How do I avoid photos like this?

[hector]

Does anyone have any pointers as to how not to end up with photos looking like this?

The photo was shot on auto on my D40, with sun facing the subjects. Why has it blown the sky completely to white? My camera seems to do this alot and I would love to know what I am doing wrong.
Would a polarizing filter sort it out?

Thanks in advance

 

[66217]

If you could give us some extra info about the photo we could help you more. Like:
-Shutter speed
-Aperture
-ISO

Also try using the manual mode. In the above photo I would guess maybe an Aperture of f/8, ISO 100 and maybe a shutter of 120 should give out a good exposure. Without using flash. Play with different combinations and tell us if it improves.

It could be a defect in the camera, but I doubt it. My D40x is far from taking good exposed photos in difficult conditions when in Auto Mode, tho in Manual Mode with the correct setting it takes excellent photos.

Make some more photos in Manual Mode and tell us how it goes.

 

[GoCubsGo]

If you're shooting on auto you're also most likely using some form of matrix metering. Matrix metering isn't always best and proof is in your results there.
The difference in light between the sky and the brick/subjects is clearly a good 2 stops at least. Matrix metering will attempt to find the best value overall and let you take your shot. Look at your histogram, I'm going to guess it's all out of whack.

Was that area where they stood darker? Perhaps not shade if the sun was on them, but darker as though they were in a corner? What time of day was it?

Your best bet would be to meter off the sky, the faces, and the area below where the brick and their bodies were and find the best neutral value. Your sky will suffer the most but the subjects won't have blown out clothing and skin.

Spot metering is often your friend.

 

[hector]

Thanks for the responses, I have been getting more confident with the manual mode since picking up a 50mm 1.8 e series cheap off ebay.
This was taken by a friend who is unfamiliar with SLRs as I am in the photo; hence auto mode.

Auto mode shot 400 ISO (?!), f5.6 and 1/125 shutter. Weird... Also yeh it will have been on matrix metering. No flash.
The whole area was in the sun, not too bright, the sky was kinda cloudy at around 8 in the morning (we were singing Jerusalem off the clock tower on may day morning to welcome in the summer!).

I haven't played around with different metering types, if i want to meter off the sky, how do i do that? Just point at the sky and half press the shutter? Wont that focus at the sky? I need to read the manual (again).

 

[worriedmac]

Those settings are too high for even a small amount of daylight. If your using your camera in auto, use exposure compensation to lower exposure for bright subjects. Although I would recommend manual settings by you and hand it over. The exposure would always be better.

 

[Westside guy]

Personally, I'd have spot-metered off of one of the faces given the conditions.

If the sun's right behind the photographer than a polarizer isn't going to help. Sometimes using the lens hood can help contrast (although not if the sun is behind the photog...).

 

[arogge]

The faces are overexposed, probably because you metered off the dark shirts. Unless you use a flash, there is nothing that you can do about the overblown sky highlights except to shoot at a different time of day. A polariser will not affect your exposure problem.

 

[Abstract]

With that sky, I bet matrix metering would normally have underexposed that shot rather than overexposed like it did in this instance. In this particular example, it seems that the camera got confused for a second and botched it up more than any other reason. I think if he took 2 photos within several seconds of each other, one of them would have turned out fine. Sometimes, the metering just gets confused for a second before seeing how bright the sky and underexposing "just in case".

Don't worry about it. Matrix metering is usually OK. I guess you could have either spot metered off their faces, or spot metered off the sky and used flash to make sure you guys (the 3 of you) weren't completely underexposed.

 

[hector]

Can anyone explain why there is such a difference in how the sky came out in these 2 photos, shot straight after each other, both on auto

Both were focussed at the building.
First shot ISO 400, shutter 1/125 and f5.6
Second shot ISO 200, shutter 1/160 and f6.3

So many photos I have been trying to take particularly of houses come out like the top pic and the blown highlight of the sky is really bugging me! It is the one thing that is spoiling the experience of DSLR ownership for since I got the D40 about a month ago, and I would love to know how to correctly expose daytime shots of buildings (like my previous point and shoot would do no problems) before I go traveling around Europe this summer (where I will obviously be taking loads of photos of cool buildings in the sunshine).

