
|
What went wrong with this shot D5100 with f/1.8 lens
4 months ago
|
Hi all.
Taking shots at a school performance in a dark gym with no flash.
Most of the shots came out very nice, but a few came out like the shot below.
What happened here? It is a mess. Due to the low light I am fighting for a usable properly exposed picture, but to me this picture seems to have gotten over-exposed (not to mention blurry due to the long shutter speed).
The camera seems to sometimes want to use a lower shutter speed resulting in blurred over-exposed images. Sometimes the same settings gave a nice shot. Other times shots taken at faster shutter speeds came out looking much better.
I guess the point is that while taking shots using P mode and A mode, most shots came out looking pretty good, but a handful came out looking like the shot below. What causes such discrepancy? How can I prevent it? For the shot below I would have much preferred a faster shutter speed but in the low light the camera tries to keep the shutter speed slow. However, on many shots it used faster speed and those shots came out better even though it was low light.
Thanks!
Juggernaut
|
Re: What went wrong with this shot D5100 with f/1.8 lens
In reply to Juggernaut122,
4 months ago
|
The big black panel in the background is the culprit. It probably confuses the metering system of the camera (especially if you used matrix), causing it to overexpose. Because the metering system sees such a big black area, it overcompensates, blowing the highlights in areas that actually matter - such as faces. In cases like these, the best thing to do is to take a test shot (e.g. before the concert begins or while the performers prepare), check exposure, then use manual mode. Or, alternatively, you could try to use center/spot metering. Personally, I prefer Manual
--
Check my blog for reviews, tips & tutorials:
Amateur Nikon
Follow me on Twitter:
@amateurnikon
|
Re: What went wrong with this shot D5100 with f/1.8 lens
In reply to DigitalPhilosopher,
4 months ago
|
I also think that mettering failed due to the black background.
Personally I prefer to use A-mode with exposure compensation if needed (like in this case)
|
ADL + SPOT
In reply to DigitalPhilosopher,
4 months ago
|
DigitalPhilosopher wrote:
The big black panel in the background is the culprit. It probably confuses the metering system of the camera (especially if you used matrix), causing it to overexpose. Because the metering system sees such a big black area, it overcompensates, blowing the highlights in areas that actually matter - such as faces. In cases like these, the best thing to do is to take a test shot (e.g. before the concert begins or while the performers prepare), check exposure, then use manual mode. Or, alternatively, you could try to use center/spot metering. Personally, I prefer Manual
You will always have a few shots that are inconsistent when combining ADL (Active D-Lighting) set to "auto" and spot metering as is the case here. ADL is meant to be used with Matrix metering and not spot. ADL set to "Auto" will simply add to the inconsistencies. If you want the same every time...use ADL with Matrix and set something other than "auto". But note...auto does get it right most of the time. Good Luck
--
My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)
|
Re: ADL + SPOT
In reply to Mako2011,
4 months ago
|
Mako2011 wrote:
combining ADL (Active D-Lighting) set to "auto" and spot metering as is the case here.
Where did you see spot metering was used for this photo? I didn't see anything like that on the exif...
--
Check my blog for reviews, tips & tutorials:
Amateur Nikon
Follow me on Twitter:
@amateurnikon
|
Re: ADL + SPOT
In reply to DigitalPhilosopher,
4 months ago
|
DigitalPhilosopher wrote:
Mako2011 wrote:
combining ADL (Active D-Lighting) set to "auto" and spot metering as is the case here.
Where did you see spot metering was used for this photo? I didn't see anything like that on the exif...
Exposure Time (1 / Shutter Speed) = 10/600 second ===> 1/60 second ===> 0.01667 second
Lens F-Number / F-Stop = 20/10 ===> ƒ/2
ISO Speed Ratings = 320
Original Date/Time = 2013:01:30 20:24:08
Digitization Date/Time = 2013:01:30 20:24:08
Exposure Bias (EV) = 0/6 ===> 0
Max Aperture Value (APEX) = 16/10 ===> 1.6
Max Aperture = ƒ/1.74
Metering Mode = spot (3)
Light Source / White Balance = unknown (0)
Flash = Flash did not fire
Focal Length = 500/10 mm ===> 50 mm
Custom Rendered = normal process (0)
Exposure Mode = auto exposure (0)
White Balance = auto (0)
--
Patco
A photograph is more than a bunch of pixels
|
Re: What went wrong with this shot D5100 with f/1.8 lens
In reply to Juggernaut122,
4 months ago
|
Juggernaut122 wrote:
Hi all.
