Experiments with Nikon D600 exposure - seems too aggressive

Started Sep 21, 2012 | Discussions
fromero Contributing Member • Posts: 907
Re: The secret is...

Sorry, but I don't understand why spot metering on the white is necessary following your rules. I am interested, this is why I ask.

eNo wrote:

  • Spot meter on what you want to be middle gray and center-meter, or

  • Spot meter on what you want to be white and open up 2 stops from center-meter

Amazing how "correct" exposures are then.

Rgds
--
Francisco Romero

Mako2011
Mako2011 Forum Pro • Posts: 23,399
I think

fromero wrote:

Sorry, but I don't understand why spot metering on the white is necessary following your rules. I am interested, this is why I ask.

eNo wrote:

  • Spot meter on what you want to be middle gray and center-meter, or

  • Spot meter on what you want to be white and open up 2 stops from center-meter

White is Zone 7. Meter on it to protect the details there so they are not blown by telling the camera that is zone 5. Then open up 2 stops which puts the metered white exactly at zone 7. The detail is protected and everything in the scene is put in it's proper place as white is indeed always exactly zone 7, and two stops open from zone 5 is exactly zone 7 also. Meter white to be zone 5...open up two stops insures it's put where it belongs....I think that's how it works.

creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: The secret is...

Hi eNo. I know how to spot meter and I understand its reliability. However, it is usually not feasible for my kind of subjects (moving around a lot, rapidly changing light conditions from outdoor sunlight to outdoor shadow to indoor, etc. etc.).

BTW thanks for your D7000 post (which Mako2011 showed me). Very informative. But don't you think it's funny that the D700's exposure seems more consistent and predictable?

Best regards,
Mic
--
http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
Mako2011
Mako2011 Forum Pro • Posts: 23,399
I would loose

creaDVty wrote:

Hi eNo. I know how to spot meter and I understand its reliability. However, it is usually not feasible for my kind of subjects (moving around a lot, rapidly changing light conditions from outdoor sunlight to outdoor shadow to indoor, etc. etc.).

Matrix meter does a pretty good job most of the time. When the scene starts to fall outside average and your intentions change from what the programers programed though...matrix becomes less desired. You also use AF-Area mode "auto" so have less control over focus point bias as it may use multiple points.

Those who use spot meter and fine control of focus point often get very fast at getting accurate/consistent exposure (regards their intention for the scene). Combine that with fine control regards AE-L and and/or AF-on and speed and accuracy regards setting exposure can indeed exceed what matrix alone is capable of in difficult scenes. Your black car example is one.

I would not bet my house in a contest of good and fast exposure with me using matrix and AF-Area mode "Auto" against eNo doing his thing. I would loose

creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: 2nd set of experiments with D600 exposure

Hi Peter. Congrats on your baby. Is s/he walking now? The fun starts there... hehehe

As for your suggestions on finding a zone v subject, they are helpful, and yes I get how a camera can't really know whether a subject is white or black or gray. I'm not saying that the D600 can't get an exposure right. Of course, with spot metering, any camera with a functional light meter should get the job done. All I'm hoping is that with the D600's technology, it's a little better about making an educated guess based on the relationship of tones in the scene. And I guess it does a fine job for the subject but to my dismay doesn't seem to try to fit the scene on the sensor's range.

As for center-weighted metering, I had no problem about the subject being correctly exposed - the D600's matrix exposure does a fine job there. My problem was that the rest of the scene - background, etc. included many blown highlights. If I used center-weighted metering I would think that the problem would be aggravated. (In fact, my s5 has the opposite tendency from the D600 - its meter is extremely conservative and when there's a highlight anywhere in the scene it will underexpose a lot which is not good for the S5 because its range is in the highlights. So what I do to overcome that is use center-weighted metering.)

Nonetheless, I'll give center-weighted a shot with the D600. Haven't tried it on the D600.

Best regards,
Mic
--
http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: I think

Well, for me, I might or might not meter that white as zone 7. It depends on the rest of the scene. If the rest of the scene is made of zones 5 through 9, then metering a zone 7 white as zone 7 could blow out zone 8 and 9 objects. So that's one situation where I would meter the zone 7 (or even zone 8) as zone 5 (for a digital camera). Then I would bring up the exposure in post.

