Pls show examples where increasing brightness in post better than increasing ISO.

Started Oct 1, 2013 | Discussions
Horshack
Veteran MemberPosts: 4,222
Like?
Re: Excellent example!
In reply to Great Bustard, Oct 3, 2013

Great Bustard wrote:

Horshack wrote:

On all the cameras I have there actually is an IQ compromise when shooting at base ISO. For example on my D800 there is more noise and shadow tinting when using ISO 100 pushed to ISO 3200/6400 than when shooting at ISO 800 and pushed to the same output brightness.

Interesting. According to sensorgen's figures (derived from DxOMark data), that should not be the case. That said, I've heard that some (most?) RAW converters may make use of the ISO setting information in the conversion. For example, an ISO 100 file will be converted differently than an ISO 800 file, based on assumptions about the scene as a function of the selected ISO. Perhaps someone can confirm or deny this.

Yeah, in my testing both the quantity and quality of noise is worse for a 100->3200 vs an 800->3200. That's not to say 100->3200 is horrible but for certain scenes I've found a definitive disadvantage to using 100. The worst is the shadow tinting/hue shifts, which on the D800 aren't so bad but are worse on the D600 and downright horrible on the D7100. The best camera/sensor in this regard is the 16MP Exmor in the D7000/D5100.

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,222
Like?
Another example (PP time savings)
In reply to Horshack, Oct 3, 2013

I posted this on a DR discussion @ Canon Rumors a month back:

I would agree that nearly every High DR scene can be captured using techniques that don't require a High DR sensor. But one benefit of such sensor is workflow time savings. Here is a recent example where I shot a home interior for a friend for his real estate listing (using a D800). I wanted maximum IQ so I used two-shot blends for all the shots which had windows, to exhibit the woodsy setting outside his home. In this example it took me 20 minutes to manually blend the image, which I did in PS using layers and masks around the windows. For kicks I also performed the same exposure adjustment using a single image, which took me about 3 minutes. The latter has more noise than the two-shot blend but it's still perfectly usable even at the native 36MP resolution...and much more so at the resolutions the images were displayed at for the MLS listing. If you multiply this by 10 photos then the time savings can be significant...compared to either blends or interior strobe set ups.

Orig lower exposure image

Two-shot blend

One-shot HDR/shadow push

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Excellent example!
In reply to Horshack, Oct 3, 2013

Horshack wrote:

Great Bustard wrote:

Horshack wrote:

On all the cameras I have there actually is an IQ compromise when shooting at base ISO. For example on my D800 there is more noise and shadow tinting when using ISO 100 pushed to ISO 3200/6400 than when shooting at ISO 800 and pushed to the same output brightness.

Interesting. According to sensorgen's figures (derived from DxOMark data), that should not be the case. That said, I've heard that some (most?) RAW converters may make use of the ISO setting information in the conversion. For example, an ISO 100 file will be converted differently than an ISO 800 file, based on assumptions about the scene as a function of the selected ISO. Perhaps someone can confirm or deny this.

Yeah, in my testing both the quantity and quality of noise is worse for a 100->3200 vs an 800->3200. That's not to say 100->3200 is horrible but for certain scenes I've found a definitive disadvantage to using 100. The worst is the shadow tinting/hue shifts, which on the D800 aren't so bad but are worse on the D600 and downright horrible on the D7100. The best camera/sensor in this regard is the 16MP Exmor in the D7000/D5100.

What raw converter are you using? and is the profile twisted?

-- hide signature --

gollywop

Reply   Reply with quote   Complain
moving_comfort
Veteran MemberPosts: 7,880Gear list
Like?
Re: Excellent example!
In reply to gollywop, Oct 3, 2013

gollywop wrote:

Horshack wrote:

Great Bustard wrote:

Horshack wrote:

On all the cameras I have there actually is an IQ compromise when shooting at base ISO. For example on my D800 there is more noise and shadow tinting when using ISO 100 pushed to ISO 3200/6400 than when shooting at ISO 800 and pushed to the same output brightness.