Taking shots at a school performance in a dark gym with no flash.
Most of the shots came out very nice, but a few came out like the shot below.
What happened here? It is a mess. Due to the low light I am fighting for a usable properly exposed picture, but to me this picture seems to have gotten over-exposed (not to mention blurry due to the long shutter speed).
The camera seems to sometimes want to use a lower shutter speed resulting in blurred over-exposed images. Sometimes the same settings gave a nice shot. Other times shots taken at faster shutter speeds came out looking much better.
I guess the point is that while taking shots using P mode and A mode, most shots came out looking pretty good, but a handful came out looking like the shot below. What causes such discrepancy? How can I prevent it? For the shot below I would have much preferred a faster shutter speed but in the low light the camera tries to keep the shutter speed slow. However, on many shots it used faster speed and those shots came out better even though it was low light.
Thanks!
Juggernaut
Welcome to the world of Event Photography... where you'll very commonly shoot in poorly illuminated auditoriums and gymnasiums. And you'll learn a few tricks!
As others have mentioned, you used Auto Exposure, probably ADL though the Exif data doesn't say so one way or the other, and spot metering. The specific problem of course is that the "spot" was on a fairly dark area right in the center of the image, so it brightened the whole image.
Next time you'll want to do it a little differently. Use Manual Exposure mode. Use a higher ISO too, to allow both a higher shutter speed (say about 1/100) and also allow you to stop the lens down a bit too. That shot was at ISO 320, 1/60 and f/2. You might want to try ISO 1600, 1/125 and then stop down to say f/4. That would give you a lot more Depth of Field, would freeze any movement much better, and would reduce exposure by about 2/3 of a stop (which may not be enough, and if so you might try ISO 800 or something in between).
The Light Value from the Exif data says 6.2 EV, which isn't really bad for this kind of thing. Some are 1 EV higher, but some are 2 EV lower too!
Eventually you'll also decide that a 50mm lens doesn't provide enough functionality. You'll want to look at a 70-200mm f/2.8 VR. Not cheap, but note that that is not the newest version with VRII. The older version will be just fine on a DX sensor body like the D5100.
|
This is why VR works well...
In reply to Juggernaut122,
4 months ago
|
One problem is using a fast lens for a group shot rather than a slower lens with VR. You shot this at F2 which has a super shallow DOF, it's not possible for everyone to be in focus. Also the iso is too low, don't be afraid to use iso 800 or even 1600.
I would have center weighted metering, fixed the iso to 800, shot it at f5.6, maybe even f8 and honestly the kit lens with VR would have been a better choice. For group shots VR beats a fast lens used opened up 2 stops hands down, even in low light.
--
Stacey
|
Re: This is why VR works well...
In reply to Stacey_K,
4 months ago
|
DOF is also dependant on the distance from the subject...He looks like he was at more than 4 meters than his subject and he's using a DX camera with a 50mm at F2...if you calculate you'll see that the DOF is 0.5 meters!
F5.6 or F8 is insanely low on that light there. He would have to use super high iso to get a decent shutter speed and his photo will be ruined by noise.
Stacey_K wrote:
One problem is using a fast lens for a group shot rather than a slower lens with VR. You shot this at F2 which has a super shallow DOF, it's not possible for everyone to be in focus. Also the iso is too low, don't be afraid to use iso 800 or even 1600.
I would have center weighted metering, fixed the iso to 800, shot it at f5.6, maybe even f8 and honestly the kit lens with VR would have been a better choice. For group shots VR beats a fast lens used opened up 2 stops hands down, even in low light.
--
Stacey
--
Pushing the Nikon D90
http://floppyrom.deviantart.com/gallery/
|
Re: This is why VR works well...
In reply to mehigh,
4 months ago
|
mehigh wrote:
DOF is also dependant on the distance from the subject...He looks like he was at more than 4 meters than his subject and he's using a DX camera with a 50mm at F2...if you calculate you'll see that the DOF is 0.5 meters!
F5.6 or F8 is insanely low on that light there. He would have to use super high iso to get a decent shutter speed and his photo will be ruined by noise.
