expose to the right?

[Ray]

POSTCARD from vacation "Desire" - Mexico

Hey! Ray,
The model you referring to is neither a big spender nor really likes to blow her budget. Matter of fact she is the biggest Polish Model discovery of 2007. Her current day/rate is a minimum of $10,000 + expenses. That is $30,000 for a three day shoot like the one below where she was not even the main model.
Andre
[a href=\"index.php?act=findpost&pid=151860\"][{POST_SNAPBACK}][/a]

Jeez! And I've been grumbling at paying just a couple of hundred dollars to persuade a Thai model to take off her clothes  .

With so much money at stake, I'd be bracketing every shot to increase my options. But I'm just an amateur.

[jing q]

Ray,
If you have experience that directly relates to this post such as your own use of a MFDB, then please share it. For example do you use ETTR with your MFD images or not? If not why? If so what do you do in post to get them looking right.

Eric
[a href=\"index.php?act=findpost&pid=151861\"][{POST_SNAPBACK}][/a]

one thing to consider (depending on how you shoot) is that blown out highlight details are alot more noticable than dense shadows...if you're working in a situation of variable exposure and you're shooting a human where the expressions are changing, practicing ETTR can risk extreme overexposure...over 1.2 stops (from what I've tried..) on the Leaf presents problems already in good highlight recovery

[Jonathan Wienke]

practicing ETTR can risk extreme overexposure...over 1.2 stops (from what I've tried..) on the Leaf presents problems already in good highlight recovery

If you have to "recover" highlights, you're going beyond what ETTR is all about. ETTR is about NOT blowing highlights in the RAW data, but rather is about keeping the highlights close to clipping, but not clipped. When shooting in fluctuating lighting, that usually means aperture priority mode, and setting a +1/3 or +2/3 exposure compensation over the meter default.

[jing q]

If you have to "recover" highlights, you're going beyond what ETTR is all about. ETTR is about NOT blowing highlights in the RAW data, but rather is about keeping the highlights close to clipping, but not clipped. When shooting in fluctuating lighting, that usually means aperture priority mode, and setting a +1/3 or +2/3 exposure compensation over the meter default.
[a href=\"index.php?act=findpost&pid=151874\"][{POST_SNAPBACK}][/a]

yes I know what ETTR is
And what I am saying is that there is more risk of blowing highlights sometimes because the meter readings are not always the most exact or trustworthy esp in fast changing light.

if I underexpose with my MFDB there's much more leeway for shadow recovery than highlight recovery. numerous people have noted this also with the output from MFDBs

[Jonathan Wienke]

It's true that you can't push the ETTR concept as hard in rapidly changing light; the risk of an unusable shot due to blown highlights is greater than the risk of an unusable shot due to excessively noisy shadows. Blown highlights are more problematic than noisy shadows whether shooting with a digicam, DSLR, or MFDB. But that doesn't negate the fact that an exposure where the highlights are close to, but not clipped (excluding small specular highlights) in the RAW results in a capture with the greatest dynamic range, lowest noise, and the most processing flexibility.

[jing q]

It's true that you can't push the ETTR concept as hard in rapidly changing light; the risk of an unusable shot due to blown highlights is greater than the risk of an unusable shot due to excessively noisy shadows. Blown highlights are more problematic than noisy shadows whether shooting with a digicam, DSLR, or MFDB. But that doesn't negate the fact that an exposure where the highlights are close to, but not clipped (excluding small specular highlights) in the RAW results in a capture with the greatest dynamic range, lowest noise, and the most processing flexibility.
[a href=\"index.php?act=findpost&pid=151881\"][{POST_SNAPBACK}][/a]

sure, theoretically.
strangely I find myself not practicing it even though I know the theory
seems that I find that having to bring the exposure down in post processing doesn't seem to retain the same kind of colour saturation I'm looking for.

maybe I'm just lazy, I find myself working better just throwing a curve on a image properly exposured or even slightly under.Gets a more saturated contrast on my images that suits my style.

I think it's great to know the theory but one should experiment with different techniques themselves to find what suits their visual taste best

I find photography quite ritualistic, sometimes we photographers will have crazy little quirks in the way we approach taking photographs and processing them. mine is just liking to see that my image on the screen properly exposured so that I can envision the right look of th eimage in my mind straight away.