Interesting. According to sensorgen's figures (derived from DxOMark data), that should not be the case. That said, I've heard that some (most?) RAW converters may make use of the ISO setting information in the conversion. For example, an ISO 100 file will be converted differently than an ISO 800 file, based on assumptions about the scene as a function of the selected ISO. Perhaps someone can confirm or deny this.

Yeah, in my testing both the quantity and quality of noise is worse for a 100->3200 vs an 800->3200. That's not to say 100->3200 is horrible but for certain scenes I've found a definitive disadvantage to using 100. The worst is the shadow tinting/hue shifts, which on the D800 aren't so bad but are worse on the D600 and downright horrible on the D7100. The best camera/sensor in this regard is the 16MP Exmor in the D7000/D5100.

What raw converter are you using? and is the profile twisted?

I've noticed the same thing with my D800, using LR 4.x/ACR. I actually started trying it with several other cameras after I read Bob's posts about the subject a few years ago. I chalked it up to my not using the right raw converter, but then never followed up with trying to find the right one for demonstrating this.

-- hide signature --

Here are a few of my favorite things...
---> http://www.flickr.com/photos/95095968@N00/sets/72157626171532197/

 moving_comfort's gear list:moving_comfort's gear list
Pentax K20D Nikon D800 Nikon AF-S DX Nikkor 35mm f/1.8G Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF Nikkor 20mm f/2.8D +10 more
Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Excellent example!
In reply to moving_comfort, Oct 3, 2013

moving_comfort wrote:

gollywop wrote:

Horshack wrote:

Great Bustard wrote:

Horshack wrote:

On all the cameras I have there actually is an IQ compromise when shooting at base ISO. For example on my D800 there is more noise and shadow tinting when using ISO 100 pushed to ISO 3200/6400 than when shooting at ISO 800 and pushed to the same output brightness.

Interesting. According to sensorgen's figures (derived from DxOMark data), that should not be the case. That said, I've heard that some (most?) RAW converters may make use of the ISO setting information in the conversion. For example, an ISO 100 file will be converted differently than an ISO 800 file, based on assumptions about the scene as a function of the selected ISO. Perhaps someone can confirm or deny this.

Yeah, in my testing both the quantity and quality of noise is worse for a 100->3200 vs an 800->3200. That's not to say 100->3200 is horrible but for certain scenes I've found a definitive disadvantage to using 100. The worst is the shadow tinting/hue shifts, which on the D800 aren't so bad but are worse on the D600 and downright horrible on the D7100. The best camera/sensor in this regard is the 16MP Exmor in the D7000/D5100.

What raw converter are you using? and is the profile twisted?

I've noticed the same thing with my D800, using LR 4.x/ACR. I actually started trying it with several other cameras after I read Bob's posts about the subject a few years ago. I chalked it up to my not using the right raw converter, but then never followed up with trying to find the right one for demonstrating this.

Are you using the Adobe Standard profile for this?

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Horshack
Veteran MemberPosts: 4,222
Like?
Re: Excellent example!
In reply to gollywop, Oct 3, 2013

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Excellent example!
In reply to Horshack, Oct 3, 2013

Horshack wrote:

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Well, your own profile is indeed not likely to be twisted, so that rules out that hypothesis.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,076
Like?
Re: Excellent example !
In reply to gollywop, Oct 3, 2013

gollywop wrote:

Horshack wrote:

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Well, your own profile is indeed not likely to be twisted, so that rules out that hypothesis.

I recall your investigation and finding that the Adobe ACR/LR "Exposure" control precedes what appears to be a (as I understand it, a silent, and a non-defeatable) non-linear transfer-function:

http://www.dpreview.com/forums/post/50786838

See also: http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fphotocrafting.wordpress.com%2F2013%2F01%2F21%2Ffollow-up-on-adobe-camera-raw%2F&act=url

One cannot scale the input to a non-linear transfer-function without changing the resultant output.

