Expose to the right...

have been using this technique from Luminous Landscape web for a long time and I find it very good. However, I am used to working with ACR and its ability to recover highlight detail. The problem comes when opening the same NEF files (I always shot RAW) in NC, since it is not capable of recovering highlight detail.

So, I have lots of NEFs, exposed to the right. that are very good when opened with ACR but that cannot be opened with NC due to the blown highlights. Given this situation, I am considering giving up this technique, since I am getting "digital negatives" (NEFS) that I´m not sure I can recover in the future.

There will always be a program like ACR which can recover highlight detail? I HOPE SO! What do you think of? What´s your opinion? Do you use "Expose to the Right" rule?

I wish to read your comments!

My first thought is "How far in the future?" I suspect Adobe will be around at least as long as Nikon. How long either of them will support the current NEF format is anybody's guess. For long term archive TIF and JPG seem to be the best bets.

On exposing to the right, I practice moderation. Reichman's arguments make sense (and I've learned a lot from his site), but unless you have a three-color histogram on camera going to far to the right risks blowing one or more of the color channels. And once blown, those highlights can never really be recovered.

--
J.R.

Somewhere south
of Amarillo

... that´s what I wanted to read. Thanks!

I suspect that highlight recovery will be a 'magic' that will only get better with time given how important many deem it to be. You have nothing to worry about.

Besides, there's nothing stopping you from keeping your current version of ACR around indefinately.
--
http://www.photoalberta.ca

http://forums.dpreview.com/forums/read.asp?forum=1020&message=14513757

In the above thread I purposefully took a scene with a very wide DR, beyond the DR of any DSLR.

See the man on the left hand side: his shirt is blown out:



Same RAW image, this time developed for the highlights:

Exp: -1.50
Shadows: 5
Brightness: 27
Contrast: +35
Saturation: -17 (ACR overstaurates Fuji images at default 0 setting)



Now developed again, this time for the shadows:

Exp: +0.05
Shadows: 5
Brightness: 95
Contrast: +35
Saturation: -15
WB also modified 'cos WB in shadows is different.



Now for the BLEND in PSCS:

Dark image over light one.
select background (light image)
CRTL+C
click Layer 1 (dark image)
Add layer mask
ALT + click on layer mask
CRTL+V
Gauss blur 60 pixels
Click on layer thumb
Flatten
(Auto levels for some punch)

Voila':



Alot of detail "hidden" in that 2-3 yr old S2. Try it out with your Nikon, & tell us.

--

'Does Phil live on this planet, or is he on his way here?'
(subtitle line in obscure 70s Indian movie

Anyone can understand the EtoR theory because of the lineal nature
of digital data collection. Yes, there is more data on the light side of
the data, but (at time) people believe this is a God like rule and can not
or should not be altered.

The final check point for this rule is your prints and your desired results
from the image. If both of these require a bending of the EtoR rule, then
by all means, do so.

For my portrait work, I want a bell shape curve and even one with some
nice rich mid to dark tones. That sort of results comes from a bell shaped
curve that is high on the scale and does not have pushed highlights.

At least with portraits, I want the extremes of the dynamic range, but the
majority of data in the mid tone range.

Therefore, the EtoR theory can be bent like most rules. Do I loose some data?

Probably so, but the prints will tell the final story and carry along that mid tones
and rich darker tones that I want in the prints. Following the EtoR rule would
simply not render the desired effect.

Hope this helps.

qwebeck said:

have been using this technique from Luminous Landscape web for a
long time and I find it very good. However, I am used to working
with ACR and its ability to recover highlight detail. The problem
comes when opening the same NEF files (I always shot RAW) in NC,
since it is not capable of recovering highlight detail.

So, I have lots of NEFs, exposed to the right. that are very good
when opened with ACR but that cannot be opened with NC due to the
blown highlights. Given this situation, I am considering giving up
this technique, since I am getting "digital negatives" (NEFS) that
I´m not sure I can recover in the future.

There will always be a program like ACR which can recover highlight
detail? I HOPE SO! What do you think of? What´s your opinion? Do
you use "Expose to the Right" rule?

I wish to read your comments!

Click to expand...

