Forums > Photography Talk > Exposure Question

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Good evening Mayhem,

I am a brand new photographer, and I have exposure screw-ups all the time, but today's really stumped me.  I had two models, one in white, and one in black, against a black background.  I did test shots, checked my histogram, the whole nine yards, and ended up with the model in black nicely exposed, and the model in white blown out like crazy.  Of course, I didn't realize this until I got home.  Gah. 

I mucked about with saturation and color a bit, and posted one that was as salvaged as I could manage to make it, if you want to see what I mean.  May I ask how you figure out how to split the difference in an instance like that, so that each model ends up at least sorta properly exposed?

Jul 25 05 01:27 am Link

Photographer

Hugh Jorgen

Posts: 2850

Ashland, Oregon, US

White is considerd the hardest to shoot against...
If you get it down you got it made!!
Practice alot!!

Your work is very good!!
Just dont give up
Saw no white backgrounds

Jul 25 05 01:31 am Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

For negative film, expose for the shadows, develop for the highlights.

For digital, expose for the highlights, develop for the shadows.

When you expose, put the whites as close to the right edge of the histogram as possible without blowing them.  Your camera may have a setting that shows blown highlights as flashing.  Then, when you're editing, you can lighten the darker areas--just watch out for noise.

Also, more practice in lighting will help you avoid this problem.

Jul 25 05 02:43 am Link

Photographer

KarlShone

Posts: 11

Maryport, England, United Kingdom

Hi well, I have doing white back grounds professionaly for 18 years and photographed, Zebras to Models on 10 x 8 to Digital, The rules seem the same regardless, aim to shoot no less than f8 ( in relation to 35mm) and get the background 1 to 1.5 stop lighter than model (flash I mean) with no hot spots. Black flats either side of model help a lot.
Histograms, white balance and all that, well I just shoot as with trditional film, but then I'm still learning and they come out ok, have a look, hey don't make fun but I have a sony cybershot f828 !!!!

Jul 25 05 03:05 am Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

Note that the background was black.  The clothes were both black and white.

Jul 25 05 03:14 am Link

Photographer

Belair

Posts: 359

Downers Grove, Illinois, US

It sounds to me like you didn't use a meter to determine exposure for these shots.

It seems that a lot of photogs are making the mistake of relying on the histogram and/or the LCD panel to determine exposre - and then wondering why they didn't get the results they wanted.

It's kinda funny how photogs will go out and spend $2,000+ on a digicam, then cede exposure decisions to a $1 LCD panel.

In reality, the histogram tells you very little about exposure.  It will provide you with a very generalized information such as whether your shadows or highlights are blocked up.  However, the histogram tells you nothing about how critical areas of the frame are exposed.

Typically, the critical area of a model is his/her face.  Therefore, my recommendation to you would be to spot meter highlight areas of the face and place them in Zone VI or so (if the subject is fair skinned) or you could take a series of incident readings around the face.  That should be your base exposure.  From there, adjust the placement and intensity of the lights lighting the clothing and background until you can be assured that you are holding the amount of shadow and highlight detail that you require.

Jul 25 05 05:46 am Link

Photographer

Columbus Photo

Posts: 2318

Columbus, Georgia, US

Belair nailed it but let me elaborate.  If your incident meter is calibrated to give perfect exposures with your camera, then both white and black will photograph correctly.  That's how wedding photographers get both the bride and groom exposed correctly in the same image.  And with an incident meter, you're measuring the light source itself, not the light reflected from the subject.

Since you're shooting digital, this article may help:

http://www.paulsportraits.com/a2z/a2z.htm

Paul

Jul 25 05 07:01 am Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Hugh  Jorgen: 
White is considerd the hardest to shoot against...
If you get it down you got it made!!
Practice alot!!

Your work is very good!!
Just dont give up
Saw no white backgrounds

Thank you for taking a look, Hugh!  (There wasn't a white background, the problem was a model wearing white next to a model wearing black.)  The encouragement is much appreciated!  smile

Jul 25 05 09:14 am Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Brian Diaz: 
For negative film, expose for the shadows, develop for the highlights.

For digital, expose for the highlights, develop for the shadows.

When you expose, put the whites as close to the right edge of the histogram as possible without blowing them.  Your camera may have a setting that shows blown highlights as flashing.  Then, when you're editing, you can lighten the darker areas--just watch out for noise.

Also, more practice in lighting will help you avoid this problem.

See, that's what's weird.  I actually did pay close attention to the histogram, and the whites weren't even crowding the blow-out point.  And yet there they are, all blown out and stuff.

Head scratch.

