Ad
Ad
Ad
Pages: « 1 [2]   Bottom of Page
Print
Author Topic: Setting exposure  (Read 10795 times)
DarkPenguin
Guest
« Reply #20 on: November 07, 2006, 08:23:51 PM »
ReplyReply

Expose to the right pushes the highlights as far as possible to the right.  One then pulls them to the left (Um, where the photographer wishes) in post processing.
« Last Edit: November 07, 2006, 08:24:57 PM by DarkPenguin » Logged
John Sheehy
Sr. Member
****
Offline Offline

Posts: 838


« Reply #21 on: November 07, 2006, 08:35:47 PM »
ReplyReply

Quote
I have never thought of white balance as part of "exposure." 

That will require some thouht. 
[a href=\"index.php?act=findpost&pid=83950\"][{POST_SNAPBACK}][/a]

WB is only relevant at shooting time if you're shooting JPEGs.  If you're shooting RAW, f-stop, shutter speed, and ISO are the only camera parameters that affect the RAW image.  There are supposed to be some exceptions, though.  The Nikon D2X is alleged to perform white balance while digitizing the RAW data.  I don't know if that's true.
Logged
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #22 on: November 07, 2006, 10:39:10 PM »
ReplyReply

Are you saying that the histogram either luminance or RGB is not one of the RAW data but an approximation of the cameras JPEG interpretation? (on a Canon DSLR)
Marc

PS. I did a quick experiment and on a 5D in auto white balance bracketing mode the displayed image and the histogram followed the changes in white balance but in manual white balance mode setting both extremes 2800K and 10,000K the image and the histograms are the same. My assumption is in manual white balance mode you are looking at the RAW histogram?
Marc
« Last Edit: November 08, 2006, 01:18:30 AM by marcmccalmont » Logged

Marc McCalmont
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #23 on: November 08, 2006, 06:37:03 AM »
ReplyReply

Quote
Are you saying that the histogram either luminance or RGB is not one of the RAW data but an approximation of the cameras JPEG interpretation? (on a Canon DSLR)
Marc

PS. I did a quick experiment and on a 5D in auto white balance bracketing mode the displayed image and the histogram followed the changes in white balance but in manual white balance mode setting both extremes 2800K and 10,000K the image and the histograms are the same. My assumption is in manual white balance mode you are looking at the RAW histogram?
Marc
[{POST_SNAPBACK}][/a]

With both Nikon and digital cameras, white balance is achieved by changing the gain in the red and blue channels so as to bring the color balance to neutral. With the Nikon D200 and daylight exposure, the red channel is multiplied by 1.8 and the blue by 1.4. Here are the multipliers for other Nikon cameras and color temps:

[a href=\"http://www.pochtar.com/NikonWhiteBalanceCoeffs.htm]http://www.pochtar.com/NikonWhiteBalanceCoeffs.htm[/url]

Here is a Macbeth Color checker exposed under daylight and processed in DCRaw with no white balance applied. The balance is decidedly green and the histograms for the channels in a neutral square are shown in Photoshop. The picture appears dark, since it is gamma 1.0. In general, raw histograms with gamma 1 are not that helpful, since everything is squeezed to the left.

http://bjanes.smugmug.com/photos/70235196-O.gif

With Nikon cameras one can get a gamma corrected histogram of the raw data by loading unity white balance into the camera so that the coefficients are all 1.0. I do not know if this is possible with Canons, but John Sheehy suggests that one can achieve a similar effect by taking a custom WB from a magenta piece of paper.
Logged
John Sheehy
Sr. Member
****
Offline Offline

Posts: 838


« Reply #24 on: November 08, 2006, 08:35:55 AM »
ReplyReply

Quote
In general, raw histograms with gamma 1 are not that helpful, since everything is squeezed to the left.
[a href=\"index.php?act=findpost&pid=84116\"][{POST_SNAPBACK}][/a]

I don't see that this would be a big issue.  Under-exposure is much easier to see with a linear histogram.  Most of the detail you need to see to expose to the right is on the right.  However, I'd like to see any kind of RAW RGB histogram, as long as it is clearly ticked for fine exposure estimates and corrections.
Logged
Chris_T
Sr. Member
****
Offline Offline

Posts: 541


« Reply #25 on: November 08, 2006, 09:03:57 AM »
ReplyReply

Quote
Thank you Bill for your thoughtful reply.  I thought exposure was still exposure.  And 'expose to the right" is not a new idea.  It worked very well (and still does) with slide film.  Expose to the right merely shifts the exposure up so the highlights are to the right as far as the photographer wishes.  Quite analogous to variable developing time for b&w negs so the highlights fall where the photographer wishes.  Only difference between film and digital is the time it takes to "see" the results.

