Discussion on exposure techniques with dslrs

[PeterD]

Hi all

Since posting my thread Little Egret here on the World of Nature section, an interesting debate started between LocoLoco and myself re exposure settings. We both hoped that other forum members would join in but we realised that it was not appropriate to have this debate in that section of the forum because we were not giving others a chance to join in.

May I ask you to visit that thread and read the debate so far and then we can continue it here on this thread. It is an important topic. We all have our own views on the subject but it would be useful for us to share them. We can learn from each other what works better and therefore improve. I am no expert but I am keen to hear the points of view from others.

Thanks for reading this and I hope we can get this discusion going on here.

 

[wilb87]

maybe a better timing is needed for the shoot, eg. when the sun isn;t that strong?

 

[PeterD]

maybe a better timing is needed for the shoot, eg. when the sun isn;t that strong?

Thank you for your reply.
What you suggest is how I used to approach this. It does make selecting the correct exposure easier.
In bright sunlight, I too have used spot measurements on the white plumage to prevent over-exposure. This though causes problems at the darker end of the tone curve. Detail is lost here. This is what I and LocoLoco found.

 

[Dream Merchant]

Lazing around before going for CNY dinner when I saw this.

Nice of you to open it up for discussion among other members Peter!


Couple of things you need to know about in-cam metering (I'm assuming your not using a hand-held?).

1) Different systems do it differently as their base-guideline of 'middle gray'.

2) In-cam 'spot' metering is NOT spot metering, but more like a 'selective-area' metering. For spot, a 1 degree circle is the norm, and some meters let you zoom out to about 4 degrees.

In cam 'spot' you get about 12 - 15 degrees, WITH bleeding if you're metering two high-contrasting scenes, e.g. the Erget's face against the darker water or conversely, the Erget's face in darkness with strong backlight. Both situations will 'fool' the 'spot' meter.

I'm not sure if later models of cameras have a much smaller and more defined area of sensitivity of their in-cam 'spot' metering.

Canon's 1D MK IV states:

Spot metering (3.8% of VIEWFINDER)

Nikons D3s states

• Spot: Meters approx. 4 mm dia. CIRCLE (about 1.5% of frame) centered on selected focus point (on center focus point when non-CPU lens is used)

Problem is, we don't have the actual dimensions of the viewfinders and whether it the length, breadth or the total area, and the description of the D3s stumps me.


Anyway, just food for thought.

If you really want very, very accurate 1 degree spot readings, use a dedicated hand-held spot meter.

Did I mention that different manufacturers use entirely different percentages to calibrate their 'middle gray'?




Then start discussing in-cam metering techniques because the tools and their parameters are slightly different.

 

[PeterD]

Lazing around before going for CNY dinner when I saw this.

Nice of you to open it up for discussion among other members Peter!


Couple of things you need to know about in-cam metering (I'm assuming your not using a hand-held?).

1) Different systems do it differently as their base-guideline of 'middle gray'.

2) In-cam 'spot' metering is NOT spot metering, but more like a 'selective-area' metering. For spot, a 1 degree circle is the norm, and some meters let you zoom out to about 4 degrees.

In cam 'spot' you get about 12 - 15 degrees, WITH bleeding if you're metering two high-contrasting scenes, e.g. the Erget's face against the darker water or conversely, the Erget's face in darkness with strong backlight. Both situations will 'fool' the 'spot' meter.

I'm not sure if later models of cameras have a much smaller and more defined area of sensitivity of their in-cam 'spot' metering.

Canon's 1D MK IV states:



Nikons D3s states



Problem is, we don't have the actual dimensions of the viewfinders and whether it the length, breadth or the total area, and the description of the D3s stumps me.


Anyway, just food for thought.

If you really want very, very accurate 1 degree spot readings, use a dedicated hand-held spot meter.

Did I mention that different manufacturers use entirely different percentages to calibrate their 'middle gray'?




Then start discussing in-cam metering techniques because the tools and their parameters are slightly different.

Thank you for your very useful reply

I own both the Olympus E3 and Olympus E5. In both cases the manual quote the spot metering is 'approx 2% for the viewfinder screen'.

My approach to spot metering has been to take the exposure on the body of the bird (usually significantly greater than the 2% circle). I am not sure if this approach skews the image exposure.

