F/ number in relation to the sensor size

[tbcass]

Yes the sarcastic tone borders on trolling.

 

[Moti]

Absolutely. And lumigraphics is a good example of that: someone who, I readily grant, is a talented photographer and who demonstrates that one doesn't need to know what's really going on to make good photographs.

Sounds like an oxymoron to me. How can someone be a talented photographer and yet, doesn't know how to make a good photo?

There are many different routes to 'making a good photo' and much of what people choose to argue about here isn't central to it. Modern cameras will do a good deal of it automatically.

I agree. But all I was asking is to see examples of good photos coming from someone rude and arrogant with a rich history of name calling names everyone who doesn't agree with him. Is it too much to ask?

But then maybe it is time for you, to enlighten us from the back of your high horse. Please explain what really is going on to make a good photograph? and I mean not only technically good.

Well, some people have an eye fora shot and others don't. Some people arrive there as if by magic, for others it's a hard learned skill.

That is true, but if this is the case here, why would one constantly offend others just because they take better photos than him?

Or even better, why won't you take the opportunity and show us some examples of good photography you have created because frankly speaking, maybe they are hidden somewhere but I haven't seen anything that I can call good photography, coming from you.

You're confusing two different things, talent and knowledge. A talented racing driver might be very good at winning races. You wouldn't expect him to know everything about how the car actually worked. Engineers are employed for that.

Bob, I'm not confusing. And I would expect a racing driver to know everything about cars and engines if this is what will help them to get better and win a race.

His value as a teacher, if it exists at all, would be via mute demonstration and not by verbiage, either spoken or written.

Not sure about this either. He may not be a good teacher as you say, but when I compare his photos to your photos, I have no doubt about the fact who can teach whom a lesson in photography, if of course you'd be clever enough to listen.

There are many aspects to photography. Being good at crafting shots doesn't mean that you know the technical stuff, and lumigraphics certainly doesn't. I've known many professional photographers and photography teachers. Some produce wonderful images. Very few know very much about the technical side of photography, especially digital photography.

It's more interesting to ask the question the other way round. Why do some people cling so dear to their photographic knowledge, even though it's shown to be completely wrong? Is it because it's all they have?

Maybe because of the same reason for which some people here are so dear of their technical knowledge even though in practice, it doesn't help them very much to take good photos ? Go figure.

One thing is sure, when it comes to photography as a craft and an art , as much as theoretical knowledge is nice to have or maybe even important in some cases, talent, skills and experience are much more important and I believe that like any other art and craft, also in photography the prove is in the pudding.

Moti

 

[Dennis]

Absolutely. And lumigraphics is a good example of that: someone who, I readily grant, is a talented photographer and who demonstrates that one doesn't need to know what's really going on to make good photographs.

Sounds like an oxymoron to me. How can someone be a talented photographer and yet, doesn't know how to make a good photo?

You're either trying REALLY hard to argue or you didn't understand what he wrote. He said that lumigraphics is a talented photographer who makes good photographs. He makes them despite not knowing what's really going on - in the context of this thread, he's obviously referring to the fact that lumigraphics doesn't understand technical details about how much light the sensor is receiving; that sort of thing. Like a race car driver who knows how to drive his car, but doesn't understand the fuel delivery system. He's acknowledging that you can be a good photographer without knowing this stuff. It would just be nice if the people who don't know this stuff don't insist that they do

• Dennis
  --
  Gallery at http://kingofthebeasts.smugmug.com

 

[Iliah Borg]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Can you please expand on that a bit?

A classic example for the first would be Canon 5D original, if one tries to lift the shadows on a shot taken at base ISO, artifacts become easily visible. To certain extent same can be said of the 5D2 and 5D3.

For the second - consider certain Panasonic cameras, for example. Here is a table of linearity limits for base ISO settings (one can say those are extended low ISO settings, but in fact the gain scales in a linear fashion from there):

DMC-FZ1000 iso 80 3277

AG-GH4 iso 100 2111

DMC-G7 iso 100 2111

DMC-GF7 iso 100 2111

DMC-GH4 iso 100 2111

DMC-GM5 iso 100 2111

DMC-GX8 iso 100 2111

DMC-GX85 iso 100 2111

DMC-LX100 iso 100 2111

DMC-GH3 iso 125 2626

DMC-GM1 iso 125 2626

DMC-GX7 iso 125 2626

 

[Pan X]

I don't disagree with your points.