Thanks again!

 

[Over Achiever]

Are you shooting JPEG? I wonder if you can recover detail if you were using RAW NEF files in the overexposed images.

In the last two images, was the sun in the same position in both photographs? It looks like the second image has the sun behind you and to your right.

 

[Mantat]

If you look at the exposure info on each picture, you will see that the one with the blowout sky is much more exposed than the other. The reason is that the lightmeter took its reading from the dark part of the picture and wanted to put that part more exposed, so it overblow the sky.

How to fix this:
- when shooting people outsite ALWAYS use a flash. Take the light reading from the sky, then let the flash expose the people.
- building: shoot in HDR, or you can also always underexpose by 1-2 stops. But this might screw up your building...

Moral of the story: if the gap in the light reading is too big, something in your picture will be overexposed-underexpose. So you have to expose the dark subject with a flash, reflector, or use HDR.

Hope it helped!

 

[juanm]

I'm of the opinion that it's likely that, by accident, you set the exposure metering to either central or (more likely) spot, instead of matrix.
Spot metering would explain the difference in the exposure between two seemingly identical pictures, given that with a very small angle of metering, the slightest movement could give substantially different exposures (notice how the shirt of the guy in the middle, is properly "exposed"?

Edit: on the house pictures, the same thing happens. A quick look in Photoshop shows that the window (painted white) is right in the middle of the frame in the properly exposed picture.

 

[Abstract]

So many photos I have been trying to take particularly of houses come out like the top pic and the blown highlight of the sky is really bugging me! It is the one thing that is spoiling the experience of DSLR ownership for since I got the D40 about a month ago, and I would love to know how to correctly expose daytime shots of buildings (like my previous point and shoot would do no problems) before I go traveling around Europe this summer (where I will obviously be taking loads of photos of cool buildings in the sunshine).

I'm sorry, but your camera would do an even worse job. The only reason it didn't blow the sky like above is because it underexposed the building, making the sky look good. Your point and shoot didn't have magic technology in it. It's just a matter of luck with regards to that camera. You'll get both exposed properly if one isn't much brighter than the other. I mean, if you want to take a photo of a building that's facing the sun (ie: it's very well lit), the photo of the sky may turn out great as well.

If you take a photo where the sky is bright, but the building isn't for some reason, you're going to have problems. You can spot meter something in the photo that you trust (grass is great), then press "Exposure lock" (AE-L) so that your cameras exposure settings don't change while you compose your shot, and then press the shutter when you're ready to take a photo. Problem is that your sky will be blown out. What I don't like about this idea is that on a D40, the metering setting may be a hard one to find in your menus, and so may take some time to change back once you're finished taking the shot. I'm not sure.

If you don't want to use spot metering, then simply use your zoom to zoom into something you want to expose properly, and then lock exposure (AE-L). That way, you don't need to change metering modes.

I guess you can also meter the sky, hit AE-L, set exposure compensation to +1, compose your shot, and take a photo. That way, you'll get the best balance between properly exposed sky and building. Maybe neither will be perfect, but that's life.

 

[PkennethV]

the d 40 also tends to overexpose. In bright daylight, I set mine to -0.7 by default. In not-so-bright light, I bring it down to -0.3 and only at night do i keep it at 0.

 

[compuwar]

Can anyone explain why there is such a difference in how the sky came out in these 2 photos, shot straight after each other, both on auto

Both were focussed at the building.
First shot ISO 400, shutter 1/125 and f5.6
Second shot ISO 200, shutter 1/160 and f6.3

So many photos I have been trying to take particularly of houses come out like the top pic and the blown highlight of the sky is really bugging me! It is the one thing that is spoiling the experience of DSLR ownership for since I got the D40 about a month ago, and I would love to know how to correctly expose daytime shots of buildings (like my previous point and shoot would do no problems) before I go traveling around Europe this summer (where I will obviously be taking loads of photos of cool buildings in the sunshine).

Thanks again!

Every camera's sensor (or type of film back in the film days) has a thing called "Dynamic Range." Dynamic range is the range between the brightest value in a scene and the darkest value in a scene that can successfully be recorded.