Stacey_K wrote:
One problem is using a fast lens for a group shot rather than a slower lens with VR. You shot this at F2 which has a super shallow DOF, it's not possible for everyone to be in focus. Also the iso is too low, don't be afraid to use iso 800 or even 1600.
I would have center weighted metering, fixed the iso to 800, shot it at f5.6, maybe even f8 and honestly the kit lens with VR would have been a better choice. For group shots VR beats a fast lens used opened up 2 stops hands down, even in low light.
--
Stacey
--
Pushing the Nikon D90
http://floppyrom.deviantart.com/gallery/
+1, not to mention that VR doesn't help with subject movement.
|
Re: This is why VR works well...
In reply to mehigh,
4 months ago
|
F5.6 would be iso 800 at 1/30. Or iso 1600 at 1/60. Hardly extreme settings. You honestly think 1.5 ft of dof (if your calculations are correct) would keep everything sharp? And of course you assume the perfect focus plane would be shot, focusing errors happen.
And maybe I'm crazy but the noise at iso 1600 on that camera isn't bad + noise can be fixed, out of focus can't. I'm not sure why so many people on this forum are obsessed with shooting group shots like this at F2. I highly doubt these kids are moving so fast that 1/30 shutter speed would be blurred with VR, esp on a 50mm setting. I've shot this exact same type event using the setting I suggested and the shots came out great, everyone was in focus etc.
I know it's a commonly parroted opinion that fast glass beats a kit lens with VR but the reality is, for many situations like this, it just doesn't. Fast glass is great for subject isolation etc but for group shots where you need DOF, it's not.
--
Stacey
|
Re: ADL + SPOT
In reply to Patco,
4 months ago
|
Patco wrote:
Max Aperture = ƒ/1.74
Metering Mode = spot (3)
*note to self: check eyesight*
--
Check my blog for reviews, tips & tutorials:
Amateur Nikon
Follow me on Twitter:
@amateurnikon
|
Re: This is why VR works well...
In reply to Stacey_K,
4 months ago
|
I have to agree with Stacey. Her advice is solid in my opinion.
|
What and How exactly is SPOT metering doing in this situation?
In reply to apaflo,
4 months ago
|
apaflo wrote:
Juggernaut122 wrote:
Hi all.
Taking shots at a school performance in a dark gym with no flash.
Most of the shots came out very nice, but a few came out like the shot below.
What happened here? It is a mess. Due to the low light I am fighting for a usable properly exposed picture, but to me this picture seems to have gotten over-exposed (not to mention blurry due to the long shutter speed).
The camera seems to sometimes want to use a lower shutter speed resulting in blurred over-exposed images. Sometimes the same settings gave a nice shot. Other times shots taken at faster shutter speeds came out looking much better.
I guess the point is that while taking shots using P mode and A mode, most shots came out looking pretty good, but a handful came out looking like the shot below. What causes such discrepancy? How can I prevent it? For the shot below I would have much preferred a faster shutter speed but in the low light the camera tries to keep the shutter speed slow. However, on many shots it used faster speed and those shots came out better even though it was low light.
Thanks!
Juggernaut
Welcome to the world of Event Photography... where you'll very commonly shoot in poorly illuminated auditoriums and gymnasiums. And you'll learn a few tricks!
As others have mentioned, you used Auto Exposure, probably ADL though the Exif data doesn't say so one way or the other, and spot metering. The specific problem of course is that the "spot" was on a fairly dark area right in the center of the image, so it brightened the whole image.
Next time you'll want to do it a little differently. Use Manual Exposure mode. Use a higher ISO too, to allow both a higher shutter speed (say about 1/100) and also allow you to stop the lens down a bit too. That shot was at ISO 320, 1/60 and f/2. You might want to try ISO 1600, 1/125 and then stop down to say f/4. That would give you a lot more Depth of Field, would freeze any movement much better, and would reduce exposure by about 2/3 of a stop (which may not be enough, and if so you might try ISO 800 or something in between).
The Light Value from the Exif data says 6.2 EV, which isn't really bad for this kind of thing. Some are 1 EV higher, but some are 2 EV lower too!
Eventually you'll also decide that a 50mm lens doesn't provide enough functionality. You'll want to look at a 70-200mm f/2.8 VR. Not cheap, but note that that is not the newest version with VRII. The older version will be just fine on a DX sensor body like the D5100.