DM ...

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Excellent example !
In reply to Detail Man, Oct 3, 2013

Detail Man wrote:

gollywop wrote:

Horshack wrote:

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Well, your own profile is indeed not likely to be twisted, so that rules out that hypothesis.

I recall your investigation and finding that the Adobe ACR/LR "Exposure" control precedes what appears to be a (as I understand it, a silent, and a non-defeatable) non-linear transfer-function:

http://www.dpreview.com/forums/post/50786838

See also: http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fphotocrafting.wordpress.com%2F2013%2F01%2F21%2Ffollow-up-on-adobe-camera-raw%2F&act=url

One cannot scale the input to a non-linear transfer-function without changing the resultant output.

This is definitely true.  That test was done on a gray scale, so it doesn't tell us how that non-linearity may operate differentially on the different color channels.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,076
Like?
Re: Excellent example !
In reply to gollywop, Oct 3, 2013

gollywop wrote:

Detail Man wrote:

gollywop wrote:

Horshack wrote:

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Well, your own profile is indeed not likely to be twisted, so that rules out that hypothesis.

I recall your investigation and finding that the Adobe ACR/LR "Exposure" control precedes what appears to be a (as I understand it, a silent, and a non-defeatable) non-linear transfer-function:

http://www.dpreview.com/forums/post/50786838

See also: http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fphotocrafting.wordpress.com%2F2013%2F01%2F21%2Ffollow-up-on-adobe-camera-raw%2F&act=url

One cannot scale the input to a non-linear transfer-function without changing the resultant output.

This is definitely true. That test was done on a gray scale, so it doesn't tell us how that non-linearity may operate differentially on the different color channels.

Would it not to be reasonable to assume that (if) no hue-twists (by whatever means) are said to exist, (then) we might as a result expect the RGB channel transfer-functions to be the same - or are we somehow discussing an entirely separate way that color-dependence may exist here ? ...

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Excellent example !
In reply to Detail Man, Oct 3, 2013

Detail Man wrote:

gollywop wrote:

Detail Man wrote:

gollywop wrote:

Horshack wrote:

gollywop wrote:

What raw converter are you using? and is the profile twisted?

ACR, with my own profile. It's not a profile hue twist issue but mostly a shadow tinting issue. I can sometimes get acceptable results with ACR's shadow tinting slider but only if the color fidelity isn't very important for the image.

Well, your own profile is indeed not likely to be twisted, so that rules out that hypothesis.

I recall your investigation and finding that the Adobe ACR/LR "Exposure" control precedes what appears to be a (as I understand it, a silent, and a non-defeatable) non-linear transfer-function:

http://www.dpreview.com/forums/post/50786838

See also: http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fphotocrafting.wordpress.com%2F2013%2F01%2F21%2Ffollow-up-on-adobe-camera-raw%2F&act=url

One cannot scale the input to a non-linear transfer-function without changing the resultant output.

This is definitely true. That test was done on a gray scale, so it doesn't tell us how that non-linearity may operate differentially on the different color channels.

Would it not to be reasonable to assume that (if) no hue-twists (by whatever means) are said to exist, (then) we might as a result expect the RGB channel transfer-functions to be the same - or are we somehow discussing an entirely separate way that color-dependence may exist here ? ...

I don't really know.  We do know that the neutral grays (redundant) remain neutral with all the alterations in the "Exposure" slider.  So some sort of balance must remain invariant.  But . . .