[Henry Goh]

I agree with jing q.  Most of the time the best starting point to an image capture is a neutral WB and properly exposed one.  From there one can tweak to taste.  The only exception is when light is so contrasty that it is beyond sensor's ability.

[Wayne Fox]

Since this subject seems to be back on topic somewhat I have a question that has had me a little curious, which can only be answered by those really understanding the 1's and 0's (I hope Andrew is still lurking).


The main difference I see between MFDB and DSLR's ... at least most of them currently, is the bit depth.  Does a 14bit MFDB by nature capture more levels so without exposing to the right as far, you essentially gain a similar amount of data?  Or are we talking about different kinds of levels here?

A second question, if LR and ACR by default is non-linear, is it important to set up a linear default, normalize the EV, and then modify other parameters?  For example if I have a medium contrast curve, and pull the EV down, is my result substantially different than if I eliminate the curve, pull the EV down, and then put the same curve back in?  Does that make any sense?  Just trying to understand the most optimum workflow when I am using ETTR.

[Henry Goh]

12-bit: 1/2 of data is in the 12th bit
16-bit: 1/2 of data is in the 16th bit

MFDB with 16-bit has more levels to capture the darker shades than 12-bit.  That is my understanding.

[digitaldog]

The main difference I see between MFDB and DSLR's ... at least most of them currently, is the bit depth.  Does a 14bit MFDB by nature capture more levels so without exposing to the right as far, you essentially gain a similar amount of data?  Or are we talking about different kinds of levels here?

No, the distribution theory is still the same since again, its linear encoded data.

A second question, if LR and ACR by default is non-linear, is it important to set up a linear default, normalize the EV, and then modify other parameters?  For example if I have a medium contrast curve, and pull the EV down, is my result substantially different than if I eliminate the curve, pull the EV down, and then put the same curve back in?  Does that make any sense?  Just trying to understand the most optimum workflow when I am using ETTR.
[a href=\"index.php?act=findpost&pid=151890\"][{POST_SNAPBACK}][/a]

For me, my normalize settings in LR is simply a move of exposure. But that's just a starting point I use for import. From there, I may have no further adjustments or many to produce a rendering appearance I desire. The exposure slider does all the major work of making what by default looks over exposed, normally exposed.

I'm pretty sure based on seeing how Michael worked in Amazon, this is his basic starting point too.

[Jonathan Wienke]

Does a 14bit MFDB by nature capture more levels so without exposing to the right as far, you essentially gain a similar amount of data?  Or are we talking about different kinds of levels here?

If the noise level was exactly the same in absolute terms (both the 12-bit and 14-bit cameras have 4 bits of noise), then your proposal would be correct. But simply using an ADC with more bits of output is no guarantee that the additional bits are all signal and no noise. In the real world, some of those extra bits will be noise, so exposing all the way to the right with a 12-bit device will still give you a better capture than underexposing 2 stops with a 14-bit device.

The exact amount of benefit those extra bits will offer depends heavily on the specific cameras and their ISO setting. Comparing a 16-bit MFDB pushed to ISO 1600 to a 14-bit 1Ds-MkIII at ISO 1600 will probably yield different results than if both cameras are shot at ISO 100.

[bjanes]

Since this subject seems to be back on topic somewhat I have a question that has had me a little curious, which can only be answered by those really understanding the 1's and 0's (I hope Andrew is still lurking).
The main difference I see between MFDB and DSLR's ... at least most of them currently, is the bit depth.  Does a 14bit MFDB by nature capture more levels so without exposing to the right as far, you essentially gain a similar amount of data?  Or are we talking about different kinds of levels here?
[a href=\"index.php?act=findpost&pid=151890\"][{POST_SNAPBACK}][/a]

Many MFDBs have 16 bit ADCs and output 16 bits per color. That means that the brightest f/stop has 32767 possible levels compared to the 2048 possible levels with a 35 mm style digital camera with a 12 bit ADC. However, the human eye can perceive only about 70 of these 32767 levels and the rest are effectively wasted. It his helpful to have more levels than the eye can perceive to provide a margin of safety in processing, but the above margin is far greater than necessary.