I'm just trying to get my head around the notion of device size versus "A Different Exposure Equivalent" as alluded to above.

You're not getting your head in there because somewhere at the back of your mind you have the idea that choosing your exposure is totally about determining the output image brightness and nothing else. It's lodged there and it's stopping you thinking about anything else.

No, I think I am trying to establish that a hand held light meter, reading a spot on an 18% grey card at noon, with ISO 100 as my pre-established sensitivity choice, will yield EV 15, or f16 * 1/125 as one of the ten possible f-stop/SS suggestions. The card/brightness level fills the frame. It is a defocussed as I can possibly make it to reduce any hotter specular reflections as much as I can. Paraphrasing the Zone system, "I have metered and placed a calibrated Zone 5 18% gray card in Zone 5, let the others fall where they may". I think this is called scene brightness, is it not? Is it called Exposure? Zone system front to back?

I have been under the impression that spot light meters (or any light meter) always give the readings taken as a Zone 5 answer. If the object was white, then 3/4 stops had to be subtracted to render it white (Zone 8/9) in the final result. If not, the white object was rendered as Zone 5 grey. Ditto, in reverse, for a black object (Zone 1/2).

Therefore, to render any scenes brightness exactly (Exposed exactly?), (without any further manipulations desired or pre-visualized) meter a grey card in the same light as the scene, and choose from those metered suggestions.

Let me set up my scene this way to remove all doubts and any artistic intents.

I take a reflective, spot meter reading on an 18% grey card, standard day, noon, no clouds, no shadow, card pointed to the sun, meter pointed at the card, no shadow from the meter, no glare from the sun. The meter result should be the "Golden Rule of Exposure". Ansel Adam's Zone Five, if you will. The meter will show, (among the various ten possibilities or so), f16 * 1/125 * ISO 100. I choose to fill the frame and fully defocus on this 18% grey card. I choose ISO 100. I choose 1/125.

Wrong way of thinking about it. The meter didn't choose, 100 ISO. You did. When you did that, you told the camera that an exposure at the sensor of 0.1 lux seconds should produce a 12.7% grey in the output file. You also told the meter to tell you which EV to set to achieve 0.1. lux seconds for an 18% grey in the scene you're photographing, so far as can be determined from light reflected from the scene (18% grey in the scene gets rendered as 12.7% grey in the output image, because lights and specular reflections need to be rendered brighter than white)

This is a simple, controlled set up. No artistic intent.

It seems to me that a "so-called" "Perfect Exposure" should yield a perfect Zone five, middle grey "Exposure" on the finished results, however viewed - LCD, CRT, Plasma, film negative, film positive. Is this true?

You're using the zone system back to front. The point of it was to place the different tones on the right place in the film characteristic to provide the most potential for processing and printing. Ansell Adams spent a lot of time manipulating his image in the darkroom, there was never an intention that some how, automatically a given tone on the scene would be rendered at a particular tone in the final print.

He wasn't at all setting up a system to control print brightness by exposure, he was setting up a system to give the most printable negatives.

But, you can use his system to nail perfect exposure, yes?

Underexposed results should/will lean towards Zone 4 and Overexposed should/will lean towards Zone 6. (reversed for film negative)

Now, for the real question, and the source of my confusion.

The meter suggested f16 for the above conditions.(Obviously, a minor plus/minus factor will be involved. Don't get hung up here.)

Do I need to make adjustments to f16 due to different image device sizes needing different "exposures" due to the "crop factor" or "equivalences" as detailed in the posts far above????

No, if you're using exposure to control brightness, it doesn't change at a set ISO, because that's what ISO defines. But apart from when you're locked into how bid an exposure you can use by base ISO, you'll tend to choose ISO according to image quality constraints, and then you'll chose ISO according to sensor size. So, if you were happy with quality of mFT at 400 ISO, you'd be equally happy with FF at 1600 ISO.