The overall house scene has more dynamic range than a digital camera is capable of recording. Either you metered off the house in spot or center-weighted mode, or in matrix mode the camera decided that the house was the subject and metered the entire scene in an effort to properly expose the house. Because the sky was so much brighter than the house a correct exposure of the house "blows out" the much brighter sky. Basically pushing it to white, which is the highest exposure value. You're also shooting into the sun (looks like it's at front right,) so the difference in range is significantly larger in the brightest part of the image, which is large in the frame (the sky.)

In the second shot, the tonal range is within the sensor's range, since that entire image doesn't have dark spots like the trees, fence, black roof, and all the areas in the shadow because of where the sun is in the first shot. You're also shooting with the sun off of your right shoulder, where it's illuminating the house- bringing the level up towards the brightness of the sky. Since the tonal range is much more compressed, the sensor is able to render it all with a pleasing exposure value that's not all the way at the high end.

You'd generally have a bigger issue with a point and shoot, as dynamic range is a little better on larger sensors than it is on smaller sensors.

Others have already suggested metering methods, though if these weren't shot in matrix mode, that may be the biggest issue. I'll add the advice to only shoot into the sun if you're doing something other than automatic metering or you're using fill flash, and to learn to read the histogram and check shots that might be questionable.

Photography is all about light and shadow- your key light is the sun, you need to pay more attention to how it's lighting your subjects.

 

[Abstract]

Really? I've never heard of that problem with the D40. I've heard about the D80's approach to metering being different than every other Nikon's (prior to D300.....maybe the D300 uses the same approach), but I haven't heard the same for the D40.

 

[PkennethV]

Really? I've never heard of that problem with the D40. I've heard about the D80's approach to metering being different than every other Nikon's (prior to D300.....maybe the D300 uses the same approach), but I haven't heard the same for the D40.

just a quick google search

Exposure: All cameras are different. The D50 and D70 are straight-shooters and give great exposures as set.The D40 tends to overexpose, so I leave it set to -0.7 exposure compensation.

If the light gets contrasty the D50 and D70 aim for the highlights, which leads less-experianced photographers think they underexpose. The D40 has the same meter as the D80, which means it's very different and exposes for the shadows and will overexposure in contrasty light. Even in awful backlight the D40 won't let anything go black.

http://www.kenrockwell.com/nikon/d40/d40-recommendations.htm




and while I don't agree with the overall point of the article, saying that this basically makes the d 40 bad camera...

I tried out my new Nikon D40 DSLR with the 18-55 G ED II lens. The camera overexposes the pictures taken with or without the flash on

http://www.viewpoints.com/Nikon-D40-Digital-Camera-with-G-II-18-55mm-Lens-review-e2415

 

[Mantat]

If you know that the camera over expose a bit, just use the exposure compensation... I dont understand why people complain about this!

Anyways, this is a DSLR, you just have to check the result pic and correct the setting afterward. Over even better, just fix it in post production in Aperture.

 

[Abstract]

That Ken Rockwell quote isn't quite right. Several things to consider:

1. It's Ken Rockwell.

2. What he said about the D50 and D70 doesn't make sense. If the D50 and D70 were both "straight shooters" giving great exposures as set, then why did they underexpose? Also, why does Ken Rockwell say "...which leads less-experianced photographers think they underexpose"? It DOES underexpose. He just knows to use exposure compensation for every situation where he knows the camera will be fooled. It hardly seems fair to say that the D50 and D70 did a great job, while the D40 and D80 do a poorer job. They both need adjustment. KR is just used to using positive compensation rather than negative, and it tricks his brain into thinking, "Wow, this is difficult".


I owned the D50 for a long time, and trust me, the exposure wasn't perfect on that camera. I'd say that's true for every DSLR, from every brand, on the market today, which all take the same approach of underexposing rather than trying to expose well. The only case where this isn't true are the D40 (apparently), the D80, and (especially the) D300, and that's because they use a built-in look-up table inside to try and guess what type of scene you are trying to photograph.