Jul 25 05 09:15 am Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by KarlShone: 
...but then I'm still learning and they come out ok...

Wow, more than okay.  Beautiful work, Karl!

Jul 25 05 09:16 am Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Belair: 
It sounds to me like you didn't use a meter to determine exposure for these shots.

It seems that a lot of photogs are making the mistake of relying on the histogram and/or the LCD panel to determine exposre - and then wondering why they didn't get the results they wanted.

It's kinda funny how photogs will go out and spend $2,000+ on a digicam, then cede exposure decisions to a $1 LCD panel.

In reality, the histogram tells you very little about exposure.  It will provide you with a very generalized information such as whether your shadows or highlights are blocked up.  However, the histogram tells you nothing about how critical areas of the frame are exposed.

Typically, the critical area of a model is his/her face.  Therefore, my recommendation to you would be to spot meter highlight areas of the face and place them in Zone VI or so (if the subject is fair skinned) or you could take a series of incident readings around the face.  That should be your base exposure.  From there, adjust the placement and intensity of the lights lighting the clothing and background until you can be assured that you are holding the amount of shadow and highlight detail that you require.

Yowch.  I've only been shooting for a month, I haven't had time to save for all the gadgets I need.  This was a workshop setting, so I'd hoped to borrow someone's, but I think everyone there hoped the same thing, so there were no meters to be had!

I hadn't realized the histogram wouldn't tell me about the critical areas.  Since I had a pale model in white, I tried to veer toward the slightly underexposed side of things, which my histogram said I did.  Which seems to prove your point.

Now, do I have to get a fancy digitalized do-everything-but-make-your-coffee kinda meter, or will a cheaper one do?  (I can't believe how expensive they are.  I should've taken up stamp collecting as a hobby instead.)

I appreciate your detailed response!

Jul 25 05 09:20 am Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Paul Ferrara: 
Belair nailed it but let me elaborate.  If your incident meter is calibrated to give perfect exposures with your camera, then both white and black will photograph correctly.  That's how wedding photographers get both the bride and groom exposed correctly in the same image.  And with an incident meter, you're measuring the light source itself, not the light reflected from the subject.

Since you're shooting digital, this article may help:

http://www.paulsportraits.com/a2z/a2z.htm

Paul

Thank you, Paul!  I'll give it a read today.  I appreciate the resource.

Jul 25 05 09:20 am Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

Posted by SitaMaeEdwards: 
See, that's what's weird.  I actually did pay close attention to the histogram, and the whites weren't even crowding the blow-out point.  And yet there they are, all blown out and stuff.

Head scratch.

Because the whites are such a small percentage of the the entire scene, it will be difficult to see exactly where the lightest point it.    Next time, try filling the frame with the white clothing to check the exposure.

Just curious--what camera are you using?

Jul 25 05 03:27 pm Link

Photographer

J O H N A L L A N

Posts: 12221

Los Angeles, California, US

Posted by SitaMaeEdwards: 

Posted by Belair: 
It sounds to me like you didn't use a meter to determine exposure for these shots.

It seems that a lot of photogs are making the mistake of relying on the histogram and/or the LCD panel to determine exposre - and then wondering why they didn't get the results they wanted.

It's kinda funny how photogs will go out and spend $2,000+ on a digicam, then cede exposure decisions to a $1 LCD panel.

In reality, the histogram tells you very little about exposure.  It will provide you with a very generalized information such as whether your shadows or highlights are blocked up.  However, the histogram tells you nothing about how critical areas of the frame are exposed.

Typically, the critical area of a model is his/her face.  Therefore, my recommendation to you would be to spot meter highlight areas of the face and place them in Zone VI or so (if the subject is fair skinned) or you could take a series of incident readings around the face.  That should be your base exposure.  From there, adjust the placement and intensity of the lights lighting the clothing and background until you can be assured that you are holding the amount of shadow and highlight detail that you require.

Yowch.  I've only been shooting for a month, I haven't had time to save for all the gadgets I need.  This was a workshop setting, so I'd hoped to borrow someone's, but I think everyone there hoped the same thing, so there were no meters to be had!

I hadn't realized the histogram wouldn't tell me about the critical areas.  Since I had a pale model in white, I tried to veer toward the slightly underexposed side of things, which my histogram said I did.  Which seems to prove your point.

Now, do I have to get a fancy digitalized do-everything-but-make-your-coffee kinda meter, or will a cheaper one do?  (I can't believe how expensive they are.  I should've taken up stamp collecting as a hobby instead.)

I appreciate your detailed response!