But exposure is still just f/stop, shutter speed and ISO.  Nothing really new.
[a href=\"index.php?act=findpost&pid=84049\"][{POST_SNAPBACK}][/a]

You are absolutely correct about ETTR! Instead of introducing this *new* acronym, the writers could have made it a lot simpler by referencing "exposing to retain highlight details on slides". The experienced slide shooters would have no problem understanding it. But that is just too boring for some.

As far as auto everything is concerned, many media and sports pros do rely upon them. Manually exposing and focusing can mean losing a shot.

There are a couple of key differences between film and digital exposure. With digital, the preview and histogram can provide immediate feedback. But of course those who can nail their exposures perfectly every shot would find these useless.

The noise at different digital ISOs vary between different camera sensors. For a particular film type and ISO, the grain, etc. remains the same regardless of which camera is used. Not so in digital.

With digital raw (and HDR), it is now possible to extend the dynamic range with a single shot (or multiple shots with different exposures). This is something very difficult or nearly impossible to achieve in a traditional darkroom.
Logged
howiesmith
Guest
« Reply #26 on: November 08, 2006, 01:21:05 PM »
ReplyReply

Quote
You are absolutely correct about ETTR! Instead of introducing this *new* acronym, the writers could have made it a lot simpler by referencing "exposing to retain highlight details on slides". The experienced slide shooters would have no problem understanding it. But that is just too boring for some.

As far as auto everything is concerned, many media and sports pros do rely upon them. Manually exposing and focusing can mean losing a shot.

There are a couple of key differences between film and digital exposure. With digital, the preview and histogram can provide immediate feedback. But of course those who can nail their exposures perfectly every shot would find these useless.

The noise at different digital ISOs vary between different camera sensors. For a particular film type and ISO, the grain, etc. remains the same regardless of which camera is used. Not so in digital.

With digital raw (and HDR), it is now possible to extend the dynamic range with a single shot (or multiple shots with different exposures). This is something very difficult or nearly impossible to achieve in a traditional darkroom.
[a href=\"index.php?act=findpost&pid=84143\"][{POST_SNAPBACK}][/a]

I sounds like a digital camera determines exposure the same way.  It just uses a new exposure meter with a histogram and an instant picture to look at.

Digial noise is a lot like film grain.  Fast film, more grain (or noise).

Yes, with digital. it is "possible to extend the dynamic range with a single shot (or multiple shots with different exposures)."  I say that is not exposure but processing.  It is possible to extend the dynamic range with film with preflashing.  Not really exposure either.

If I had a new camera with new technology and all, I wouldn't want to say it really worked just like an old fashioned film camera.  And that is a lot easier to say with a straight face if you have no idea how an old fashioned film camera works.

When Adams "invented" the Zone System, he was really just applying an old methodolgy developed (no pun intended) by others.  Variable processing is not Adams'.

And again, I have nothing against autofocus or autoexposure.  But I suspect those "media and sports pros" really understand how their camera works and what it is doing for them.
« Last Edit: November 08, 2006, 01:24:12 PM by howiesmith » Logged
John Sheehy
Sr. Member
****
Offline Offline

Posts: 838


« Reply #27 on: November 08, 2006, 04:38:38 PM »
ReplyReply

Quote
You are absolutely correct about ETTR! Instead of introducing this *new* acronym, the writers could have made it a lot simpler by referencing "exposing to retain highlight details on slides".

No, that's about keeping it enough to the left that you don't blow the highlights.  ETTR is about going to the right as much as is possible, with no regard for reflecting real scene tonality.  There's a difference.  Does slide film record contrast best in its upper stop?  Have the weakest grain after increasing its contrast? Have the best saturation?  Digital *does*.

Quote
The experienced slide shooters would have no problem understanding it. But that is just too boring for some.

Not exactly.  This is different.  Would a slide shooter shoot a dark-grey-on-black scene so that the dark grey is just short of clipping?  That is what ETTR can entail.

Quote
There are a couple of key differences between film and digital exposure. With digital, the preview and histogram can provide immediate feedback. But of course those who can nail their exposures perfectly every shot would find these useless.