If you're trying to photograph a scene where the top half of the frame is bright, but the bottom half is dark, it assumes that you're shooting the ground on the bottom half (which can be a grassy field, lots of buildings, streets, or anything else) and the sky on the top half. It assumes that you, the photographer, wants the ground to be properly exposed, while the sky can be blown out. To do this, it puts greater emphasis on the bottom half of the frame when metering the light and calculating the exposure. It assumes that you're most interested with whatever is on bottom half. It's actually doing a pretty good job of judging this, as most people really do want the ground to look great. Of course, if you want the sky to be perfect, and you're happy with the bottom half of the photo being a bit too dark, you're out of luck. Exposure compensation will have to be used.

 

[compuwar]

The only case where this isn't true are the D40, D80, and D300, and that's because they use a built-in look-up table inside to try and guess what type of scene you are trying to photograph.

Actually, this isn't true. Every Nikon body that has Matrix Metering uses the same approach- Nikon derives an algorithm from a bunch of scenes that the camera uses to make exposure decisions. That's everything from film bodies like the 8008s to the D3.

https://forums.macrumors.com/showthread.php?p=5380824

 

[GoCubsGo]

Can anyone explain why there is such a difference in how the sky came out in these 2 photos, shot straight after each other, both on auto

Both were focussed at the building.
First shot ISO 400, shutter 1/125 and f5.6
Second shot ISO 200, shutter 1/160 and f6.3

So many photos I have been trying to take particularly of houses come out like the top pic and the blown highlight of the sky is really bugging me! It is the one thing that is spoiling the experience of DSLR ownership for since I got the D40 about a month ago, and I would love to know how to correctly expose daytime shots of buildings (like my previous point and shoot would do no problems) before I go traveling around Europe this summer (where I will obviously be taking loads of photos of cool buildings in the sunshine).

Thanks again!

In addition to what compuwar said you made the statement that they were shot on auto one right after the other. I have to tell you that auto or manual the sequence in which you shot had nothing to do with the results. The settings were completely different, the angle of the glass to the sunlight also varied. Those things alone will result in two totally different photos as far as exposure goes (among other things).

 

[James_C]

Looking at all the pictures I suggest that he has the D40 set to Spot metering rather than Matrix Metering, Matrix metering should not blown the faces out on the first picture. Basically the spot meter system will meter just in a small circle in the centre of the image, and because the main subject is wearing a dark color, that is what is throwing the metering off.

Just my best guess

 

[Abstract]

Actually, this isn't true. Every Nikon body that has Matrix Metering uses the same approach- Nikon derives an algorithm from a bunch of scenes that the camera uses to make exposure decisions. That's everything from film bodies like the 8008s to the D3.

https://forums.macrumors.com/showthread.php?p=5380824

I heard it was now a lot more aggressive at how they approached metering, relying more heavily on the scene judgement. That's why so many people have complained about the D80's uncanny ability to blow out skies, thinking that the camera was wrong, while it's actually doing what it's told to do......to expose the likeliest subject perfectly while letting the big bright band in the top half of the frame get overexposed. It never used to do that, but I guess exposing a bit more also lowers signal-to-noise, so not only do you get more S/N, but potentially better exposure.

Also, I heard the D80 gets tricked far easier than the D200 in scenes where there is a small bright light near the centre of the frame (or several small bright lights) and you're using the center focus point. This supposedly wasn't a problem with the D50 (it didn't seem to be), D70, or D200, but was supposed to be a problem with the D80.

I'm not really sure though. It may have just been a monster-sized thread full of whiny people at DPReview who said it.

 

[66217]

How to fix this:
- when shooting people outsite ALWAYS use a flash. Take the light reading from the sky, then let the flash expose the people.

Just curious, but why would you recommend that?

I completely hate using flash. I only use it when I don't have sufficient light available. And in the conditions of the first photo from the OP I think that there was more than sufficient light.

If you are having trouble getting a pleasant sky color, then I think the best thing to do is to expose correctly the principal subject (in this case the three guys) and then in Aperture, Photoshop, etc. try making the sky look more vivid.

 

[PkennethV]

In summary, how not to get shots like that with the d 40...

-Use exposure compensation
-Review your shots (with the histogram)
-Make further tweaks if necessary