-- hide signature --

http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
Yves P. Forum Pro • Posts: 18,646
Actually ...

Mine is spot on ...

creaDVty wrote:

I did some experiments to test whether the Nikon D600 has a tendency to overexpose or underexpose. Based on the results it appears the D600's exposure is far too aggressive for me, even with ADL.

Here is the blogpost I did on this:

http://betterfamilyphotos.blogspot.com/2012/09/experiments-on-nikon-d600-exposure.html

If there's something wrong with the way I did the tests or what pls let me know! TIA.

Best regards,
Mic
--
http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

-- hide signature --

Yves P. Photographe
Share the Knowledge

NPS Member
PBASE Supporter

Some pictures I like:
http://www.pbase.com/yp8/

 Yves P.'s gear list:Yves P.'s gear list
Fujifilm X-T2
Mako2011
Mako2011 Forum Pro • Posts: 23,399
detail

creaDVty wrote:

All I'm hoping is that with the D600's technology, it's a little better about making an educated guess based on the relationship of tones in the scene.

It is an improvement over last gen in a big way

As for center-weighted metering, I had no problem about the subject being correctly exposed - the D600's matrix exposure does a fine job there. My problem was that the rest of the scene - background, etc. included many blown highlights.

I would note that when you went back and actually examined the raw data...it was much less the case as detail was indeed very recoverable. Had you shot with a Uni-WB picture control it may have been readily evident from the beginning, but I find Uni-WB very impractical. It's really only blown if you can't get to the detail

Mako2011
Mako2011 Forum Pro • Posts: 23,399
Re: I think

creaDVty wrote:

Well, for me, I might or might not meter that white as zone 7. It depends on the rest of the scene. If the rest of the scene is made of zones 5 through 9, then metering a zone 7 white as zone 7 could blow out zone 8 and 9 objects. So that's one situation where I would meter the zone 7 (or even zone 8) as zone 5 (for a digital camera). Then I would bring up the exposure in post.

Very true and another reason why getting fast and accurate with spot metering might have an advantage over matrix in that situation. Very easy to make a zone 8 a zone 7 with spot metering...harder to tweak matrix at times as it requires review.

creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: nice Job

Of note, the three examples using the wall, the FOV of the AF array is actually predominately very close to Zone 5 in all 3.

If you are basing that on the image on the screen I think it might not be accurate because I was using a low-contrast neutral picture style, so the scene's tones are artificially compressed.

In test shot one and test shot two of your son on the playground apparatus...note that the FOV of both AF arrays is very close to zone five in both (red is zone five most often)

Yes that was my intention (that's why I picked those samples, even though the shots sucked) and indeed the subject was rendered correctly. The d600 does a pretty good job of that. It's the rest of the scene that I'm worried about. I mean if the sensor does have the range to cover the whole scene yet the camera insists on a metering that is correct for the subject but pushes the brighter tones outside that range, that is something I need to be aware of.

You hit upon what we know... the D600 like the D7000, seems geared towards shooting RAW and not like the D5100 which seem geared towards shooting jpeg (in default setup).

Well I think from the example below that (my son in the living room), the D600 is not really doing ETTR. It just seems that way because of the scene's highlights and the D600's seemingly wide highlight range - see the comment from mshafik who articulated it better than me. My guess is that the D600 seems to be trying to get the subject exposed correctly, without much regard for highlights.
--
http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: Actually ...

I think it depends on what you consider the correct exposure. If you mean that the subject is correctly exposed, then yes the D600 matrix metering is pretty good. If you mean that the exposure maximizes the information available for postprocessing by avoiding blown highlights, then I don't think the D600 matrix metering does that.

(If your D600 does protect highlights I would like to know that.)

-- hide signature --

http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
Mako2011
Mako2011 Forum Pro • Posts: 23,399
Re: nice Job

creaDVty wrote:

Of note, the three examples using the wall, the FOV of the AF array is actually predominately very close to Zone 5 in all 3.

If you are basing that on the image on the screen I think it might not be accurate because I was using a low-contrast neutral picture style, so the scene's tones are artificially compressed.

Possible but the histograms and the blinkies are based on the scene as rendered given the picture control applied in ViewNX. If your picture being displayed is with Neutral PC then the CS6 check should be close.....if I understand it correctly and that's a big if.