In real world photography, dynamic range is limited by noise rather than quantization  and the quantization theory of the advantages of ETTR is overblown IMHO. The real advantage of ETTR is reduced noise and better dynamic range. If your exposure does not result in near full well photo sites, then you are not taking advantage of the capabilities of the sensor. If you reduce exposure by 1 stop and you lose 1 stop of DR. The signal drops by a factor of 1/2 (0.5) and the shot noise falls by 1/sqrt(2) or or 0.707.  The resultant signal to noise (S:N) is 0.5/0.707  or 0.707 of what it was previously. Contrary to popular belief, noise is actually higher in the highlights, but the S:N is better in the highlights.

The Kodak KAF-39000 39 MP sensor used in many high end MFDBs has a full well capacity of 60,000 electrons and a read noise of 16 electrons, giving a dynamic range of 3750:1 or 11.87 f/stops. The data sheet states a DR of 12 stops. This is no better than the latest Canon sensors and I see no reason to believe that the principles of ETTR are different for the MFDBs than 35 mm style DSLRs.

A second question, if LR and ACR by default is non-linear, is it important to set up a linear default, normalize the EV, and then modify other parameters?  For example if I have a medium contrast curve, and pull the EV down, is my result substantially different than if I eliminate the curve, pull the EV down, and then put the same curve back in?  Does that make any sense?  Just trying to understand the most optimum workflow when I am using ETTR.
[a href=\"index.php?act=findpost&pid=151890\"][{POST_SNAPBACK}][/a]

The adjustments are done on the linear data before the tone curve is applied, so the manipulations you suggest are not necessary. In my previous post, I held the white point constant, but let the black point vary. If you keep both of these constant, the tone curve does not change that much when you increase exposure in the raw converter.

In any event, you have full control over the tone curve in ACR, but the signal:noise is controlled by the actual in camera exposure, according to the number of photons captured. With this in mind, I can think of no reason for not exposing to the right, perhaps bracketing or else leaving a bit of headroom so as to not clip the highlight. Small amounts of overexposure can often be corrected with highlight recovery in the raw converter.

Bill

[jonstewart]

Bill, just curious: which MFDB do you use?

[samuel_js]

the brightest f/stop has 32767 possible levels compared to the 2048 possible levels with a 35 mm style digital camera with a 12 bit ADC. However, the human eye can perceive only about 70 of these 32767 levels and the rest are effectively wasted. It his helpful to have more levels than the eye can perceive to provide a margin of safety in processing, but the above margin is far greater than necessary.
Bill
[a href=\"index.php?act=findpost&pid=151961\"][{POST_SNAPBACK}][/a]

If the human eye has only 70 levels against 2048 of a 35mm. How do you explain that your eyes can register much greater DR?  
And how do you explain that my P21's DR is still under the human eye's DR capacity if so much levels are "efectively wasted"?

[Jonathan Wienke]

Could somebody PLEASE post two MFDB RAWS, one ETTR (but not so far to the right that the highlights are clipped), and one with less exposure, same subject, lighting, focus, etc. where the ETTR RAW is inferior to the RAW with less exposure?

So far, no proof that ETTR is flawed has been offered other than vague anecdotal assertions that a non-ETTR shot is easier to process or has better color saturation or superior tonality or whatever. If ETTR is as bad an idea when shooting with MFDB as Andre and some others here have claimed, then presenting such a pair of RAWS should be easy, and the advantage of processing one over the other should be straightforward and obvious. The advantages of ETTR have a solid basis in both mathematical theory and the practical real-world experience of many photographers and digital imaging experts. It's up to the critics of ETTR to prove that the earth is really flat after all. Is anyone up to the challenge of presenting some tangible evidence for their assertions, or do you all prefer to continue with personal attacks against anyone who dares to express skepticism?

[jonstewart]

If the human eye has only 70 levels against 2048 of a 35mm. How do you explain that your eyes can register much greater DR?   
And how do you explain that my P21's DR is still under the human eye's DR capacity if so much levels are "efectively wasted"?
[a href=\"index.php?act=findpost&pid=152006\"][{POST_SNAPBACK}][/a]

Actually, nobody knows for certain how many colours the eye (which is very adaptive in its perception) can perceive.

