Lack of contrast sample (1 Viewer)

hampers said:

Personally, I think the consistent high quality of Roy's images demonstrate that he has a good system for his exposures. The thread started with an example for discussion, it provoked that and has served to show that people have different views and techniques and of course everyone is entitled and should have their own views.
Personally I think I will try going more to the right than I currently do, that is normally 1/3 stop, gradually and see if it improves the amount of detail in my images.
Thanks for starting the debate Roy, I have watched it with interest.

Phil

Click to expand...

Thanks for that Phil :t: I am glad that someone has got a little out of it.

For everyone out there, I did not start this thread with a view to converting anyone to ETTR but more to show anyone that if you do get a washed out look image then it is simply a 10 sec job to correct in post. Having said that it has certainly provoked some healthy debate which can only be a good thing IMHO.

For what it is worth I have been using the ETTR method for bird photography some three or four years now and would not do it any other way. The fact that a lot of folk do not use this method is of no concern to me.

For those that have not followed up on Paul's link to Art Morris's site on the subject of ETTR HERE it is well worth a read if you are interested in what ETTR is all about. I was a member of Art's site http://www.birdphotographers.net/forums/forum.php well before Art was a convert and can still remember the timewhen he was first converted to ETTR - BTW there are some excellent tutorials on processing on the site.

Paul Corfield said:

ETTR wasn't mentioned till post No.9 which is why some (including me) got confused as to the intitial point of the discussion. I think the thread should have been started with ETTR as the subject from the outset. Apart from that I've learnt something new from Roy's posts.

Paul.

Click to expand...

Why I started the thread was simply to show how easy and quick it is to correct a 'washed out' look image Paul. Pretty soon it was turned into a 'you are doing something wrong' OR 'there is some wrong with your gear' debate. This is why I decided to disclose that the washed out look is common place if you ETTR. I incorrectly assumed that a lot of folk on here would be doing this anyway and I did not envisage that it would turn into a ETTR thread. Lots of threads seem to change course mid way through but it all helps stimulate debate which is what it is all about IMHO.

Paul Corfield said:

ETTR wasn't mentioned till post No.9 which is why some (including me) got confused as to the intitial point of the discussion. I think the thread should have been started with ETTR as the subject from the outset. Apart from that I've learnt something new from Roy's posts.

Paul.

Click to expand...

My thoughts as well.
have already begun looking into this ettr-thing. Can't wait to try it out in the field.

Histogram - What does it really show?

All this discussion on ETTR, JPGs and RAW causes me to wonder what does the histogram really show in a camera.

Does the histogram represent the full dynamic range of a RAW image (12-14 bits) or the smaller dynamic range of a JPG (8 bits)? Does the histogram automatically change to allow for the larger dynamic range when shooting RAW? The 8 bit JPG values are presumably created by discarding data, but is the discarded data from the top or bottom of the dynamic range? Is the x axis of the histogram linear or logarithmic?

TIA,
Brian

Roy,
I have never obtained results with contrast as low as the first one. (On the other hand I do not over-expose as much as you seem to do).
/Tord

Edit - I now realize the thread is not a complaint about the SW80ED contrast, it is about ettr...

Brian O'C said:

All this discussion on ETTR, JPGs and RAW causes me to wonder what does the histogram really show in a camera.

Does the histogram represent the full dynamic range of a RAW image (12-14 bits) or the smaller dynamic range of a JPG (8 bits)? Does the histogram automatically change to allow for the larger dynamic range when shooting RAW? The 8 bit JPG values are presumably created by discarding data, but is the discarded data from the top or bottom of the dynamic range? Is the x axis of the histogram linear or logarithmic?

TIA,
Brian

Click to expand...

Hi Brian

Here is a link to an article that may cast more light on the subject. Hope it helps.
http://www.cambridgeincolour.com/tutorials/digital-exposure-techniques.htm

BR
Jaco

Roy C said:

Why I started the thread was simply to show how easy and quick it is to correct a 'washed out' look image Paul. Pretty soon it was turned into a 'you are doing something wrong' OR 'there is some wrong with your gear' debate.

