As above, light skin has about twice the usual 18% gray card reflectance, so if you meter off the face, you want one stop more. I don't know about other skin colors.</p>

[mike_chalmers]

<p>Thanks for your responses guys - I'm not sure I was very clear with my questions, but appreciate everyone's input and I'm going to take some pictures to test with different lighting conditions and settings today and see what results are best for the film type with skin.</p>

<p>@Lee Sarile</p>

<p>As I've not shot with this camera yet, I've not had disappointing results. I know that Portra is the film is want to use from its characteristics and will be doing a grade and retouch in post (something I do in my line of work with fashion photographers, but generally all digital). </p>

<p>@James Dainis</p>

<p>Thanks, this is essentially what my thinking was. I'm basically trying to work out if I should open up one stop (/meter at half box speed) if I want to get the richest, deepest skin tones possible. Because of the latitude range, a lot of people have said to do this, but I was wondering if others had specific experience. I'll test both ways and maybe bracket further (although my my film supply is limited today).</p>

<p>@Barry Fisher</p>

<p>Yeah, thanks - This was my plan for all shots and I'm gonna be noting everything down so I can reflect on it after. </p>

<p>@glen herrmannsfeldt </p>

<p>Yeah as mentioned I'm using incident light metering so it should be okay, but that's an interesting point about the gamma gamma compensation.</p>

[glen_h]

<p>I used to use VPS mostly with my Nikon FM, with the match LED meter.</p>

<p>If there was plenty of light, I would set the exposure just a little more than the 0 of the meter. <br>

No need to change the ISO setting, just adjust the exposure when actually metering.</p>

<p>But other than manual mode in most cameras, you either change the ISO, or use an exposure +/- setting. I would probably do a half stop, but you could do a whole stop.</p>

[mike_chalmers]

<p>@ Les Sarile<br /> No - I wasn't using Portra or 1/2 box speed. Was mostly using Gold or Fuji Superia - cheaper and more available to me at the time. <br /> I'm confident I could shoot these films better now and will maybe revisit. I was tending to replicate my exposure settings that I was using on digital, but I was watching for blown highlights, not aware that this would make for bad film photos. I would have been better relying on in-camera metering all of the time.<br>

I was shooting mostly on Canon 35mm - T70s, a couple of different mid-range EOS too. Scanning with Epsom V500 Perfection.</p>

<p>@ Glen Herrmannsfeldt</p>

<p>Yeah that's interesting and what I was trying to do manually yesterday. My issue was lack of available light as my camera has a reputation for slight blur so you have to double the (and inverse) the focal length to guarantee no shake when handheld and my aperture was already wide open, so would be concerned the shutter would go to a danger area with exposure comp. Luckily the faster iso film supply I ordered has arrived.</p>

[scott_murphy5]

<p>I suggest you start using an incident meter. Or use an 18% gray card and measure directly off of that.</p>

<p>Although I do use an incident meter a fair amount of the time, when I want to meter critically I use my trusty 30+ year old Pentax 1º Spotmeter V and apply Adams' Zone System. Caucasian skin (not heavily tanned) is around Zone VI (one stop above neutral gray), You can meter off of a highlight (not shadow) and set the meter to indicate the measured value 1 stop above the average reading. </p>