Holy Sh*t. I just got it. Thank you Bob. I'm dense at times, but your words right here somehow penetrated the fog. Got it, got it. This equivalent stuff isn't about "upfront" exposure settings at all, as I had stubbornly thought.

Sorry about taking up your time, but thanks for the 2x4 upside my head.

The proponents seem to be very insistent that f-stop crop factors/equivalences for the different device sizes are very real.

They are very real. They determine very much what the final picture will look like, so long as you select the ISO to keep the equivalence. With two cameras of a generation, set to equivalent settings, it can be very difficult indeed to tell which was the smaller and which the larger sensor. Or turn t the other way round, at equivalent settings, sensor size isn't much of an issue.

Will they influence the results in my test scene above?

Not if you're using that 'exposure to set brightness and don't think about which ISO I'm using' method.

--
Bob.
DARK IN HERE, ISN'T IT?

 

[bobn2]

I don't disagree with your points.

I'm just trying to get my head around the notion of device size versus "A Different Exposure Equivalent" as alluded to above.

You're not getting your head in there because somewhere at the back of your mind you have the idea that choosing your exposure is totally about determining the output image brightness and nothing else. It's lodged there and it's stopping you thinking about anything else.

No, I think I am trying to establish that a hand held light meter, reading a spot on an 18% grey card at noon, with ISO 100 as my pre-established sensitivity choice,

I would counsel against thinking of ISO as 'sensitivity'. It's not what ISO calls it (it's either a 'speed' or an 'exposure index'. The problem with the word 'sensitivity' is that it sets up conceptual models that go nowhere.

will yield EV 15, or f16 * 1/125 as one of the ten possible f-stop/SS suggestions.

What the exposure meter should be doing is giving you an EV setting, which along with the scene illuminance, will give an exposure of the 18% grey card of 0.1 lux seconds. ISO 100 says that the 0.1 lux seconds exposure should be rendered as a 12.7% grey scale.

The card/brightness level fills the frame. It is a defocussed as I can possibly make it to reduce any hotter specular reflections as much as I can. Paraphrasing the Zone system, "I have metered and placed a calibrated Zone 5 18% gray card in Zone 5, let the others fall where they may". I think this is called scene brightness, is it not? Is it called Exposure? Zone system front to back?

Exposure is how much light per unit area is incident on the image plane, or the luminous energy density to be more succinct.

I have been under the impression that spot light meters (or any light meter) always give the readings taken as a Zone 5 answer. If the object was white, then 3/4 stops had to be subtracted to render it white (Zone 8/9) in the final result. If not, the white object was rendered as Zone 5 grey. Ditto, in reverse, for a black object (Zone 1/2).

Therefore, to render any scenes brightness exactly (Exposed exactly?), (without any further manipulations desired or pre-visualized) meter a grey card in the same light as the scene, and choose from those metered suggestions.

Yes, that's right. The chain goes, set ISO to 100. That dictates that an exposure of 0.1 lux seconds is going to be rendered at 12.7% grey. Now arrange for a 18% grey card on the subject to give an exposure of 0.1 lux seconds - that's what the exposure meter is trying to do for you. It has no way of knowing what in the scene is 18% grey, so it's based on some heuristics, like assume that the scene as a whole averages to 18% grey, or assume that the centre of the scene is 18% grey. It's when these heuristics fail to work that you need to apply some compensation to get the correct 0.1 lux seconds for an 18% grey.

Incident light metering circumvents that problem by actually measuring the light on the subject. It surprises me that no-one has come up with a tiny incident light sensor which you can clip on the bride's wherever, and gives a reading by Bluetooth which the camera can pick up directly and use to set the exposure. I suppose you risk the bride going off on her honeymoon with your light meter still attached.

Let me set up my scene this way to remove all doubts and any artistic intents.

I take a reflective, spot meter reading on an 18% grey card, standard day, noon, no clouds, no shadow, card pointed to the sun, meter pointed at the card, no shadow from the meter, no glare from the sun. The meter result should be the "Golden Rule of Exposure". Ansel Adam's Zone Five, if you will. The meter will show, (among the various ten possibilities or so), f16 * 1/125 * ISO 100. I choose to fill the frame and fully defocus on this 18% grey card. I choose ISO 100. I choose 1/125.