Really though... An incident meter is not a luxory, it is a necessity if you're going to be working beyond the rank amateur level at all and want anything approaching predictable and consistent results. They are not that expensive - couple of hundred - bite the bullet and make the purchase.

John

Jul 25 05 03:41 pm Link

Photographer

Herb Way

Posts: 1506

Black Mountain, North Carolina, US

The situation you describe is the same as shooting a wedding with the bride in a white gown and the groom in a black tuxedo.  Famed wedding photographer, Gary Fong, suggests going into the camera's parameters and lowering the contrast and saturation.  I've done this with my Canon 10D and 20D and have realized improved results all around.  I rarely use my flash meter anymore. 

Jul 25 05 03:42 pm Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

I mean this purely for the sake of discussion, and not to be contradictory, but what can one do with a meter that one cannot do with a digital camera and a grey card?

Jul 25 05 04:16 pm Link

Photographer

Columbus Photo

Posts: 2318

Columbus, Georgia, US

Thoretically nothing but I think an incident meter is easier to use than getting the model to hold a gray card.

Paul

Jul 25 05 04:34 pm Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Brian Diaz: 
Because the whites are such a small percentage of the the entire scene, it will be difficult to see exactly where the lightest point it.    Next time, try filling the frame with the white clothing to check the exposure.

Just curious--what camera are you using?

Oh, of course, that makes perfect sense.  Thank you Brian!

I'm using a Nikon D70.

Jul 25 05 06:21 pm Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by John Allan: 
Really though... An incident meter is not a luxory, it is a necessity if you're going to be working beyond the rank amateur level at all and want anything approaching predictable and consistent results. They are not that expensive - couple of hundred - bite the bullet and make the purchase.

John

Though I'm not sure I'm aiming higher than rank amateur, the lack of predictability *is* beginning to chafe.  I'll take it under advisement with my next paycheck.  wink

Jul 25 05 06:23 pm Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Herb Way: 
The situation you describe is the same as shooting a wedding with the bride in a white gown and the groom in a black tuxedo.  Famed wedding photographer, Gary Fong, suggests going into the camera's parameters and lowering the contrast and saturation.  I've done this with my Canon 10D and 20D and have realized improved results all around.  I rarely use my flash meter anymore.   

Herb, how did you figure out how much you wanted to lower your contrast and saturation?

Jul 25 05 06:25 pm Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Brian Diaz: 
I mean this purely for the sake of discussion, and not to be contradictory, but what can one do with a meter that one cannot do with a digital camera and a grey card?

May I ask which method you use, Brian?

Jul 25 05 06:40 pm Link

Photographer

S

Posts: 21678

Pittsburgh, Pennsylvania, US

Posted by Paul Ferrara: 
Thoretically nothing but I think an incident meter is easier to use than getting the model to hold a gray card.

Paul

Sounds like perhaps I should try both methods a few times and see which works better for me, maybe?

Jul 25 05 06:40 pm Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

Posted by SitaMaeEdwards: 

Posted by Paul Ferrara: 
Thoretically nothing but I think an incident meter is easier to use than getting the model to hold a gray card.

Paul

Sounds like perhaps I should try both methods a few times and see which works better for me, maybe?

Definitely rent one, and spend a weekend with it around your neck.  Should go for about $15-$20 at Calumet.  Then you'll be able to make an informed decision as to whether it's worth the money.

Personally, I will on occasion use a grey card, but often, I just histogram it.  And even if it's off half or 2/3 of a stop, that's easily fixable in post.  If I'm really concerned, I'll take a test shot or two and pop it into Photoshop and really see the results.  (Makeup artists also like this method...)

Now for film, I wouldn't dare work without a flash meter, even though the exposure latitude is wider.  It's just that with digital, it's immediate and the cost per click is so much lower.

Jul 25 05 08:14 pm Link

Photographer

Tito Trelles-MADE IN NY

Posts: 960

Miami, Florida, US

Posted by SitaMaeEdwards: 
Good evening Mayhem,

I am a brand new photographer, and I have exposure screw-ups all the time, but today's really stumped me.  I had two models, one in white, and one in black, against a black background.  I did test shots, checked my histogram, the whole nine yards, and ended up with the model in black nicely exposed, and the model in white blown out like crazy.  Of course, I didn't realize this until I got home.  Gah. 
My advice? buy an incident light meter and forget the camera's. A sekonik can cost you less than 200 bucks and you wont fail one shot.
I can get into a retoric and boring explanation about Zone System and crap. Buy a light meter, you wont regret. Take care. T
I mucked about with saturation and color a bit, and posted one that was as salvaged as I could manage to make it, if you want to see what I mean.  May I ask how you figure out how to split the difference in an instance like that, so that each model ends up at least sorta properly exposed?