Click to expand...

To be fair to others and myself, your first post was asking a lot questions like you weren't sure why you were getting a lack of contrast. I think it's perfectly reasonable to assume people are going to come back with answers/opinions to your questions, especially as you specifically asked for peoples take on this lack of contrast. In that respect it's perfectly reasonable to assume it was fogging of the lens because the effect would have been exactly the same. Everyone here is of a different ability or takes photos for different reasons, so why assume everyone is using ettr or that they should automatically know this is what you are doing in post No.1. You created the problem and then get defensive when people offer up opinions based on the minimum of facts.

Paul.

Brian O'C said:

All this discussion on ETTR, JPGs and RAW causes me to wonder what does the histogram really show in a camera.

Does the histogram represent the full dynamic range of a RAW image (12-14 bits) or the smaller dynamic range of a JPG (8 bits)? Does the histogram automatically change to allow for the larger dynamic range when shooting RAW? The 8 bit JPG values are presumably created by discarding data, but is the discarded data from the top or bottom of the dynamic range? Is the x axis of the histogram linear or logarithmic?

TIA,
Brian

Click to expand...

Brian, When shooting in RAW the in-camera preview and histogram is just a jpeg representation of the RAW image and will almost certainly not be accurate.
Although it has no baring on the RAW file itself it is the in camera shooting style that you have set which determines the jpeg preview.

This raises a good question when it comes to ETTR - "when is a blinky not a blinky?". If you are using a normal shooting style and the Camera preview shows a slight blinky (highlight clipping) then it is almost certain that when you actually load your RAW file into your converter there will be no clipping whatsoever. The reason is that the in camera preview would have increased contrast and saturation ..... but this is not present in the actual RAW file.

There are two ways to counteract this when trying to shoot ETTR:-

1) Leave in in-camera shooting style as it is and just be aware of how much clipping you can get away with before it actually effects the RAW image. This is not a bad solution for anyone who is worried about blowing highlights when shooting to the right - providing the highlights are just clipping you can be sure that your RAW file will not be blown.

2) Create a custom in-camera shooting style that produces an in-camera jpeg/histogram that is more reflective of the RAW file. This is done by reducing Contrast and saturation. Remember this will not change your RAW file in any way but the preview and histogram will be a lot more reflective of the RAW file.

For what is is worth option 2 is my preferred method of doing it but either one will aid the quest for ETTR.

Paul Corfield said:

To be fair to others and myself, your first post was asking a lot questions like you weren't sure why you were getting a lack of contrast. I think it's perfectly reasonable to assume people are going to come back with answers/opinions to your questions, especially as you specifically asked for peoples take on this lack of contrast. In that respect it's perfectly reasonable to assume it was fogging of the lens because the effect would have been exactly the same. Everyone here is of a different ability or takes photos for different reasons, so why assume everyone is using ettr or that they should automatically know this is what you are doing in post No.1. You created the problem and then get defensive when people offer up opinions based on the minimum of facts.

Paul.

Click to expand...

I am not having a go at anyone and did admit that I mistakenly assumed that a lot of folks on here would be using something like ETTR. Its not as if this is anything new as there has been so much discussion about it all over the net for years.
I also freely admit that I do not know much at all about photography when compared to most of the folks on here. With my limited photographic skills I try to gain every advantage I can when trying to get a reasonable image which is why I use things like ETTR. I only post things to try and help folks but in the future will just leave it you more knowledgeable guys to spread the word B

As a parting shot I will say that in my limited experience with different telephoto lenses over the years that the 'faster' the lens then the better out-of-Camera contrast you get, this is especially true when using a converter as they are bound to reduce contrast somewhat, no what lens you are using.
Baring in mind that an astroscope like the SW80ED at f7.5 is 'slower' than any Current DSLR lens available then it is little wonder to me that the contrast it produces can be limited at times. Obviously most folks on here are not seeing this but I can only call it as I see it.

Well, I enjoyed this discussion and it made me learn a lot even if it started strangely. Thanks Roy.