Wrong way of thinking about it. The meter didn't choose, 100 ISO. You did. When you did that, you told the camera that an exposure at the sensor of 0.1 lux seconds should produce a 12.7% grey in the output file. You also told the meter to tell you which EV to set to achieve 0.1. lux seconds for an 18% grey in the scene you're photographing, so far as can be determined from light reflected from the scene (18% grey in the scene gets rendered as 12.7% grey in the output image, because lights and specular reflections need to be rendered brighter than white)

This is a simple, controlled set up. No artistic intent.

It seems to me that a "so-called" "Perfect Exposure" should yield a perfect Zone five, middle grey "Exposure" on the finished results, however viewed - LCD, CRT, Plasma, film negative, film positive. Is this true?

You're using the zone system back to front. The point of it was to place the different tones on the right place in the film characteristic to provide the most potential for processing and printing. Ansell Adams spent a lot of time manipulating his image in the darkroom, there was never an intention that some how, automatically a given tone on the scene would be rendered at a particular tone in the final print.

He wasn't at all setting up a system to control print brightness by exposure, he was setting up a system to give the most printable negatives.

But, you can use his system to nail perfect exposure, yes?

'Perfect exposure' is another pretty loaded term which I tend not to use. Using the zone system as Ansell Adams used it won't result in an image exposed as the ISO setting would demand, it's more related to today's ETTR, about getting the best exposure for onward processing, not the one that will render an 18% grey as 12.7% grey. But then he carefully characterised his emulsions for himself, he didn't rely on ASA ratings.

To try to understand what he was trying to do, have a look at this graph of the characteristic of Tri-X Pan

(this was taken from this: http://photo.net/learn/making-photographs/film , which is a pretty good article, apart from playing fast and loose with the word 'sensitivity')

So, to understand what is going on here, the curves show the density of the negative against exposure for different development times. You' see that more development increases the contrast, but also that the curve has a distinct bend (unlike digital, which is pretty much linear). The consequence of this bend is that you can't simply compensate for exposure in the camera with a different exposure time in the enlarger, because the bend changes the relative tonal values depending on where you are on the curve. The point of the zone system was to get the best distribution along the curve for a particular development regime.

Underexposed results should/will lean towards Zone 4 and Overexposed should/will lean towards Zone 6. (reversed for film negative)

Now, for the real question, and the source of my confusion.

The meter suggested f16 for the above conditions.(Obviously, a minor plus/minus factor will be involved. Don't get hung up here.)

Do I need to make adjustments to f16 due to different image device sizes needing different "exposures" due to the "crop factor" or "equivalences" as detailed in the posts far above????

No, if you're using exposure to control brightness, it doesn't change at a set ISO, because that's what ISO defines. But apart from when you're locked into how bid an exposure you can use by base ISO, you'll tend to choose ISO according to image quality constraints, and then you'll chose ISO according to sensor size. So, if you were happy with quality of mFT at 400 ISO, you'd be equally happy with FF at 1600 ISO.

Holy Sh*t. I just got it. Thank you Bob. I'm dense at times, but your words right here somehow penetrated the fog. Got it, got it. This equivalent stuff isn't about "upfront" exposure settings at all, as I had stubbornly thought.

Sorry about taking up your time, but thanks for the 2x4 upside my head.

No problem, I post here so that people will have that 'holy sh*t' experience. It's not just your problem. There is so much disinformation around about the technical side of photography that people often end up swallowing it and then have real trouble getting rid of it, because it conditions the whole conceptual model that they have.

The proponents seem to be very insistent that f-stop crop factors/equivalences for the different device sizes are very real.

They are very real. They determine very much what the final picture will look like, so long as you select the ISO to keep the equivalence. With two cameras of a generation, set to equivalent settings, it can be very difficult indeed to tell which was the smaller and which the larger sensor. Or turn t the other way round, at equivalent settings, sensor size isn't much of an issue.