Jul 25 05 08:24 pm Link

Photographer

Herb Way

Posts: 1506

Black Mountain, North Carolina, US

Posted by SitaMaeEdwards: 

Posted by Herb Way: 
The situation you describe is the same as shooting a wedding with the bride in a white gown and the groom in a black tuxedo.  Famed wedding photographer, Gary Fong, suggests going into the camera's parameters and lowering the contrast and saturation.  I've done this with my Canon 10D and 20D and have realized improved results all around.  I rarely use my flash meter anymore.   

Herb, how did you figure out how much you wanted to lower your contrast and saturation? 

You'll have to experiment.  Gary Fong's recommendation for Canon cameras was two notches down on the saturation and one notch down on the contrast.  I got better results by doing just the opposite, two down on contrast and one down on saturation.  The first four shots in my portfolio were done at this setting.

Jul 25 05 08:31 pm Link

Photographer

Belair

Posts: 359

Downers Grove, Illinois, US

Posted by Brian Diaz: 
I mean this purely for the sake of discussion, and not to be contradictory, but what can one do with a meter that one cannot do with a digital camera and a grey card?

If you spot meter a gray card and take an incident meter reading at the same location, the EV should be the same.  Still, the most accurate means of metering is to spot meter a critical area, and place that area in the Zone you desire.

About 95% of the time, proper use of an incident meter will provide you with the contrast control you are looking for.

Jul 26 05 10:45 pm Link

Photographer

Saerbreathach_Photos

Posts: 2398

Toronto, Ontario, Canada

This sort of goes on a different take. The OP seems to know how to use the histogram, I however do not.  Can someone provide a bit of an explanation, i have a very basic understanding of it.

Jul 26 05 11:50 pm Link

Photographer

Jerry Nickels

Posts: 52

Phoenix, Arizona, US

The D70 does have the feature of showing blown out areas as flashing. I sometimes use this feature to check for blown out highlights. I use my meter but then double check by checking this feature. It helps if your making adjustments for filters and want to make sure you have it right.

Jul 27 05 12:03 am Link

Photographer

Brian Diaz

Posts: 65617

Danbury, Connecticut, US

Posted by Morphine Dream: 
This sort of goes on a different take. The OP seems to know how to use the histogram, I however do not.  Can someone provide a bit of an explanation, i have a very basic understanding of it.

Understanding Histograms

Jul 27 05 12:36 am Link

Photographer

Belair

Posts: 359

Downers Grove, Illinois, US

Posted by Morphine Dream: 

This sort of goes on a different take. The OP seems to know how to use the histogram, I however do not.  Can someone provide a bit of an explanation, i have a very basic understanding of it.

The histogram only provides you with a graphic illustration of the distribution of pixels across the tonal range from pure black to pure white.  It doesn't tell you where those pixels are...only a spot meter can do that.

I don't want to get too deep in the weeds on this histogram thing.  But, in my mind, there are two ways the histogram.

First, and I suspect the most common use, would be for the histogram to drive the exposure settings.  That is, the shooter will fiddle around with the f/stop until the histogram looks like it is "supposed" to look.  To me, this process sort of defeats the purpose of metering the scene and causes the shooter to defer authority to the camera's meter.  In fact, I believe that putting the camera in full auto program mode with no EV compensation will provide the "perfect" histogram in most cases.  Whether the shot is optimally exposed is another question.

The second, and I feel superior, use of the histogram is to use it as a confirmatory tool.  With practice, the experienced photographer should be able to look at a scene and pretty much know in his/her head how tones are distributed.  He/she would then confirm this by looking at the histogram.  Of course, the assumption is that the in-camera meter is being used to evaluate exposure.  Again, the hand held spot meter is the best means of evaluating exposure and renders the histogram display as nothing more than a novelty.

Jul 27 05 06:10 am Link

Photographer

Saerbreathach_Photos

Posts: 2398

Toronto, Ontario, Canada

Thanks guys, much appreciatted.

Jul 27 05 10:32 am Link

Photographer

Saerbreathach_Photos

Posts: 2398

Toronto, Ontario, Canada

Posted by Brian Diaz: 

Posted by Morphine Dream: 
This sort of goes on a different take. The OP seems to know how to use the histogram, I however do not.  Can someone provide a bit of an explanation, i have a very basic understanding of it.

Understanding Histograms

Hey thanks brian, just read it, very good to know.

Jul 27 05 10:40 am Link