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,943Gear list
Like?
Re: Another example (PP time savings)
In reply to Horshack, Oct 3, 2013

Horshack wrote:

I posted this on a DR discussion @ Canon Rumors a month back:

I would agree that nearly every High DR scene can be captured using techniques that don't require a High DR sensor. But one benefit of such sensor is workflow time savings. Here is a recent example where I shot a home interior for a friend for his real estate listing (using a D800). I wanted maximum IQ so I used two-shot blends for all the shots which had windows, to exhibit the woodsy setting outside his home. In this example it took me 20 minutes to manually blend the image, which I did in PS using layers and masks around the windows. For kicks I also performed the same exposure adjustment using a single image, which took me about 3 minutes. The latter has more noise than the two-shot blend but it's still perfectly usable even at the native 36MP resolution...and much more so at the resolutions the images were displayed at for the MLS listing. If you multiply this by 10 photos then the time savings can be significant...compared to either blends or interior strobe set ups.

Orig lower exposure image

Two-shot blend

One-shot HDR/shadow push

I agree wholeheartedly on the time saving. I think it has to do with what is actually happening in the final image.

The two-shot-blend is based on a complex pixel selection and tone mapping process often initially resulting in unnatural looking artifacts and requiring much fine tuning.

The one-shot-push is just simple tone mapping. At worst it is a contrast curve with some highlight recovery, something most of us are quite adept at.

Jack

Reply   Reply with quote   Complain
rubank
Contributing MemberPosts: 619Gear list
Like?
Re: Examples
In reply to gollywop, Oct 3, 2013

gollywop wrote:

rubank wrote:

bobn2 wrote:

rubank wrote:

bobn2 wrote:

rubank wrote:

The simple fact is
that you have to choose: will you sacrifice the highlights or the shadows. With a scene where the DR is beyond what your camera can capture in a single exposure, this choice is of greater importance than what ISO routine you prefer.

Do you want to preserve highlights it is often better to choose underexposure at low ISO, at the cost of shadow colour and noise (esp. banding). If shadows are more important it is often better to choose high ISO, at the cost of highlight rendering.

Heavy shadow lifting commonly leeds to colour shift and banding noise, at least with badly lit scenery where major parts of the image is in deep shadow.
In daylight situations with minor parts of the image in deep shadow, lifting in post is mostly not a problem (unless you have an ISO-full camera).

These are my findings, based on extensive practice and not on science.

The point about knowing the science is that it lets you take out of the loop a lot of the 'extensive practice'.

Extensive practice has a lot of other benefits. You need 10 000 hours of practice (whatever you engage in) to get proficient. Try it.

That's afigure plucked out of the air with no evidence.

Actually not
http://www.wisdomgroup.com/blog/10000-hours-of-practice/

The point is that science provides direction much faster than trial and error.

There is no such contradiction. Most science is based on trial and error.

Well, a lot of engineering is trial and error. And trial and error has indeed helped science. But the major breakthroughs that have led to modern science, special and general relativity and quantum mechanics, are hardly trial and error. These Kuhn-sian revolutions are greatly thought beyond trial.

Even thought-processes can be, and often are, trial and error.
And how would you categorise the search for the Higgs particle...

Likewise, the Bacon-ian notion of the hypothetico-deductive process of science, which is really the heart of scientific progress, is hardly properly described by trial and error.

Beyond that, if you're going to come up with a description of "most science," I think it would have to be that most science is experiments conducted to bear out existing theory. And most of these experiments do bear out the existing theory.

Yes, but in more earthly domains, like electronics, you don´t know if the theory holds up before its tested, i.e. you try it.

Reply   Reply with quote   Complain
rubank
Contributing MemberPosts: 619Gear list
Like?
Re: You might be right, you might be wrong
In reply to Reilly Diefenbach, Oct 3, 2013

Reilly Diefenbach wrote:

Pushing two stops in LR5 from 1600 to 6400, the 6400 is slightly better. Slightly finer grain, no color blotching (D800e.) So yeah, I'll go with the higher ISO up until the spike climbs the wall, thanks anyway, gentlemen.

As for practical results I agree.

But it is very hard to deduct from the actual output what is hardware induced and what is software induced. Or how the camera in reality is programmed to operate.

To get the output that serves me best I go with what looks best to my eyes, others may neglect the eye-balling of actual output in favour of theory. Each to his own.

Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,236
Like?
Protecting important highlights
In reply to Horshack, Oct 3, 2013

Horshack wrote:

The example, pushed 5EV in post. Notice how the image on the television is retained by shooting at ISO 100. If I had shot at ISO 3200 the TV image would have been blown out.

Yes, but like other similar examples, then all it really shows is that it's a bad thing/idea to blow important highlights. That's true if using ETTR at base ISO in good light, and also true if using a higher ISO in low light.

Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,236
Like?
Re: Excellent example!
In reply to Great Bustard, Oct 3, 2013

Great Bustard wrote:

Horshack wrote:

IMO the two biggest benefits of ISOless shooting are dynamic range and not having to worry about image brightness at the time of shooting. For the latter if you're shooting in conditions that would otherwise require elevated ISO you simply set the ISO to base (or to whatever level is optimal for that sensor) and never worry about setting ISO thereafter.

The example, pushed 5EV in post. Notice how the image on the television is retained by shooting at ISO 100. If I had shot at ISO 3200 the TV image would have been blown out.

Oriignal (ISO 100)

Post-processed

Nicely demonstrated!

On all the cameras I have there actually is an IQ compromise when shooting at base ISO. For example on my D800 there is more noise and shadow tinting when using ISO 100 pushed to ISO 3200/6400 than when shooting at ISO 800 and pushed to the same output brightness.

Interesting. According to sensorgen's figures (derived from DxOMark data), that should not be the case. That said, I've heard that some (most?) RAW converters may make use of the ISO setting information in the conversion. For example, an ISO 100 file will be converted differently than an ISO 800 file, based on assumptions about the scene as a function of the selected ISO. Perhaps someone can confirm or deny this.

According to Bill Claff the D800/D800e read noise decreases from app. 4.2 e- at ISO 100 to app. 2.7 e- at ISO 800.

Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,943Gear list
Like?
You are welcome
In reply to Reilly Diefenbach, Oct 3, 2013

Reilly Diefenbach wrote:

Pushing two stops in LR5 from 1600 to 6400, the 6400 is slightly better. Slightly finer grain, no color blotching (D800e.) So yeah, I'll go with the higher ISO up until the spike climbs the wall, thanks anyway, gentlemen.

Going from ISO 1600 to 6400 is all 'software' on the D800s - i.e. the ISO6400 setting on the D800s is really the ISO1600 setting (pushed two stops digitally in-camera). You can see it clearly here:

On the D800e ISO 6400 raw values are really ISO 1600 raw values times four

This means that if anybody is looking at two identical captures by a D800/e at the same exposure, one taken at ISO 1600 and one taken at ISO 6400 - and one sees differences in renderings of the two images once brightness is equalized, the differences are entirely due to differing raw conversion processes - because the Raw information is exactly the same.

So if one saw differences in that situation and one were using LR, the differences would have been introduced by LR applying slightly different recipes of its various parameters, including noise reduction, at ISO 6400 vs 1600.  Anybody not knowing that LR is causing the differences itself (and that therefore one could replicate at will when converting) would leave a couple of stops of potentially useful DR at the scene.

Jack

Reply   Reply with quote   Complain
Jeff
Veteran MemberPosts: 4,470
Like?
Re: Pls show examples where increasing brightness in post better than increasing ISO.
In reply to gollywop, Oct 3, 2013

gollywop wrote:

Jeff wrote:

texinwien wrote:

Jeff wrote:

bobn2 wrote:

ultimitsu wrote:

bobn2 wrote:

So, mostly the metering is checking highlights. The easy technique is to set 2 1/2 or a bit more stops of EC and spot meter off the highlights (if your camera lets you use EC in M). On most camera there are a few ISO settings to be working with, and you know those pretty well anyway.

This sounds quite interesting, can you elaborate a bit on how it works? why 2.5 stops?