Here are some thoughts:

We are very far from the old traditional film camera days... Over-exposing was then a no-no because if was more difficult to recuperate than under-exposition. There was not much room for error. And this was only true if you processed your own photos: there was no margin for error with commercial photo finishing done with machines. Most serious photographers had a light-meter hanging from their neck and we were trying to set exposition as accurately as possible. There was no digital noise then - yes, there was grain but it was behaving differently.

When I bought my first digital reflex, a Canon 10D , I used to set exposure at -1/3 to prevent blown highlights, as many photographers did. It became a habit and I kept that setting until recently when Adobe LightRoom 4 came out. In LR4, adjusting a RAW file became extremely easy and straightforward with the Highlights, Shadows, Whites and Blacks sliders. It made me change from -1/3 to +1/3, changing the setting only if I got Blinkies or for special situations. I feel this improved dynamic range quite a bit.

I still have problems with setting exposure as much to the right as possible, and even more at times, as Roy suggests. First, I think it is looking for problems because a photo with clipped highlights is ruined. Second, I still wonder if some details in the whites and highlights are not lost. There are still strong discussions about the benefits of ETTR: while everybody seems to agree with the benefits of doing it with moderation, many have doubts about extreme ETTR like Roy suggest.

Extreme ETTR means that the exposure settings must be adjusted for each picture. For me, this is a problem when using regular camera lenses for birding. I like to do my birding walking around slowly, looking for birds instead of being stationary, waiting for the bird. This means that many of my shots must be taken rapidly and I often have no time for manual adjustments. I try to set the camera for the expected light condition and speed requirements. Trying to manually set exposure for maximum ETTR without clipping would cost me a lot of birds and would mean a lot of photos with burned highlights.

Astro-scoping is a different story. First, it requires manual focus. Second, because of the nature of the beast, it requires a tripod and we must set up the equipment. Third, I shoot from longer distances and the birds are usually less weary. I find that I usually have more time available for a shot on average compared to regular birding so it may be interesting to try extreme ETTR.

Quebec winter is almost over and it will be time for digiscoping again. I will definitely try it and see how it goes.

Roy C said:

Baring in mind that an astroscope like the SW80ED at f7.5 is 'slower' than any Current DSLR lens available then it is little wonder to me that the contrast it produces can be limited at times. Obviously most folks on here are not seeing this but I can only call it as I see it.

Click to expand...

It's not that we don't see it, but as you stated already, a little less contrast is so easily and quickly corrected, so I guess noone here have thought that as an issue. And we already know that focking all that not is focked improves contrast.

Roy C said:

Brian, When shooting in RAW the in-camera preview and histogram is just a jpeg representation of the RAW image and will almost certainly not be accurate.
Although it has no baring on the RAW file itself it is the in camera shooting style that you have set which determines the jpeg preview.

Click to expand...

So in PP what does the ACR histogram display when ACR is set up in 16 bit mode? Does the information in the ACR histogram vary when you change the bit depth? Given that ACR only supports 8 and 16 bit modes, does this mean the histogram x axis should be 256 times longer in 16 bit mode? Or is it like an audio signal where increasing the bit depth does not increase the maximum signal level but only lowers the recovered analogue noise floor? Most of what I have read in the photographic world implies an increase in maximum signal level when shooting raw, but in the audio world extra bit depth translates to a better recovered signal to noise ratio with little or no maximum recorded signal level improvement. Do the extra bits get used differently in cameras - extra magnitude rather than increased resolution?

If the maximum recordable signal level is the same, independent of the recorded bit depth, why would raw blinkies be different to JPG blinkies? In each case you have exceeded the maximum signal level when the blinkies appear and are losing IQ.