I find all this very new and interesting to me. What or how exactly is Spot Metering doing in this scenario? When you say it "brightened the whole image"? What is it that gets increased automatically, pure exposure compensation, what about ISO ?? On the exif file it does not show for example: "+2" exposure compensation.
-
|
How to get more consistent exposure.
In reply to Juggernaut122,
4 months ago
|
If you are in an auditorium and the lights aren't changing, then use P or A mode to find the right exposure you like and then switch over to manual mode and use those exposure settings.
If you are in P or A mode, everytime you aim your camera at someone with a white or black shirt for instance, your camera will increase or lower exposure based on what it seeing, thus potentially ruining everything else in the scene.
If you are under non-changing lighting conditions, it's best to take your camera off any kinda auto exposure modes because there is no reason to keep changing exposure. Obviously, if you want to go up and down in aperture for creative reasons, then you should adjust shutter speed or ISO accordingly.
--
--
Antonio
http://www.intensitystudios.com
|
Re: What and How exactly is SPOT metering doing in this situation?
In reply to mistermejia,
4 months ago
|
mistermejia wrote:
I find all this very new and interesting to me. What or how exactly is Spot Metering doing in this scenario? When you say it "brightened the whole image"? What is it that gets increased automatically, pure exposure compensation, what about ISO ?? On the exif file it does not show for example: "+2" exposure compensation.
-
The camera will want to expose so that whatever is under the metering point appears as middle grey. If the point is over black, the scene will appear over exposed; if it is over white, the scene will be under exposed. How it does this will depend on what mode you are in.
Note too that even PAS and M with AutoISO are affected in matrix metering mode because the exposure is biased towards the selected focus point.
For this type of application where lighting is constant, I would either meter a tone in the scene, lock the exposure and adjust EC or find the exposure I like in A mode with matrix metering and then switch to manual using those parameters.
|
Re: What and How exactly is SPOT metering doing in this situation?
In reply to mistermejia,
4 months ago
|
Reflective light meters, as in cameras whether set to matrix, center-weighted or spot, presume a scene that approximates to mid grey (18% grey card). For most photographs this is fine as most scenes contain a range of tones. The problem arises when the scene to be photographed is excessively dark or excessively light ie contains a lot of black or a lot of white. In the the scene of the OP there is a large black back drop, matrix might have evened out the exposure between the white shirts and the black back drop but it looks like the narrow "spot" metered a dark area and presuming mid-grey opened up the exposure to render the black grey and hence blowing the highlights or whites for these were now overexposed.
The camera does not apply exposure compensation it is just the way it meters the scene. The answer is to apply exposure compensation "-" for blacks "+" for whites either by changing the aperture or the shutter speed. Changing ISO does not apply exposure compensation it just changes the sensitivity of the sensor. It might appear that is does for the exposure will change but the light on the subject is exactly the same. The amount exposure compensation depends on the amount of dark or light in the scene. A snow covered landscape in bright sunlight may need as much as 2 stops exposure compensations. At least with digital you have an instant display and histogram to help you judge. If in doubt bracket.
In camera metering has come a long way from when it was first introduced but it still needs to be used intelligently by reading the scene and applying adjustment accordingly. 80-90% of shots will probably meter effectively with matrix. You can of course buy 18% grey cards to meter the light falling on you subject and then use those set manually on your camera.
I hope that helps.
|
Re: What and How exactly is SPOT metering doing in this situation?
In reply to GRimage,
4 months ago
|
GRimage wrote:
Reflective light meters, as in cameras whether set to matrix, center-weighted or spot, presume a scene that approximates to mid grey (18% grey card). For most photographs this is fine as most scenes contain a range of tones. The problem arises when the scene to be photographed is excessively dark or excessively light ie contains a lot of black or a lot of white. In the the scene of the OP there is a large black back drop, matrix might have evened out the exposure between the white shirts and the black back drop but it looks like the narrow "spot" metered a dark area and presuming mid-grey opened up the exposure to render the black grey and hence blowing the highlights or whites for these were now overexposed.
The camera does not apply exposure compensation it is just the way it meters the scene. The answer is to apply exposure compensation "-" for blacks "+" for whites either by changing the aperture or the shutter speed. Changing ISO does not apply exposure compensation it just changes the sensitivity of the sensor. It might appear that is does for the exposure will change but the light on the subject is exactly the same. The amount exposure compensation depends on the amount of dark or light in the scene. A snow covered landscape in bright sunlight may need as much as 2 stops exposure compensations. At least with digital you have an instant display and histogram to help you judge. If in doubt bracket.