--
Find the answer to your photography question here:
http://www.researchetc.com/scanners/2/index.htm

I would convert the Nikon NEFs to Adobe DNG (Digital Negative Format), it's an open format, it's documentation is freely available so even in 10 years from now it wouldn't be a problem to program a converter to file format then in use, even if both Nikon and Adobe would then be out of business. You may also embed the original raw file (NEF in your case) inside the DNG file. At this time it's directly supported by Leica (for their digital back) and Hasselblad.

I wouldn't use JPGs for archival because of the not lossless compression. TIFFs are better in this regard, but you loose the advantage of the raw format if you archive your images in TIFFs.

--
Best regards,
Vlad

How can I convert from NEF to DNG? do I need a plugin?

Thaks a lot!

qwebeck said:

How can I convert from NEF to DNG? do I need a plugin?

Thaks a lot!

Click to expand...

Go to the Adobe website and look for downloads of the DNG. I have not actually used it but I believe you can download a standalone application that will do the NEF ---> DNG convert for you. And, as far as I know it's completely free!

Brian

qwebeck said:

How can I convert from NEF to DNG? do I need a plugin?

Click to expand...

You can download a free batch DNG converter from the Adobe web site.

Another detail to bear in mind: if you are shooting compressed NEFs, the expose-to-the-right theory does not apply because Nikon's lossy compression method reduces the precision in the highlights.

But the DNG converter has a lossless compression method that produces smaller files than Nikon's lossy compression.

'Exposing to the right' is an amateur technique for getting around problems and ensuring you often get a resonable, nothing special, but usable file.

Of course the real answer is to exposure correctly. If you have a minor technical challange, then that is what you are paid for to get around as a pro.

When I am shooting panorama shots that span a large brightness range, capturing the brightest shot to the right allows the most brightness range to be captured above the noise level. Doing a normal exposure (which I did when shooting jpg) can loose a significant amount of brightness range and, hence, detail in the shadows in some panoramas. This also applies to single shots if the scene brightness range is large enough. Obviously, you need to be fairly skilled at raw processing to make this work.
--
Leon
http://homepage.mac.com/leonwittwer/landscapes.htm

Provided that the image fits in between the 'goal posts' of the histogram, it IS correctly exposed. And it WILL have less noise, and better shadow detail.

I've done it for years, with zero ill consequences, such as the nasty shadow noise you'll get in a left-biased histo.
Ken

--



http://www.ahomls.com/gallery.htm
Voted Best of the City 2004 by Cincinnati Magazine
I don't believe in fate, but I do believe in f/8!

Why is ensuring maximum color fidelity and minimal noise classed as 'bad technique'?

I don't expect a logical answer here.

Just keep in mind that digital has changed the rules of exposure somewhat. 'Correct' exposure no longer means what it did in the film world. Exposing to the right and pulling the histogram back left after the fact can often lead to a better result than shooting to the left in the first place.

--
http://www.photoalberta.ca

Leon Wittwer said:

When I am shooting panorama shots that span a large brightness
range, capturing the brightest shot to the right allows the most
brightness range to be captured above the noise level. Doing a
normal exposure (which I did when shooting jpg) can loose a
significant amount of brightness range and, hence, detail in the
shadows in some panoramas.

Click to expand...

This is the sort of example I meant when I said that you should be dealing with it more professionally.

1. You shouldn't be shooting .jpg for a start. There is no excuse not to shoot RAW.

2. The problems with most panaramics like this, can be completely solved by the use of ND filters. Not only will the end result will be so much better, you'll have a lot less post-processing to do.

Nice job of explaining how to blend shots in PSCS for enhanced dynamic range.

But... if you were able to recover detail in the 'blown' shirt by changing the exposure/brightness level, then you didn't blow out the highlights (by definition). 'Blown highlights' means when the pixel value is maxed for all three primary colors, e.g., an RGB value of 255-255-255 in an 8-bit system. Because there is absolute white at this location you can never 'underexpose' it and get information back. Instead, blending the two variations of the same image is the digital version of "dodging and burning" and is really a simpler way of creating complicated layers for each EV layer and blending them.

There's a lot more going on in these raw files than many of us appreciate. I think that the raw pixel data is combined with recorded EV level and thus what SEEMS blown-out really isn't... especially when the blown area is more than one pixel wide and adjacent pixels can be used to recalculate image data (Bayer sensor-equipped cameras). IOW, 'blown' areas really aren't blown because there is some pixel data there that is adjusted by the recorded EV.