ETTR is not without it's risks and tradeoffs. It is fine if you have the time and if the subject allows. Here is the original tutorial on the subject by Michael Reichmann which is clear and easily understood. He and his friend Thomas Knoll of PS and Camera Raw fame, were the pioneers on the subject.
But he warns "Also be aware though that by doing this you are in fact effectively lowering the ISO used to capture the image, requiring slower shutter speeds and/or larger apertures. If you are hand-holding the camera, or shooting moving objects, the tradeoff may not be worth the reduced noise level." If this doesn't apply to photographing birds with our "slow" lenses, I don't know what does.
If you have the time to shoot, check, adjust, shoot again, check again, adjust again until the histogram looks just right, fine. But how often does that happen while photographing notoriously uncooperative birds? That is why I bias my exposure +2/3 with spot metering. This gives a noticeable improvement in the noise levels, and thus, detail, without running too much risk of blowing highs. If you can get away with +2EV, which is what the original picture looks like to me, great. Just push the black slider to the right and bingo. But +2EV as a starting point IMHO too risky.

tried the ettr method today. It was hard to adjust all the time (with mixed lighting conditions), but I manage to have 1 1.3 1.7 or more ev-compensation most of the time. Most shots were iso500, but last ones was iso1000 (unthinkable with my E-3 ;-) )

On this particular image it did not get too much exposed to the right, but what I found surprising was that the iso1000 was very nice. The samples here are processed in LR and the noise slider up to 20.

Female Blackbird
iso 1000
1/1600s

ETTR is much easier with subjects like this with a rather narrow tonal range. Have fun trying it on an Egret!

I think there maybe a bit of confusion with regards to ETTR for bird photography. Figures like Ev +1/3 or Ev + 2 are completely meaningless when it comes to correctly exposing the bird – each situation requires its own consideration as to how much (if any) exposure compensation is needed. Remember in bird photography the aim is to correctly expose the bird often at the expense of the background.

The three things to consider is:-
a) The tone of the bird against the tone of the background
b) How big the bird is in the frame (e.g. how much bird against how much background is occupying the frame)
c) The metering method you use.
All I would say about the metering method is that it is best to use one method and to get to know exactly how the Camera reacts in various situations with this method – if you keep switching methods you will not know where the heck you are. Most of the top bird photographers use evaluative (canon speak) all the time but I tend to use partial. Spot metering has its use but should be used very carefully IMHO, in some circumstances it can give strange results and trying to gauge the amount of exposure comp needed in all the different scenarios is nigh on impossible.

I guess the classic example to bird photography exposure is where there is a dark bird perched high on a tree against a light sky and where the bird is small in the frame. We have all been here when we first started, you let the camera determine exposure and end up with a perfectly exposed sky with a little black blob for the bird!
Sure you could selectively lift the bird in post but you cannot put back detail and tones that were not captured to begin with and also any noise will be amplified. In this case you may need up to Ev +3 to correctly expose the bird. It is far better to correctly expose the bird in the first place and then deal with the background later on in post.

On the other hand I was shooting a little egret a few weeks ago where the bird was in bright sunlight (not a good idea for white birds) against a dark background and it needed Ev-1 to ETTR on the bird.

ETTR is nothing more than correctly exposing the bird while making sure that you shoot as far to the right as you dare. It is most certainly not about over exposing.

If you are shooting in jpeg then the Cameras histogram/blinkies warning is you absolute guide, ‘what you see is what you get’ If you do happen to clip some of the highlights it is very difficult to do much about it in post.

On the other hand if you shoot in RAW things are a little different. For a start the histogram/blinkies warning is not a true reflection of the RAW file, if you expose so that parts of the bird are slightly clipping you will almost certainly find that when you open your RAW file there will be no clipping whatsoever (but you will have exposed pretty much to the right).
Also with a good RAW processor you still have a little leeway in even pulling back detail on a clipped area, certainly good for up to 2/3 of a stop in my experience. If all else fails and you still have some small areas of ‘blown’ detail then even these can be ‘fixed’ in Photoshop.

A side benefit to shooting to the right is that noise levels are kept to an absolute minimum level on the bird as has been mentioned above – noise on the background is of little consequence and can easily be fixed in post whereas noise on the bird itself can be difficult to deal with without losing fine detail.
The other consequence of shooting to the right is that the black level will often need correcting in post as Cameras are not capable of capturing the full dynamic range – this is a ten second job as previously mentioned.