In test shot one and test shot two of your son on the playground apparatus...note that the FOV of both AF arrays is very close to zone five in both (red is zone five most often)

Yes that was my intention (that's why I picked those samples, even though the shots sucked) and indeed the subject was rendered correctly. The d600 does a pretty good job of that. It's the rest of the scene that I'm worried about. I mean if the sensor does have the range to cover the whole scene yet the camera insists on a metering that is correct for the subject but pushes the brighter tones outside that range, that is something I need to be aware of.

I would look at the raw data before concluding the highlights are truly blown....Also, I would note that matrix metering may not be trying to keep things inside the cameras DR as a priority. That's an assumption I think not true. You can look at scenes with nothing below zone 5 and get a feel for it. The programmers intent with many of the 300,000 plus scenes is not intuitively obvious to me...but it does get it pretty darn right very often.

You hit upon what we know... the D600 like the D7000, seems geared towards shooting RAW and not like the D5100 which seem geared towards shooting jpeg (in default setup).

Well I think from the example below that (my son in the living room), the D600 is not really doing ETTR.

Correct.

It just seems that way because of the scene's highlights and the D600's seemingly wide highlight range - see the comment from mshafik who articulated it better than me. My guess is that the D600 seems to be trying to get the subject exposed correctly, without much regard for highlights.

Scene dependent....but no, I don't think they had a strict ETTR mentality when designing Matrix algorithms. Surprising though how many times it does seem to protect the highlight detail (when examing the RAW).

Thanks for your time and work.

Mako2011
Mako2011 Forum Pro • Posts: 23,399
It did

creaDVty wrote:

I think it depends on what you consider the correct exposure. If you mean that the subject is correctly exposed, then yes the D600 matrix metering is pretty good. If you mean that the exposure maximizes the information available for postprocessing by avoiding blown highlights, then I don't think the D600 matrix metering does that.

Not necessarily designed to

(If your D600 does protect highlights I would like to know that.)

When you went back and examined just how much detail could be pulled out of the raw files in your examples...you metering did exactly that in many ways.

" ....But what I found is that the clipping display is not accurate and that there is indeed more recoverable information than is shown by the clipping display.....So, after all, the D600 wasn't losing anywhere near as much info as I previously thought. Indeed, it was great that the camera appeared to be maximizing the amount of light, short of blowing relevant highlights, i.e., exposing to the right. This could be awesome for minimizing noise"

-- hide signature --

http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: nice Job

Thanks. As for matrix metering, I agree it doesn't seem to fit the DR within the sensor's range and neither does it do ETTR. I'm beginning to wonder whether, if my goal is to maximize the info available for postprocessing, perhaps a better approach would be to use average metering (one of the options for center-weighted).

-- hide signature --

http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
creaDVty
OP creaDVty Senior Member • Posts: 1,268
Re: It did

Not necessarily designed to

I agree it's not designed for that unfortunately.

(If your D600 does protect highlights I would like to know that.)

When you went back and examined just how much detail could be pulled out of the raw files in your examples...you metering did exactly that in many ways.

For now I don't agree. From what I've seen the D600 did not 'go out of its way' to protect the highlights. The fact that the highlights were not lost appears to be a testament to the sensor's range, an "accident," and not really about the meter's avoidance of blown highlights. At least it seems that way based on what I've seen so far. Maybe I'm wrong, in which case, so much the better!

-- hide signature --

http://www.BetterFamilyPhotos.com

TTL Flash Tutorial and over 400+ other articles for taking better candid and family photos.
http://betterfamilyphotos.blogspot.com/p/better-family-photos-index.html

 creaDVty's gear list:creaDVty's gear list
Fujifilm FinePix F31fd Panasonic Lumix DMC-LX5 Fujifilm FinePix Real 3D W3 Sony Cyber-shot DSC-RX10 Olympus Stylus 1 +29 more
Mako2011
Mako2011 Forum Pro • Posts: 23,399
All modes

creaDVty wrote:

Thanks. As for matrix metering, I agree it doesn't seem to fit the DR within the sensor's range and neither does it do ETTR. I'm beginning to wonder whether, if my goal is to maximize the info available for postprocessing, perhaps a better approach would be to use average metering (one of the options for center-weighted).