Your meter should be calibrated to centre when it gets 18% grey as the nominal exposure (10/ISO lux seconds). So, you need to miscalibrate it to regard 100% as nominal. According to ISO, the ratio of highlight to 18% is 78/10 = 7.8 = 2.96 stops (funny what these discussions can do - I see I miscalculated a long time ago, and have been giving my highlights a bit less than I could - dial in 3 stops). So, if the meter reads highlight and thinks it's 18% grey, it will give 3 stops less exposure than it should - so you correct that with EC, or recalibrating the meter (most will not let you recalibrate that far). In practice if you know how much headroom your camera has, you could dial in more than that.

(Just going back through my working, I realise where the extra half a stop came from, ISO leaves 1/2 stop for specular highlights over white - I must have taken it off somewhere along the line - 3 stops is better)

-- hide signature --

Bob

This is more or less what I've learned to do when shooting in challenging light, but with a small twist.

My E-P5 allows me to adjust the level for shadow and highlight 'blinkies', which I've set at jpeg levels 5 and 250, respectively. In A priority, the rear control wheel adjusts aperture, and the front control wheel adjusts EC.

When taking a shot, exposure is set by keeping an eye on the blinkies while adjusting EC, and while monitoring everything else to be sure I'm not totally killing the shadows. EC will generally fall between -0.5 and -2.5, but the critical issue is watch where the highlights and shadows fall on the live histogram and, via the blinkies, on the image.

The resulting jpeg preview may look off (generally too dark) but this procedure will give me a good raw file for subsequent processing in LR. ISO is a tertiary consideration. I generally try to work at base ISO, but will increase ISO as necessary to keep the viewfinder and jpeg previews usable.

Here are some results http://www.flickr.com/photos/jck_photos/sets/72157635431417147/

It becomes second nature pretty quickly.

I started typing a similar reply, took a break, and here you've gone and done the work, yourself!

The live-view 'blinkies' overexposure indicators are a boon to ETTR raw photogs. Add sufficient controls (i.e. Olympus E-M5, E-P5, E-M1) and you have an ETTR dream machine.

I, too, started out in A mode, mostly, but since Anders W convinced me that there was little advantage to A or S over M when using the blinkies method of ETTR, I have switched, and stay in M 99% of the time.

What comes out are images that are usually within a (<1/3) fraction of a stop from optimal ETTR. As much light as possible collected without blowing any important highlights - consistently.

Since the E-M5 (and E-P5) sensor is not 100% ISO-invariant, I do switch between ISO 200, 400 and 800 (rarely 1600). My sincere hope is that the next generation of OM-D models will be truly ISO-invariant, so I can forget that useless (to me) throwback to ancient history and get on with making properly-exposed photographs - one less useless variable to have to keep in mind.

Couldn't agree more. I've been experimenting a bit with the level settings for the blinkies. 5 and 250 may be unnecessarily conservative, but produce good files. 2 and 253

Anders W is probably right (as usual), but I just like how the controls work in A mode. Back wheel A, front wheel EC, and it's totally second nature. Flick the 2x2 control and then ISO is on the front wheel. I am using M mode more often, but A mode just feels more intuitive to me. Admittedly, this is a personal quirk.

Oly could really own this with a few more tweaks of their UI. They're pretty close now.

With the E-M5 and UniWB, I've found no need to adjust the histogram sensitivity at all. I keep it at 0, 255 and find it works splendidly. I believe Anders found the same thing using autoWB. The typical 1/3EV difference between UniWB and autoWB settings was made up by the fact that I adjust EC (or exposure) to "just under" blinkies, while Anders adjusted to "just showing" blinkies.

I tried Anders' method extensively while traveling this past spring, and I have to agree that it works just as well as the UniWB. I came back with 450 shots (not counting WhiBal readings) and absolutely none of them was problematic for exposure as assessed by RawDigger.

That's not to say I didn't have shots that were incorrect, but I was able to determine that situation immediately with the post-shot blinkies and retake as needed.