Correctly exposing the bird is by far the most difficult thing in bird photography and it does not come easy, just letting the Camera auto expose or dialling-in a set amount of exposure compensation all the time will get the occasional correct result but overall it is a sure way to screw things up.

"Figures like Ev +1/3 or Ev + 2 are completely meaningless when it comes to correctly exposing the bird"
How can you say that? You are contradicting yourself. Your original shot looks to me, as I said, to be taken at about +2 Ev. I wasn't looking at anything other than the bird and I was looking at it as I would using the smallest possible spot metering I can (as I always do) and measuring a part of the bird of medium brightness. That is the whole point of spot metering. You measured the light on the bird and pushed the histogram to the right as far as possible, (I am assuming you did it deliberately) and that looks like about +2Ev. So, where is the problem? Ev compensation applies to what you want it to apply to, be it the underparts of a bird in a tree against the sky or an entire landscape.
Setting a + or - Ev shift in camera as a default is nothing more than a starting point, depending on what the sensor will generally allow, for times when there is not enough time to measure, shoot, check, adjust, shoot, check again etc.. Nobody said anything about sticking by it on a P+S basis. On my Nikon D7000 I feel safe with +2/3 Ev, whereas on my old Olympus E-510 I had to set it at -2/3 because the risk of blowing the high was so much greater. These are STARTING POINTS and nothing more. Sometimes you have to adjust one way, sometimes the other, depending on the subject.
Your shot has a rather narrow tonal spectrum and that makes ETTR and histogram reading easy. If the bird were an Avocet you would have had a much harder time because the histogram cannot know that you are only interested in the bird and not the background, unless there is a really clear spike on the right as there would be with your Egret in the sun on a dark background. The higher the contrast, the more difficult ETTR becomes, no mater what the subject mater is.
Exposure techniques for birds is in no way different than for anything else. Cameras can't think and make selective decisions. They can only help us make them, but in the end, we have to tell the camera what to do.

I do not get it?? You started a thread complaining about "lack of contrast you can get 'straight out of the Camera' with the SW80ED (at least with my set-up)", then covered your tracks and turned it into a lecture on ETTR.
ETTR is nothing other than selective exposure compensation/manipulation to get the most out of a digital sensor. No big deal and nothing all that new.
None of the rest of seem to be plagued by such low contrast images with our SW scopes. On the contrary. Most of us find the contrast quite good. But then maybe the rest of us don't like to push the histogram so far to the right all the time. It is not always called for.
Enough.

Dan, when it comes to photography I have always thought that a picture is better than a thousand words and to this end I invite you to have a quick look at my Photostream and 500px gallery if you have time (links in my signature). Just about all the shots were taken with my ETTR method.
I am only too aware that the general quality of my shots is not up to a lot of folk’s standards but I do think that most of the exposures are about right. You may well think otherwise which is fine – we all have our own opinions.
In this case we should agree to disagree on the subject and leave it there.

BTW I have deleted my last post - it was made in haste and was in poor taste.

I will not be bothering you folks again soon as I have now got a new 'toy' to play around with

DanC.Licks said:

If you have the time to shoot, check, adjust, shoot again, check again, adjust again until the histogram looks just right, fine. But how often does that happen while photographing notoriously uncooperative birds? That is why I bias my exposure +2/3 with spot metering. This gives a noticeable improvement in the noise levels, and thus, detail, without running too much risk of blowing highs. If you can get away with +2EV, which is what the original picture looks like to me, great. Just push the black slider to the right and bingo. But +2EV as a starting point IMHO too risky.

Click to expand...

I would agree with your sentiments Dan. I am generally a birder-photographer where speed of reaction to capture the bird often doesn't allow the luxury of experimenting with different settings on a single subject. I have recently been on a birding holiday where I set my default to give +2/3 with spot metering (aiming to reduce noise) but as described, this then reduces shutter speed or increases ISO which hampered me in a few situations. Generally I was pleased with my results however I've still got much to learn in being able to change my settings faster than I do.