Will they influence the results in my test scene above?

Not if you're using that 'exposure to set brightness and don't think about which ISO I'm using' method.

--
Bob.
DARK IN HERE, ISN'T IT?

--
Bob.
DARK IN HERE, ISN'T IT?

 

[lovEU]

... (the degree and the highest useful for this ISO setting varies depending on the camera model).

Iliah, what's the way to find out that ISO setting for a given camera? I guess you've shared the answer somewhere already. Thank you for further advice.

 

[Great Bustard]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Can you please expand on that a bit?

A classic example for the first would be Canon 5D original, if one tries to lift the shadows on a shot taken at base ISO, artifacts become easily visible. To certain extent same can be said of the 5D2 and 5D3.

OK -- this is interesting. Could you go into more details?

For the second - consider certain Panasonic cameras, for example. Here is a table of linearity limits for base ISO settings (one can say those are extended low ISO settings, but in fact the gain scales in a linear fashion from there):

DMC-FZ1000 iso 80 3277

AG-GH4 iso 100 2111

DMC-G7 iso 100 2111

DMC-GF7 iso 100 2111

DMC-GH4 iso 100 2111

DMC-GM5 iso 100 2111

DMC-GX8 iso 100 2111

DMC-GX85 iso 100 2111

DMC-LX100 iso 100 2111

DMC-GH3 iso 125 2626

DMC-GM1 iso 125 2626

DMC-GX7 iso 125 2626

I'm not sure what the numbers represent. Are the first set of numbers the lowest ISO setting before non-linearity occurs and the last set of numbers the highest ISO setting?

 

[Iliah Borg]

OK -- this is interesting.

I try

Could you go into more details?

Well, that's very simple experiment. Set some decent exposure for base ISO. Set base + 1EV ISO, same exposure. Set base + 1 EV ISO, exposure -1 EV. Try lifting shadows on all 3, compare noise, banding, other artifacts.

For the second - consider certain Panasonic cameras, for example. Here is a table of linearity limits for base ISO settings (one can say those are extended low ISO settings, but in fact the gain scales in a linear fashion from there):

DMC-FZ1000 iso 80 3277

AG-GH4 iso 100 2111

DMC-G7 iso 100 2111

DMC-GF7 iso 100 2111

DMC-GH4 iso 100 2111

DMC-GM5 iso 100 2111

DMC-GX8 iso 100 2111

DMC-GX85 iso 100 2111

DMC-LX100 iso 100 2111

DMC-GH3 iso 125 2626

DMC-GM1 iso 125 2626

DMC-GX7 iso 125 2626

I'm not sure what the numbers represent. Are the first set of numbers the lowest ISO setting before non-linearity occurs and the last set of numbers the highest ISO setting?

The last number is linearity limit at that ISO. Max = 4095.

 

[Iliah Borg]

if you can't expose to the right at base ISO, increasing ISO setting helps to get cleaner image (the degree and the highest useful for this ISO setting varies depending on the camera model).

what's the way to find out that ISO setting for a given camera?

Here is a shortcut, http://www.photonstophotos.net/Charts/PDR_Shadow.htm

Note an ISO setting with less then 1/3 EV shadow improvement accumulated from that ISO setting to the highest.

Further, meter from some gray card under incandescent light, setting the camera to ISO from above. Shoot some scene with deep shadows at the resulting settings, and go down on ISO only from there. Lift the shadows on all the shots and see the point where you are comfortable with the results.

 

[PK24X36NOW]

Great answer - but weird from a dpreview standpoint. A wrench, hunh? Hope I didn't insult you. Just curious. Usually I use a wrench to fix things . . . or totally screw them up. I like to keep things simple. Honest. Brain's getting tired, I guess.

Certainly no insult. I too have been able to use a wrench to many ends – not all good.

Heh sometimes use of a wrench seems to be best way to skin knuckles.

 

[lovEU]

if you can't expose to the right at base ISO, increasing ISO setting helps to get cleaner image (the degree and the highest useful for this ISO setting varies depending on the camera model).

what's the way to find out that ISO setting for a given camera?