 

[wootsk]

In studio, everything is base on what is in the studio how you control it. You can have full control of your lighting with strobe, lamp, honeycomb, light meter, reflector and the list goes on. Under sun, you will have to made do with what you have brought along as well as the sun and cloud.

One of my favorite way of shooting is when the sun is strong, I would use reflector to get a strong light shining from the rear of the model so as to get a "glow" around the model while using side bounced by umbrella flash for front lighting. Distracting background will be wipe off with a shallow DOF.

 

[daredevil123]

In studio, everything is base on what is in the studio how you control it. You can have full control of your lighting with strobe, lamp, honeycomb, light meter, reflector and the list goes on. Under sun, you will have to made do with what you have brought along as well as the sun and cloud.

One of my favorite way of shooting is when the sun is strong, I would use reflector to get a strong light shining from the rear of the model so as to get a "glow" around the model while using side bounced by umbrella flash for front lighting. Distracting background will be wipe off with a shallow DOF.

That will be a little hard for TS, especially when he is shooting birds mostly... u know the ones with wings and real feathers.

 

[allenleonhart]

That will be a little hard for TS, especially when he is shooting birds mostly... u know the ones with wings and real feathers.

hmm. isnt this when post processing kicks in?:think:

sometimes, we just do our best to get the best exposure. we tweak afterwards. i'm not saying that on the spot exposure is not important, but under tricky situations, u pretty much cant do anything except pray that the shot is correct, right?

my take on this. u meter the situation, shoot in raw, get enough detail in ur exposure, then tweak in photoshop. check the histogram to see how much "information" is being registered.

there isnt something called correct exposure. probably best exposure i guess:think:

just my take on the practicality of the thing lar. i find it easier and less time consuming if i do it that way.

but ultimately, rubbish in, rubbish out.

 

[divinemoment]

This is an interesting subject and taking birds of white feathers is really tricky and one could have easily overblown the subject if not exposed correctly. I have seen far too many wedding photos with overblown white gowns (even taken by professional) and this is the least thing I would expect. Imaging paying top dollar for the gown with sequins and intricate lace not showing up. The reflection of the water is adding to the challenge where it can gives out overblown white spot, and the task is made even more formidable when it is around mid day.

My method (which I also use in general photography) has been always av. metering and using EV to compensate. This method suit me best without needing to readjust different mode of metering whatever genre of photography. As you have mentioned in your previous thread, at times I do find minus one stop is essential to get the feathers properly exposed but also at the expense of having all other areas under exposed. Its really no other choice when taking such subject (unless someone can offer a better method). I believed HDR may work to a certain extend if the bird remain stationary but personally most of the times I do not like the outcome of the final image (looks arty to me).

In my experience and using my GH-1, I find details can be recovered without any noticeable noise from under one stop and even at times up to 1.5 stop under. and that considering using jpeg.

Here's a collection of my white swans taken during different times of the day.
http://david-low.smugmug.com/Other/Largest-Flying-Bird/11608874_yonxc#822189441_6sG3P

P/S. The two egrets are well exposed.

 

[wootsk]

That will be a little hard for TS, especially when he is shooting birds mostly... u know the ones with wings and real feathers.

Oh sorry, I didn't find where TS says about shooting bird, my bad. I don't shoot birds, but regarding the title to bird shooting, my guess should be spot metering and exposure lock. Because luxury timing for stuff like lighting might miss the shot. Maybe after the first shot, then can consider about the rest.;p

 

[CamInit]

Not into birds but would it be feasible to expose to right in RAW and then do single shot HDR in PP?

 

[Gengh]

I've taken pictures of egrets in various lighting conditions, and I understand the potential difficulty with these birds.

Sometimes I avoid metering on the birds altogether, and choose an easier reference object instead. Like the green grass beside the egret that's getting the same light. Spot meter on the grass in M mode and give another +1/3EV ('cos I know that that gives good exposure on my cam, generally the average green grass works like a "green card" for me. The +1/3EV may or may not be necessary depending on what the sun is like that day), then go back to the egret and shoot. When I do that, the egret will be bright, but not blown out. I find that this works best for me, since I don't have to worry about the colour of the bird I'm shooting.