How do you know that you're "nailing" your exposure?  I bet lots of people who pat themselves on the back for nailing exposure are wasting a stop of DR, which is most likely what you're doing in a RAW file if the JPEG looks "nailed".
Logged
DarkPenguin
Guest
« Reply #28 on: November 08, 2006, 05:14:17 PM »
ReplyReply

http://www.luminous-landscape.com/tutorial...ose-right.shtml
Logged
howiesmith
Guest
« Reply #29 on: November 08, 2006, 07:49:14 PM »
ReplyReply

Quote
http://www.luminous-landscape.com/tutorial...ose-right.shtml
[a href=\"index.php?act=findpost&pid=84218\"][{POST_SNAPBACK}][/a]

Thanks for the link.  I read the information and I think I have it now.

Film and digital have some common ground and some real differences.  If I understand it right, digital is linear and flim is not.  There is no warning and no forgiveness for overexposing digital.  Film is more forgiving.  Maybe that is why it is easy to get the "correct" exposure.  AS a highlight approaches "blown out" on film, it takes an increasing amount of exposure to cross the line with film.  With digital, the highlights just fall off the end.

It appears to me that digital exposure is still just f/stop, shutter speed and ISO, same as film.  The method for bumping the image to the right is the same, but knowing how far to bump is different, and of course processing is different.

Because film is non-linear, slightly underexposed highlights are also hard to see.  The highligh latitude is more forgiving in both directions.  Underexpose a digital highlight is OK, it just wastes information space,
Logged
Eric Myrvaagnes
Sr. Member
****
Offline Offline

Posts: 7790



WWW
« Reply #30 on: November 08, 2006, 08:09:32 PM »
ReplyReply

Quote
Thanks for the link.  I read the information and I think I have it now.

Film and digital have some common ground and some real differences.  If I understand it right, digital is linear and flim is not.  There is no warning and no forgiveness for overexposing digital.  Film is more forgiving.  Maybe that is why it is easy to get the "correct" exposure.  AS a highlight approaches "blown out" on film, it takes an increasing amount of exposure to cross the line with film.  With digital, the highlights just fall off the end.

It appears to me that digital exposure is still just f/stop, shutter speed and ISO, same as film.  The method for bumping the image to the right is the same, but knowing how far to bump is different, and of course processing is different.

Because film is non-linear, slightly underexposed highlights are also hard to see.  The highligh latitude is more forgiving in both directions.  Underexpose a digital highlight is OK, it just wastes information space,
[a href=\"index.php?act=findpost&pid=84239\"][{POST_SNAPBACK}][/a]
I think that is an excellent summary, Howie. Well expressed.

Eric
Logged

-Eric Myrvaagnes

http://myrvaagnes.com  Visit my website. New images each season.
dbell
Full Member
***
Offline Offline

Posts: 131


« Reply #31 on: November 09, 2006, 08:53:54 AM »
ReplyReply

Quote
I was trout fishing near Bishop one day when a guy with a thousand dollar fly rod asked me what I was using for bait.  I said "worms."  While he was explaining how n one can catch trout on worms, I had to ask him to move over while I landed dinner.
[a href=\"index.php?act=findpost&pid=83903\"][{POST_SNAPBACK}][/a]



I don't mean to imply that there's anything wrong with using modern equipment, only that cameras and lenses are just tools for doing a job. If the old tools are doing the job, I'd go right on using them. There are certain things about digital capture, postprocessing and printing that I  find compelling enough to convince me to work that way.

That said, I enjoy the experience of shooting with mechanical cameras much more than using my automated cameras. I just make more good images, faster, with the modern setup.

If you do your best work with a Nikon F or Leica M (or 4x5 field camera), why mess with anything else?


--
Daniel Bell
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2756



« Reply #32 on: November 09, 2006, 10:06:40 AM »
ReplyReply

Quote
I think that is an excellent summary, Howie. Well expressed.

Eric
[a href=\"index.php?act=findpost&pid=84243\"][{POST_SNAPBACK}][/a]

Yes, after reading a few posts and MR's ETTR essay Howie has come up to speed very quickly with digital. That is because he understands exposure. The same principles involved in 4x5 exposure with black and white negative film or color transparencies also apply to digital: one places the exposure to optimize the characteristics of the medium.

Bill
Logged
Pages: « 1 [2]   Top of Page
Print
Jump to:  

Ad
Ad
Ad