Here is a shortcut, http://www.photonstophotos.net/Charts/PDR_Shadow.htm

Note an ISO setting with less then 1/3 EV shadow improvement accumulated from that ISO setting to the highest.

Further, meter from some gray card under incandescent light, setting the camera to ISO from above. Shoot some scene with deep shadows at the resulting settings, and go down on ISO only from there. Lift the shadows on all the shots and see the point where you are comfortable with the results.

Thank you very much! But sorry, I do not quite understand the first part "an ISO setting with less then 1/3 EV shadow improvement accumulated from that ISO setting to the highest". Does "highest" mean "highest ISO setting"? If so would it suggest i.e. for a ...

- D810 > ISO 12.800?
- D700 > ISO 800?
- D7200 > n.a. ("ISO-less")??

Since this would provide ISO 819.200 for a D500 I guess I got it wrong.

 

[PravdaAmI]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Typically the base is is the one where the full well capacity is the largest. Maybe one or two cameras have extended low ISO where FWC is even larger than on base ISO, but with significant non linearity effect. But in general base ISO has max FWC and every doubling of ISO effectively halves the FWC.

Cleaniness of shadows vs other ISOs depends on exposure. With the same exposure higher ISO tends to be cleaner.

2. if you can't expose to the right at base ISO, increasing ISO setting helps to get cleaner image (the degree and the highest useful for this ISO setting varies depending on the camera model).

Exactly.

 

[bobn2]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Typically the base is is the one where the full well capacity is the largest. Maybe one or two cameras have extended low ISO where FWC is even larger than on base ISO, but with significant non linearity effect. But in general base ISO has max FWC and every doubling of ISO effectively halves the FWC.

'Full well capacity' is a misused term in the context of CMOS sensors. They are rarely used at anything very near full well capacity. In any case it isn't 'full well' capacity, it's 'source follower voltage swing' capacity, and one of the issues is that the sensor can become quite non-linear at the top end of the range. The characteristics of different sensor architectures can make the effects of saturation quite different. For instance, Sony sensors typically use a CCD like stripe, rather than individual isolated photodetectors. One effect of that is that saturation will tend to cause blooming rather than hard clipping.

--
Bob.
DARK IN HERE, ISN'T IT?

 

[Iliah Borg]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Typically the base is is the one where the full well capacity is the largest. Maybe one or two cameras have extended low ISO where FWC is even larger than on base ISO, but with significant non linearity effect. But in general base ISO has max FWC and every doubling of ISO effectively halves the FWC.

That's how it used to be, and still more or less is; however "one or two cameras" is a tad of understatement (given a baker's dozen of Panasonics, if you include FZ2000/FZ2500; or we can re-define base ISO to include those FWC-limited designs), and with switched capacitors and gain it is not a given that the FWC halves.

However, to me the most important part is about shadow noise at base ISO.

With the same exposure higher ISO tends to be cleaner.

Yes, but in certain cases shadows may be cleaner even with less exposure, just because of the effect of amplification. Something to check for any particular camera, design decisions tend to change.

 

[bobn2]

ETTR) which has to be done at base ISO, by definition

It is a little bit different.

1. Not all cameras have the cleanest shadows / max well capacity at base ISO;

Typically the base is is the one where the full well capacity is the largest. Maybe one or two cameras have extended low ISO where FWC is even larger than on base ISO, but with significant non linearity effect. But in general base ISO has max FWC and every doubling of ISO effectively halves the FWC.

That's how it used to be, and still more or less is; however "one or two cameras" is a tad of understatement (given a baker's dozen of Panasonics, if you include FZ2000/FZ2500; or we can re-define base ISO to include those FWC-limited designs), and with switched capacitors and gain it is not a given that the FWC halves.

However, to me the most important part is about shadow noise at base ISO.

Presumably because of the character of the shadow noise. You'll almost always (John Sheehy knows of an exception) get the greatest available DR with full exposure at the very bottom of the ISO range, since the ISO control clips a bit off the top each step. So, given that if the highlights are exposed right at the top, that must in theory leave the shadows least noisy. Unfortunately the base ISO read noise can have significant banding and tartan effects in it, the higher ISO read noises less so.