Many thanks. I'll be opening up the blinkie window.

I'm also intrigued by your comments regarding autoWB vs uniWB.  Normally I use autoWB unless I'm in a tricky, bad lighting situation where I can set up a custom WB.  What would be the advantages of uniWB?

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Examples
In reply to rubank, Oct 3, 2013

rubank wrote:

gollywop wrote:

rubank wrote:

bobn2 wrote:

rubank wrote:

bobn2 wrote:

rubank wrote:

The simple fact is
that you have to choose: will you sacrifice the highlights or the shadows. With a scene where the DR is beyond what your camera can capture in a single exposure, this choice is of greater importance than what ISO routine you prefer.

Do you want to preserve highlights it is often better to choose underexposure at low ISO, at the cost of shadow colour and noise (esp. banding). If shadows are more important it is often better to choose high ISO, at the cost of highlight rendering.

Heavy shadow lifting commonly leeds to colour shift and banding noise, at least with badly lit scenery where major parts of the image is in deep shadow.
In daylight situations with minor parts of the image in deep shadow, lifting in post is mostly not a problem (unless you have an ISO-full camera).

These are my findings, based on extensive practice and not on science.

The point about knowing the science is that it lets you take out of the loop a lot of the 'extensive practice'.

Extensive practice has a lot of other benefits. You need 10 000 hours of practice (whatever you engage in) to get proficient. Try it.

That's afigure plucked out of the air with no evidence.

Actually not
http://www.wisdomgroup.com/blog/10000-hours-of-practice/

The point is that science provides direction much faster than trial and error.

There is no such contradiction. Most science is based on trial and error.

Well, a lot of engineering is trial and error. And trial and error has indeed helped science. But the major breakthroughs that have led to modern science, special and general relativity and quantum mechanics, are hardly trial and error. These Kuhn-sian revolutions are greatly thought beyond trial.

Even thought-processes can be, and often are, trial and error.
And how would you categorise the search for the Higgs particle...

Difficult.  But, seriously, how would you characterize the "knowledge" that the Higgs particle was even worth searching for?

Likewise, the Bacon-ian notion of the hypothetico-deductive process of science, which is really the heart of scientific progress, is hardly properly described by trial and error.

Beyond that, if you're going to come up with a description of "most science," I think it would have to be that most science is experiments conducted to bear out existing theory. And most of these experiments do bear out the existing theory.

Yes, but in more earthly domains, like electronics, you don´t know if the theory holds up before its tested, i.e. you try it.

Well, of course.  That's what the hypethetico-deductive method is all about.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,681
Like?
Re: Pls show examples where increasing brightness in post better than increasing ISO.
In reply to Jeff, Oct 3, 2013

Jeff wrote:

gollywop wrote:

Jeff wrote:

texinwien wrote:

Jeff wrote:

bobn2 wrote:

ultimitsu wrote:

bobn2 wrote:

So, mostly the metering is checking highlights. The easy technique is to set 2 1/2 or a bit more stops of EC and spot meter off the highlights (if your camera lets you use EC in M). On most camera there are a few ISO settings to be working with, and you know those pretty well anyway.

This sounds quite interesting, can you elaborate a bit on how it works? why 2.5 stops?

Your meter should be calibrated to centre when it gets 18% grey as the nominal exposure (10/ISO lux seconds). So, you need to miscalibrate it to regard 100% as nominal. According to ISO, the ratio of highlight to 18% is 78/10 = 7.8 = 2.96 stops (funny what these discussions can do - I see I miscalculated a long time ago, and have been giving my highlights a bit less than I could - dial in 3 stops). So, if the meter reads highlight and thinks it's 18% grey, it will give 3 stops less exposure than it should - so you correct that with EC, or recalibrating the meter (most will not let you recalibrate that far). In practice if you know how much headroom your camera has, you could dial in more than that.