Michael Reichmann's advice on "shooting to the right" is spot-on for getting the most out of a digital camera's dynamic range, and those who don't understand this don't understand how digital cameras work and how a little judicious post-processing, akin to darkroom work in the wet film era, is another tool in the arsenal of the skilled photographer that separates him from the digicam-wielding point-and-shooter. It works for basically the same reason that slightly overexposing chromogenic film works.

Quite frankly, the need to deliberately do this with many subjects has been eliminated by the superb matrix metering capabilities of many of the best dSLRs, but when conditions are especially tricky keeping "shoot to the right" in mind, and exposing for the shadows, can make or break a great shot.

I shoot with a Sigma SD10 with a Foveon sensor and it does display all three RGB channels on the histogram. I also "shoot to the right" but only far enough to just-about-max-out a channel (and spot-check each exposure/histogram via the camera's LCD display). I use the Sigma-supplied raw processing software to get the histograms where they need to be and then export to TIFF in 16-bit and finish up using Picture Window Pro. The dynamic range is ample and I am very happy with the results I get on most subjects. For really high-dynamic range subjects that can't be effectively captured with one photo, e.g., an inside shot of a stained glass window where I also wanted interior detail to show, I'd shoot two exposures and blend them during post-processing. Of course, my multi-channel histograms show me when this is the case.

The neat thing about all of this, though, is that a skilled photographer can get more dynamic range out of a quality digital camera than with any film camera. What a great time to be a photographer!

OBVIOUSLY the dynamic range wasn't too wide for your DSLR.

Maybe it was wider than the camera's built-in processing would normally accomodate for writing to a JPEG file.

And maybe it was wider than the basic RAW conversion would accomodate with the default options. Although that's not suprising, since the basic RAW conversion for most systems simply does more or less the same thing as the camera's built-in processing.

But obviously, if you were able to get the information at all, the dynamic range wasn't too much for the camera itself to handle.

More than anything, this just points out the fact that few, if any, RAW converter programs (including ACR) have any facility to automatically maintain the maximum dynamic range of each individual image.

riddell said:

riddell said:

When I am shooting panorama shots that span a large brightness
range, capturing the brightest shot to the right allows the most
brightness range to be captured above the noise level. Doing a
normal exposure (which I did when shooting jpg) can loose a
significant amount of brightness range and, hence, detail in the
shadows in some panoramas.

Click to expand...

This is the sort of example I meant when I said that you should be
dealing with it more professionally.

1. You shouldn't be shooting .jpg for a start. There is no excuse
not to shoot RAW.

Click to expand...

In my early days, RAW was not an option. I have been using RAW for several years on my panos.

riddell said:

2. The problems with most panaramics like this, can be completely
solved by the use of ND filters. Not only will the end result will
be so much better, you'll have a lot less post-processing to do.

Click to expand...

No, ND filters do not do the job as the variations in brightnesses can be almost anywhere in the panorama including very local regions. In addition, such filters can cause bad problems when doing stitching. Have you shot any panoramas?

--
Leon
http://homepage.mac.com/leonwittwer/landscapes.htm

Leon Wittwer said:

Leon Wittwer said:

2. The problems with most panaramics like this, can be completely
solved by the use of ND filters. Not only will the end result will
be so much better, you'll have a lot less post-processing to do.

Click to expand...

No, ND filters do not do the job as the variations in brightnesses
can be almost anywhere in the panorama including very local
regions. In addition, such filters can cause bad problems when
doing stitching. Have you shot any panoramas?

Click to expand...

Yes loads. I use a 1ds and have made some huge panaramas for businesses. One recently was 14' across without any interpolation. I used an ND grad, and had no problems stitiching together all the multiple images.

of these 2 alternatives (capture/photoshop) nikon capture is 1st to disappear. ACR is much better RAW converter imo. Hold hoghlights better and integrates seamlessly to Photosho (obviously). I buoght into Nikon system in 2003 mainly BECAUSE of capture and sadly 3 yrs later went to canon camp for the very same reason.
Last capture version just was not usable for me.

juha
--
http://www.pbase.com/juuso