In camera metering has come a long way from when it was first introduced but it still needs to be used intelligently by reading the scene and applying adjustment accordingly. 80-90% of shots will probably meter effectively with matrix. You can of course buy 18% grey cards to meter the light falling on you subject and then use those set manually on your camera.
I hope that helps.
Nice! Thank you.
|
Re: This is why VR works well...
In reply to Stacey_K,
4 months ago
|
Stacey_K wrote:
One problem is using a fast lens for a group shot rather than a slower lens with VR. You shot this at F2 which has a super shallow DOF, it's not possible for everyone to be in focus. Also the iso is too low, don't be afraid to use iso 800 or even 1600.
I would have center weighted metering, fixed the iso to 800, shot it at f5.6, maybe even f8 and honestly the kit lens with VR would have been a better choice. For group shots VR beats a fast lens used opened up 2 stops hands down, even in low light.
--
Stacey
Hi Stacey. Thanks for the ideas.
I have tried shots like this with the kit lens - I personally am not able to get anything that looks good - always have to increase the ISO too high - like 3200 - too high for my liking (I don't have time to post process).
I also took shots with the 55-200 kit lens. A few were acceptable.
The f/1.8 lens gets decent shots most of the time, but as you mentioned the shallow DOF is not always desired. But how can I get bright shots without using very high ISO, unless I am using a large aperature? Most of the professionals at these venues are using the f/2.8 70-200 lenses & shooting wide open. I think the kit lenses only open up to 3.5.
(hehe, I wish it were easier to change lenses so I can compare more easily!!)
Thanks!
Juggernaut.
|
Re: This is why VR works well...
In reply to Stacey_K,
4 months ago
|
Stacey_K wrote:
F5.6 would be iso 800 at 1/30. Or iso 1600 at 1/60. Hardly extreme settings. You honestly think 1.5 ft of dof (if your calculations are correct) would keep everything sharp? And of course you assume the perfect focus plane would be shot, focusing errors happen.
And maybe I'm crazy but the noise at iso 1600 on that camera isn't bad + noise can be fixed, out of focus can't. I'm not sure why so many people on this forum are obsessed with shooting group shots like this at F2. I highly doubt these kids are moving so fast that 1/30 shutter speed would be blurred with VR, esp on a 50mm setting. I've shot this exact same type event using the setting I suggested and the shots came out great, everyone was in focus etc.
I know it's a commonly parroted opinion that fast glass beats a kit lens with VR but the reality is, for many situations like this, it just doesn't. Fast glass is great for subject isolation etc but for group shots where you need DOF, it's not.
I've shot a lot of event photography in poorly lit school gymnasiums for several years. Most of this I agree with, and a little I don't.
That camera will easily get good shots using ISO 1600 (and for that matter at ISO 6400 too). Up to ISO 1600 there won't be enough noise to worry about, though the dynamic range does go down and it is imperative to expose to get highlights with desired detail (all those white shirts on the kids) at just below clipping. That is the functionality that histograms provide so well.
It is simply operator error to shoot a group shot at f/2! At f/2 the camera to subject distance has to be more than 30 feet to get 10 feet of DOF. Hence I agree that f/5.6 is a valid target for aperture. At a 20 foot camera to subject distance f/5.6 gives 12 feet of DOF, which pretty much guarantees everything will be in focus. f/4 gives 8 feet of DOF and that is probably good enough, particularly if the subject distance is more like 25 feet or greater, and if one is careful about where to focus in order to center the DOF on the group.
I don't quite agree about VR though. With a 50mm lens on a DX camera VR is almost useless. Camera motion is not the problem and subject motion is much more likely. The shutter speed would best be kept at least at 1/60, and I'd rather have 1/125.
Note that the image is over exposed, so it actually needs to have exposure reduce by perhaps 1 fstop. ISO 1600 would add 2-1/3 fstops, and using f/5.6 would reduce that by 3 stops giving a total reduction of about 2/3 of an fstop. Hence f/4 at 1/125 might be okay too, if the subject distance is greater than 25 feet or so.
The big issue is to use Manual Exposure, setting it just one time by looking at the histogram. The lights aren't going to change...