Of course, this only works 'cos egrets and other wading birds are generally so calm and cooperative, can take my time to meter and compose.

 

[PeterD]

In studio, everything is base on what is in the studio how you control it. You can have full control of your lighting with strobe, lamp, honeycomb, light meter, reflector and the list goes on. Under sun, you will have to made do with what you have brought along as well as the sun and cloud.

One of my favorite way of shooting is when the sun is strong, I would use reflector to get a strong light shining from the rear of the model so as to get a "glow" around the model while using side bounced by umbrella flash for front lighting. Distracting background will be wipe off with a shallow DOF.

That will be a little hard for TS, especially when he is shooting birds mostly... u know the ones with wings and real feathers.

Thank you both for your replies. Studio work is a particular situation where you do have more control and time to set up a shoot. It can still be difficult but as daredevil points out, it is very much more difficult to do outside. in natural light.

 

[PeterD]

hmm. isnt this when post processing kicks in?:think:

sometimes, we just do our best to get the best exposure. we tweak afterwards. i'm not saying that on the spot exposure is not important, but under tricky situations, u pretty much cant do anything except pray that the shot is correct, right?

my take on this. u meter the situation, shoot in raw, get enough detail in ur exposure, then tweak in photoshop. check the histogram to see how much "information" is being registered.

there isnt something called correct exposure. probably best exposure i guess:think:

just my take on the practicality of the thing lar. i find it easier and less time consuming if i do it that way.

but ultimately, rubbish in, rubbish out.

You make some very interesting and valid points.

1. 'Rubbish in, rubbish out'. Here you highlight the need to get it right BEFORE any post processing as PP cannot correct for all exposure issues. If any part of the image is over-exposed, information in that part has been irretrievably lost. Information in the shadow or dark areas is still there and can be recovered albeit with some noise.

2. 'There isn't something called correct exposure. Probably best exposure I guess'. I think that correct exposure and best exposure are one and the same thing. Any image you take has its own unique bright and dark tones and the image histogram shows this. Each camera model has its own way of dealing with this and all are limited by the fact that brightness of each channel has to be encoded digitally. These two things will determine the image output from the camera. Given this, the best (or correct) exposure will give an image that retains all of the detail.

3. Shooting in raw gives the most opportunity to improve the final image. If you shoot in jpg then the image processor within the camera has made its own decisions on how the image data should be corrected. Yes, this can give good results but not always optimal. It also gives you less information to make adjustments in PP.

4. Most of the discussion so far has been on spot measurements but this is not the only way to set the exposure.

 

[PeterD]

This is an interesting subject and taking birds of white feathers is really tricky and one could have easily overblown the subject if not exposed correctly. I have seen far too many wedding photos with overblown white gowns (even taken by professional) and this is the least thing I would expect. Imaging paying top dollar for the gown with sequins and intricate lace not showing up. The reflection of the water is adding to the challenge where it can gives out overblown white spot, and the task is made even more formidable when it is around mid day.

My method (which I also use in general photography) has been always av. metering and using EV to compensate. This method suit me best without needing to readjust different mode of metering whatever genre of photography. As you have mentioned in your previous thread, at times I do find minus one stop is essential to get the feathers properly exposed but also at the expense of having all other areas under exposed. Its really no other choice when taking such subject (unless someone can offer a better method). I believed HDR may work to a certain extend if the bird remain stationary but personally most of the times I do not like the outcome of the final image (looks arty to me).

In my experience and using my GH-1, I find details can be recovered without any noticeable noise from under one stop and even at times up to 1.5 stop under. and that considering using jpeg.

Here's a collection of my white swans taken during different times of the day.
http://david-low.smugmug.com/Other/Largest-Flying-Bird/11608874_yonxc#822189441_6sG3P

P/S. The two egrets are well exposed.

Thanks for your reply

You are quite right that this issue is not just for wildlife photography but is a general one affecting most types of photography.

I have had a look through your images in your gallery link and you have managed to get some superb images of the swans. They do show that trying to achieve this result in the highlights is at the expense of the darker area of the images. I am confident that you should be able to recover these, whilst retaining the detail in the highlights, using post processing. Thank you for adding this link.:thumbsup:

I agree with you that HDR is not really appropriate for this type of image. We strive to produce images of wildlife as naturally as posible and take steps to ensure the scene is as we saw it at the time.