Very tricky. There are those that do indeed practice ETTR with every scene as the DR range of current gen sensors has gotten so good. Problem with that is the noise in the shadows when pulled and you get some starting results that look just plane wrong as ETTR is just not always the best solution. I'm starting to think that the "expose for what your really want the first time" might be a better approach or a "hey, I'm going to do it this way today cause I want to" is better.

I'm going to learn how it all really works and then expose the way that makes me happy at that particular moment Striving to be competent in all modes. Good Luck

Mako2011
Mako2011 Forum Pro • Posts: 23,399
Keep it

creaDVty wrote:

Not necessarily designed to

I agree it's not designed for that unfortunately.

(If your D600 does protect highlights I would like to know that.)

When you went back and examined just how much detail could be pulled out of the raw files in your examples...you metering did exactly that in many ways.

For now I don't agree. From what I've seen the D600 did not 'go out of its way' to protect the highlights. The fact that the highlights were not lost appears to be a testament to the sensor's range, an "accident," and not really about the meter's avoidance of blown highlights. At least it seems that way based on what I've seen so far. Maybe I'm wrong, in which case, so much the better!

Keep that open mind....I've noticed your blog has already greatly improved because of it.

Jack Hogan Veteran Member • Posts: 6,154
'White' Spot Exposure mode: 2012 and yet nowhere to be found.

fromero wrote:

Sorry, but I don't understand why spot metering on the white is necessary following your rules. I am interested, this is why I ask.

eNo wrote:

  • Spot meter on what you want to be white and open up 2 stops from center-meter

Most modern DSLR meters (including Nikon and Canon's) are designed to place average scene luminous exposure half way between perceived white and black - with an additional half stop as a highlight safety guard. Overall that's 3 raw stops below brightest white (saturation).

The dumb meter does not know what the luminous exposure under the 'spot' represents (snow on a sunny day or a black cat in a coalmine), it just produces camera settings for shutter speed and aperture so that whatever is under the 'spot' will appear three raw stops down from whitest white. That works pretty well if the subject is a mid-gray card. Not so well in the other two cases above

So, if the spot meter was placed on what we want to be the brightest object in the capture (say the whitest cloud) and leave camera settings as set by the meter, the whole image will be exposed so that the cloud will appear 3 stops below brightest white (saturation): it will look gray. But we know (because we purposedly spot metered the brightest object in the scene) that the meter is 3 stops off, so we increase exposure by 2+ stops (why not 3? long story) et voilà, the cloud is now white again and the rest of the capture is properly exposed relative to it.

As a landscape am-photog it's ridiculous the number of times I do that on a single outing. What I'd like to know is why in 2012 we do not have a metering mode that will do it for us : spot meter on something, hit the 'brightest/white' button, and have the camera set exposure so that what's under the spot shows up near the top of the histogram.

Cheers,
Jack

Jack Hogan Veteran Member • Posts: 6,154
ETTR

creaDVty wrote:

if my goal is to maximize the info available for postprocessing, perhaps a better approach would be to use average metering (one of the options for center-weighted).

Nein, that's also my goal and that's not a good way to do it, imo. Spot meter the brightest object and increase exposure by 2.3EV or so is my preference. But it's not easy to do in a dynamic situation. In that case you may want to consider matrix+auto ADL, one of Mako's specialties

Cheers,
Jack

Mako2011
Mako2011 Forum Pro • Posts: 23,399
Re: ETTR

Jack Hogan wrote:

creaDVty wrote:

if my goal is to maximize the info available for postprocessing, perhaps a better approach would be to use average metering (one of the options for center-weighted).

Nein, that's also my goal and that's not a good way to do it, imo. Spot meter the brightest object and increase exposure by 2.3EV or so is my preference. But it's not easy to do in a dynamic situation. In that case you may want to consider matrix+auto ADL, one of Mako's specialties

Yep...ADL in Auto can indeed be a useful tool! I'm learning now to be more effective with spot but do not relinquish my hold on ADL Been experiminting with spot just as you suggest. Amazing at how well it can work and what it does in forcing your to "see" the scene in terms of exposure with more intimacy. A great learning experience.

Keyboard shortcuts:
FForum MMy threads