Commonly perceived knowledge now is that it is in the millions, and even suggestions of up to 100 million. I'm glad I have a 16 bit sensor with all those 'extra' colours!

[jonstewart]

Could somebody PLEASE post two MFDB RAWS, one ETTR (but not so far to the right that the highlights are clipped), and one with less exposure, same subject, lighting, focus, etc. where the ETTR RAW is inferior to the RAW with less exposure?

So far, no proof that ETTR is flawed has been offered other than vague anecdotal assertions that a non-ETTR shot is easier to process or has better color saturation or superior tonality or whatever. If ETTR is as bad an idea when shooting with MFDB as Andre and some others here have claimed, then presenting such a pair of RAWS should be easy, and the advantage of processing one over the other should be straightforward and obvious. The advantages of ETTR have a solid basis in both mathematical theory and the practical real-world experience of many photographers and digital imaging experts. It's up to the critics of ETTR to prove that the earth is really flat after all. Is anyone up to the challenge of presenting some tangible evidence for their assertions, or do you all prefer to continue with personal attacks against anyone who dares to express skepticism?
[a href=\"index.php?act=findpost&pid=152007\"][{POST_SNAPBACK}][/a]

But who cares? Most people in this forum seem to spend more time actually doing photography in a way that produces pleasing results for themselves and ultimately their clients. Most, if not all, have an understanding of what ETTR is, and how to do it, if they so want.

I happen to agree: expose correctly and you should have a little headroom, but why make all the fuss about it.

Just let it drop. You seem very intent on prolonging the argument long beyond it's sell by date, just to prove your point. I can see how this grates on some.

[samuel_js]

Could somebody PLEASE post two MFDB RAWS, one ETTR (but not so far to the right that the highlights are clipped), and one with less exposure, same subject, lighting, focus, etc. where the ETTR RAW is inferior to the RAW with less exposure?

So far, no proof that ETTR is flawed has been offered other than vague anecdotal assertions that a non-ETTR shot is easier to process or has better color saturation or superior tonality or whatever. If ETTR is as bad an idea when shooting with MFDB as Andre and some others here have claimed, then presenting such a pair of RAWS should be easy, and the advantage of processing one over the other should be straightforward and obvious. The advantages of ETTR have a solid basis in both mathematical theory and the practical real-world experience of many photographers and digital imaging experts. It's up to the critics of ETTR to prove that the earth is really flat after all. Is anyone up to the challenge of presenting some tangible evidence for their assertions, or do you all prefer to continue with personal attacks against anyone who dares to express skepticism?
[{POST_SNAPBACK}][/a]

Well, if you have the time to play with them, I've got these two files for you. Personally, and because i took the pictures, I think ETTR with a digital back is a bad idea, but that's personal experience. Please don't look at the artistic side of the photos, they are just part of a test...  
Hasselblad h2 with P21 and 35mmHC lens on tripod and mirror up, tif unprocessed:

[a href=\"http://www.samuelaxelsson.com/temp/ETTR.tif]ETTR.tif[/url]

NORMALEXP.tif

Good luck with the test.

[marcmccalmont]

Well I downloaded both (wish they were RAW's) using levels I set the white point the same and adjusted the normal exposure to match the ETTR exposure (the normal seemed a bit dark to me) and the ETTR looked better? More local contrast and dynamic range.
Marc

[bjanes]

If the human eye has only 70 levels against 2048 of a 35mm. How do you explain that your eyes can register much greater DR?   
And how do you explain that my P21's DR is still under the human eye's DR capacity if so much levels are "efectively wasted"?
[{POST_SNAPBACK}][/a]

The number of levels in one f/stop has nothing to do with dynamic range. I suggest you do a bit of reading, rather than making foolish statements. I would suggest you start [a href=\"http://www.normankoren.com/digital_tonality.html]here[/url].

The human eye can not see that wide of a dynamic range all at once without adaption as explained in this article on Wikipedia. 30 db is 1000:1, less than what your camera can record. Shifting your gaze about a high dynamic range scene and letting the eye accommodate is like taking a number of shots with your camera at various exposures and combining them into a HDR image. The trouble with digital sensors are that they are linear rather than logarithmic as is the eye, and they therefore waste a lot of bandwidth in the higher f/stops of the scene.

Bill