Thanks for your comments on the Egrets

 

[PeterD]

Oh sorry, I didn't find where TS says about shooting bird, my bad. I don't shoot birds, but regarding the title to bird shooting, my guess should be spot metering and exposure lock. Because luxury timing for stuff like lighting might miss the shot. Maybe after the first shot, then can consider about the rest.;p

Yes I think you are correct in saying review the image results in the histogram, make adjustments and try again. This assumes of course that you have the opportunity to do this.

Not into birds but would it be feasible to expose to right in RAW and then do single shot HDR in PP?

Whilst I have already said I am not in favour of HDR I would be VERY interested if any forum members have attempted to do this.

Thank you both for your replies

Please feel free to post your own images on this thread if they will further the course of this discussion:thumbsup:

 

[PeterD]

I've taken pictures of egrets in various lighting conditions, and I understand the potential difficulty with these birds.

Sometimes I avoid metering on the birds altogether, and choose an easier reference object instead. Like the green grass beside the egret that's getting the same light. Spot meter on the grass in M mode and give another +1/3EV ('cos I know that that gives good exposure on my cam, generally the average green grass works like a "green card" for me. The +1/3EV may or may not be necessary depending on what the sun is like that day), then go back to the egret and shoot. When I do that, the egret will be bright, but not blown out. I find that this works best for me, since I don't have to worry about the colour of the bird I'm shooting.

Of course, this only works 'cos egrets and other wading birds are generally so calm and cooperative, can take my time to meter and compose.

Thank you for your reply

You raise an interesting method here and I shall give it a try and report back my experience.

 

[allenleonhart]

Each camera model has its own way of dealing with this and all are limited by the fact that brightness of each channel has to be encoded digitally. These two things will determine the image output from the camera. Given this, the best (or correct) exposure will give an image that retains all of the detail.

mm. i think this is where dynamic range would start to kick in. one way to max the information would be to shoot in the native base ISO i guess?

for me, often the dynamic range is never enough (i love shooting landscapes), thus there is pretty much no correct exposure, cause the difference between the highlights and shadows are often too great:bsmilie: so i'll usually do exposure blending/hdr to get the details back. perhaps thats why i have a best exposure. one best one for shadows, one for midtones, last one for highlights:bsmilie:

 

[wootsk]

mm. i think this is where dynamic range would start to kick in. one way to max the information would be to shoot in the native base ISO i guess?

for me, often the dynamic range is never enough (i love shooting landscapes), thus there is pretty much no correct exposure, cause the difference between the highlights and shadows are often too great:bsmilie: so i'll usually do exposure blending/hdr to get the details back. perhaps thats why i have a best exposure. one best one for shadows, one for midtones, last one for highlights:bsmilie:

There is limitation for electronic sensor. That's one of the reason I am shoot shooting film now. Anyway when you view on your monitor, most monitor doesn't have that high range as well.:cry:

Enjoy some hidden dark truth about most marketing post the Analog to Digital convertor D-range value, but never the actual sensor D-range value. Can actually plot test chart, but I am lazy;p

Many digital sensors have less dynamic range than color print film. However, some newer CCDs such as Fuji's Super CCD, which combines diodes of different sensitivity, have improved upon this issue.
When highlights burn out, they burn to white without details, while film cameras retain a reduced level of detail, as discussed above.

 

[PeterD]

Thank you for your reply.

mm. i think this is where dynamic range would start to kick in. one way to max the information would be to shoot in the native base ISO i guess?

for me, often the dynamic range is never enough (i love shooting landscapes), thus there is pretty much no correct exposure, cause the difference between the highlights and shadows are often too great:bsmilie: so i'll usually do exposure blending/hdr to get the details back. perhaps thats why i have a best exposure. one best one for shadows, one for midtones, last one for highlights:bsmilie:

This is very true. To get the best IQ, every camera model does have a best ISO value. In the case of Olympus E3/5 this is ISO 200. This however does not mean that you should only use that value.

This is where PP does come in as it allows you to amplify or attenuate the image tone curve to adjust the brightness through the image. Noise etc are also dealt with in PP.
Can you please post some example images that you have using your methods? I am sure that we would all like to see them.