(Just going back through my working, I realise where the extra half a stop came from, ISO leaves 1/2 stop for specular highlights over white - I must have taken it off somewhere along the line - 3 stops is better)

-- hide signature --

Bob

This is more or less what I've learned to do when shooting in challenging light, but with a small twist.

My E-P5 allows me to adjust the level for shadow and highlight 'blinkies', which I've set at jpeg levels 5 and 250, respectively. In A priority, the rear control wheel adjusts aperture, and the front control wheel adjusts EC.

When taking a shot, exposure is set by keeping an eye on the blinkies while adjusting EC, and while monitoring everything else to be sure I'm not totally killing the shadows. EC will generally fall between -0.5 and -2.5, but the critical issue is watch where the highlights and shadows fall on the live histogram and, via the blinkies, on the image.

The resulting jpeg preview may look off (generally too dark) but this procedure will give me a good raw file for subsequent processing in LR. ISO is a tertiary consideration. I generally try to work at base ISO, but will increase ISO as necessary to keep the viewfinder and jpeg previews usable.

Here are some results http://www.flickr.com/photos/jck_photos/sets/72157635431417147/

It becomes second nature pretty quickly.

I started typing a similar reply, took a break, and here you've gone and done the work, yourself!

The live-view 'blinkies' overexposure indicators are a boon to ETTR raw photogs. Add sufficient controls (i.e. Olympus E-M5, E-P5, E-M1) and you have an ETTR dream machine.

I, too, started out in A mode, mostly, but since Anders W convinced me that there was little advantage to A or S over M when using the blinkies method of ETTR, I have switched, and stay in M 99% of the time.

What comes out are images that are usually within a (<1/3) fraction of a stop from optimal ETTR. As much light as possible collected without blowing any important highlights - consistently.

Since the E-M5 (and E-P5) sensor is not 100% ISO-invariant, I do switch between ISO 200, 400 and 800 (rarely 1600). My sincere hope is that the next generation of OM-D models will be truly ISO-invariant, so I can forget that useless (to me) throwback to ancient history and get on with making properly-exposed photographs - one less useless variable to have to keep in mind.

Couldn't agree more. I've been experimenting a bit with the level settings for the blinkies. 5 and 250 may be unnecessarily conservative, but produce good files. 2 and 253

Anders W is probably right (as usual), but I just like how the controls work in A mode. Back wheel A, front wheel EC, and it's totally second nature. Flick the 2x2 control and then ISO is on the front wheel. I am using M mode more often, but A mode just feels more intuitive to me. Admittedly, this is a personal quirk.

Oly could really own this with a few more tweaks of their UI. They're pretty close now.

With the E-M5 and UniWB, I've found no need to adjust the histogram sensitivity at all. I keep it at 0, 255 and find it works splendidly. I believe Anders found the same thing using autoWB. The typical 1/3EV difference between UniWB and autoWB settings was made up by the fact that I adjust EC (or exposure) to "just under" blinkies, while Anders adjusted to "just showing" blinkies.

I tried Anders' method extensively while traveling this past spring, and I have to agree that it works just as well as the UniWB. I came back with 450 shots (not counting WhiBal readings) and absolutely none of them was problematic for exposure as assessed by RawDigger.

That's not to say I didn't have shots that were incorrect, but I was able to determine that situation immediately with the post-shot blinkies and retake as needed.

Many thanks. I'll be opening up the blinkie window.

I'm also intrigued by your comments regarding autoWB vs uniWB. Normally I use autoWB unless I'm in a tricky, bad lighting situation where I can set up a custom WB. What would be the advantages of uniWB?

It provides post-shot color histograms that are more nearly like the raw-data histograms.  This means that, for most situations, multipliers have not been applied to the B and R channels, allowing them to indicate higher R and B levels than in fact exist.  For most cases where G leads the way, this doesn't make too much difference.  But in cases where R or B lie further to the right, you can't be sure what's really happening using WB other than UniWB.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads