Crop sensor size in relation to exposure

[daralough]

So unfortunately I have been doing some thinking and have drawn the conclusion that crop bodies gather less light. In clear thinking, this would mean the following:

A full frame body at ISO100, 1/60th shutter at f2.8 should have a brighter exposure than a crop body with the exact same parameters (iso100, 1/60th and the same lens at f2.8)

Is this true? Or have I thought to much?

My explanation comes from this fact, say I have a 70-200mm f2.8 that is focussing an image onto a 35mm sensor, that 35mm sensor has an area of just over 870mm squared, now that quantity of light projected from the rear of the lens into the body is covering the sensors full area, now if I put a crop body onto the same lens, the lens projects the same image and the same quantity of ligh, but the crop sensor is only reading a portion of it equal to the crop sensor size. The ff sensor is about 2.5x bigger than the crop sensor, so essentially when an image is projected from that lens, the ff sensor absorbs all the light it projects, whereas the crop sensor will absorb less due to its smaller area.

So this would mean that because the crop sensor is cropping the image projected from the lens, it is also cropping the total light available, hence the exposure would appear darker. In my head the difference will be minimal, but should still be able to be seen. I know the crop factor would not be the way to work this out mathematically (ie, saying that a crop absorbs 1.6x less light is incorrect because this crop factor comes from the diagonal measurement of the sensor which affects the field of view).

So my thinking is that if a ff sensor is roughly 2.5x bigger than the crop in area, would then ff sensor absorb 2.5x more light?

Sorry for the mighty question, hopefully I haven't opened a can of worms and someone can prove my thinking wrong!

 

[dinners]

'More' light rather than 'brighter' light is falling on the full frame sensor.

Too late for me but this link discusses it.

http://photo.stackexchange.com/ques...e-sensors-gather-more-light-than-crop-sensors

I couldn't find it but I'm sure is was discussed before and a couple of analogies were given but It's too late for me to get my head around whether they make sense or not.

Stick a 50cm diameter bucket out in the rain next to a 30cm bucket. They will both fill up to the same level even though the wider bucket will collect more rain.

Light comes through a window and hits a wall on which a painting is hanging. Although the wall is much bigger the light falling on the smaller painting is just a bright.

Somebody like Hoppy will probably explain it properly.

 

[daralough]

Yeah the link makes some good points, but there's some flawed maths, surely you can't divide a f stop by a crop factor. The f stop range from say f2 to f8 is an exponential decay of light, because each f stop decreased the light by a factor of two. That link states that f2/1.6 is f1.25, what it should be is f2 less 60% of a stop, so 60% less than the difference between f2 and f1.4, correct me if I'm wrong however.

And also, the bucket principal is true, but it doesn't apply to cameras unless you involve volume. Exposure is made up of light intensity and time, the same way the with the buckets the volume is made up of rain intensity and time, the depth of the water is irrelevant. The bigger bucket will result in a greater volume of water, the same way a larger sensor will get exposed to a greater 'quantity' of light.

Sorry for using the term brighter, I know the same light intensity will come through the lens regardless of body, but it's the quantity of that light captured that gives ff the advantage.

 

[woof woof]

The total light argument has never made any sense to me and I've never read an argument that's convinced me that it should be true.

I've never read the painting on the wall analogy before but I like it. It's simple and it makes the point. If you use the same lens and settings on two cameras one with a smaller sensor than the other the same light will fall on the smaller sensor as will fall on an identically sized centre portion of the larger sensor and any light falling outside of the central crop area of the FF sensor can not influence the image quality in the centre crop area of the frame just as any light falling on the top of the wall can't make the painting appear brighter.

These days many of us have cameras with different sized sensors and it should be easy enough to point them at something and see the results. Big is usually better but I think it's for reasons other than simply total light.

Certainly exposure wise if I point my FF camera and my MFT camera at the same scene I get near enough the same settings and any differences are no doubt due to what Sony and Panasonic think the ISO/any other setting should be rather than a result of sensor size. I don't get hugely different or longer exposure settings when using MFT.

Exposure wise FF with 50mm at f2.8 will be the same as MFT at 25mm and f2.8. The FF image may be better... but maybe because of other factors including the degree of magnification applied to the final image. Equivalence only muddies the waters IMVHO.

 

[redhed17]

ISO100, 1/60th shutter at f2.8 should record the same exposure on cameras with different sized sensors. You can try it with any cameras you have with different sensors. The size of the sensor becomes relevant when dynamic range and/or high ISO performance may be important because the larger sensor (possibly more efficient light gatherer) may be able to collect more photons of light for each (possibly larger) photosites. I don't know the science of it though.

Depending on age of sensor of course. A Nikon D5300 compared to a Canon 5D gives better dynamic range and higher ISO performance. And that is a smaller sensor with 11 million more pixels.

That is what 8 years of technological advancement can give you.

 

[Nod]

Put a glass on the floor (let's call it a crop sensor for a moment) and pour water onto it from a watering can with a rose fitted. The glass will fill in T seconds. Empty the glass and place another one beside it - this doubling of size is approximately the same as making the 2 glasses a full frame sensor. Repeat the watering. BOTH glasses will fill in T seconds. The same amount of water has fallen on the whole image circle (the ground wetted by the watering can) in the same time (T seconds) but twice as much has been captured in the FF sensor as in the crop one. However, the amount captured over each area is still just a glassful. So more light has been captured by the FF sensor but it's at the same concentration as that captured by the crop.

 

[Phil V]

Sorry for the mighty question, hopefully I haven't opened a can of worms and someone can prove my thinking wrong!

It's wrong.

As above, it's not more light, it's a larger area of the same amount of light.

But larger photosites are an interesting point for understanding dynamic range.

 

[zednini]

A 10 x 8 plate camera has about 60 times the area of 35 mm. That would be about a 5 1/2 stop advantage

 

[TheBigYin]

ISO100, 1/60th shutter at f2.8 should record the same exposure on cameras with different sized sensors.

^^^ This....

I often use my 7D (or even the ridiculously small sensor in my cameraphone) to provide metering details for my MF Bronica.

ISO100, 1/60th shutter at f2.8 is the same on a crop sensor, a "full frame" digital, 4.5x6cm on my ETRS, 6x6cm on a Hasselblad 500 or 5"x4" / 10"x8" on REAL large "sensor" cameras. If the "sensor" size mattered, there would have to be a "sensor size" setting you'd have to adjust on dedicated lightmeters.

 

[Lemaildetom]

And also, the bucket principal is true, but it doesn't apply to cameras unless you involve volume. Exposure is made up of light intensity and time, the same way the with the buckets the volume is made up of rain intensity and time, the depth of the water is irrelevant. The bigger bucket will result in a greater volume of water, the same way a larger sensor will get exposed to a greater 'quantity' of light.

I believe this analogy works, the depth of water is the same in a small bucket and in a large bucket which is the same as the exposure. Let's say that to have a proper exposure you need the bucket to be full (half full = underexpose, flowing over the top = unrecoverable clipped highlight). Then you will need exactly the same exposure to the rain to fill both bucket. It's the volume which is irrelevant.

If you use the same lens and settings on two cameras one with a smaller sensor than the other the same light will fall on the smaller sensor as will fall on an identically sized centre portion of the larger sensor and any light falling outside of the central crop area of the FF sensor can not influence the image quality in the centre crop area of the frame just as any light falling on the top of the wall can't make the painting appear brighter.

I think that a really good explanation Allan! If you are using a FF frame camera. When you crop an image this equates to using a smaller sensor. You do not need to change the exposure.

 

[daralough]

I think we over simplify ISO when comparing systems. When we change the ISO on our cameras, the sensor gain increases or decreases in accordance with the manufacturer parameters.

What if manufacturers tweak the cameras ISO parameters to get an equivalent exposure with a crop sensor? By this I mean what if ISO 100 on a 5d iii is mapped to sensor gain Z, but ISO 100 on a 7d ii is mapped to gain Z multiplied by the difference in area (roughly 2.5x). This would mean that exposures are the same when the settings are applied, but the crop body's sensor is using more power and inducing more noise. Its accepted that full frame has a cleaner image, is this possibly the real reason why?

I could be loosing my marbles here, but it makes an interesting conversation.

I fully accept that in the examples above the same intensity of light falls on each sensor, but when we take a 20mp ff sensor and a 20mp crop, we know that the crop records less light per photo site due to the higher pixel density (smaller pixel area).
I think this may validate my theory that a crop sensor has to run higher sensor gain to keep up with the full frame in terms of exposure when the same setting are applied.

 

[woof woof]

I think that a really good explanation Allan! If you are using a FF frame camera. When you crop an image this equates to using a smaller sensor. You do not need to change the exposure.

Er... thank you

It would be nice if we could easily compare a FF camera to a MFT or APS-C camera with known identical technology but I don't know if we can do that as we can't be sure what different hardware and programming changes have been incorporated into each camera even if they appear quite similar. To me the only way to be sure of getting exactly the same technology would be to use a FF camera in crop mode as then we'd know that the hardware and processing was exactly the same but if we do that we may as well just take a shot with the FF camera and crop it in our photo processing package of choice and if we do that the only image quality differences we should see should be due to the different amounts of enlargement applied to each image.

I'm not an optical engineer (electronics and computing was my thing) but reading some of the discussions on this subject I often feel I'm missing something as some of the arguments put forward seem to rely on some magical property inherent in sensor size whereas I'd attribute the quality gains or losses more to things that I personally can understand and measure... such as magnification of the image and better circuit design. For example many people would be very surprised if they saw the effect a track (including its properties and it's route) has on a signal and these tangible and measurable and engineerable things will explain why some small sensors apparently outperform some larger sensors whereas if simply size and total light was the answer this couldn't be the case.

 

[woof woof]

What if manufacturers tweak the cameras ISO parameters to get an equivalent exposure with a crop sensor? By this I mean what if ISO 100 on a 5d iii is mapped to sensor gain Z, but ISO 100 on a 7d ii is mapped to gain Z multiplied by the difference in area (roughly 2.5x). This would mean that exposures are the same when the settings are applied, but the crop body's sensor is using more power and inducing more noise. Its accepted that full frame has a cleaner image, is this possibly the real reason why?

I could be loosing my marbles here, but it makes an interesting conversation.

I'm sure that ISO measurements are out there on the various technical review sites and I'm sure they can be found by anyone who wishes to Google them and I'm sure that some cameras will say that they're using ISO 400 when in reality they're using 360 (or something...) but I think ISO variances will possibly be more to do with a manufacturer than a sensor size. True ISO v Camera ISO is no doubt an issue and what ISO is programmed into cameras may well vary from manufacturer to manufacturer or even from model to model but I don't think it's an answer to this question.

As I may have said above, if I point my FF and MFT cameras at the same scene I get the same exposure or something near enough not to raise any concerns.

 

[Lemaildetom]

I think we over simplify ISO when comparing systems.

I think you are actually overcomplicating this a bit too much . Iso is not a number supposed to represent how much gain is apply to each pixel because this is not important to the photographer.

Look at a canon 30D with a 8.2MP cropped sensor
And look at a canon 5D mark 3 with a full frame 22MP sensor
The canon 30D sensor, if enlarged to FF would equates at 8.2MP x 2.5more surface=20.5MP

So they are a fair comparison with similar pixel density. So they must apply the same gain/pixel because each pixel must receive the same number of photon. But at the end of the day it doesn't matter how much gain is apply. What matter is how well it is apply and how much more resulting noise there is in your picture. The gain itself isn't important to you but it should be considered to the constructor which sometimes seems more attracted by increasing pixel density for arguable increase in resolution.

But yes, larger pixel mean more photon per pixel, also mean less noise

 

[daralough]

Il have to really look into this further, intriguing stuff.
So talking about sensor gain, a 5d iii at 1/60th, f2.8, ISO 100 will have a sensor gain of say 'X'. But a 7d ii at the same settings at ISO 100 will have a sensor gain equalivent to X multiplied by 2.5 (the 2.5 factor is steming from the size difference in the sensors).

So what I'm trying to say is that manufacturers up the sensitivity of smaller sensors to put their exposures in 35mm terms, hence why a crop and full frame sensors exposure will look the same at the same settings. This would tie in with the fact that a full frame sensor is cleaner because it's actually running at a lower sensitivity. ISO isn't a actual physical number of sensitivity of the sensor, it's mesrly an index. We won't ever know what sensitivity a sensor runs at without stripping a camera and probing the circuits for current.

Just my ramblings

 

[steveo_mcg]

No, sorry mate that isn't correct. If there was any difference between the iso and "sensor" size then film speeds across format would be impossible to keep a handle on. As it is fp4 is exactly the same speed from miniature (35mm) right through to 8x10 and beyond. That is to say it will respond predictably with the same sensitivity in the same conditions regardless of the format.

 

[daralough]

No, sorry mate that isn't correct. If there was any difference between the iso and "sensor" size then film speeds across format would be impossible to keep a handle on. As it is fp4 is exactly the same speed from miniature (35mm) right through to 8x10 and beyond. That is to say it will respond predictably with the same sensitivity in the same conditions regardless of the format.

Yes that makes perfect sense. Im thinking into things too much

 

[Phil V]

No, sorry mate that isn't correct. If there was any difference between the iso and "sensor" size then film speeds across format would be impossible to keep a handle on. As it is fp4 is exactly the same speed from miniature (35mm) right through to 8x10 and beyond. That is to say it will respond predictably with the same sensitivity in the same conditions regardless of the format.

This^

Yes that makes perfect sense. Im thinking into things too much

However it was pointed out earlier in similar ways...
ISO is ISO (as much as it ever was*) It's an international standard.

* Notwithstanding different meters will give different results from the same scene, different batches of film may have variance in sensitivity, reciprocity failure, etc. etc.

 

[Lemaildetom]

I would had to this that an handheld lightmeter can be used for 35mm, 120, 8x10, digital crop sensor, digital full frame...

If manufacturer tweak the gain in the sensor in order that 1/60th, f/2.8, ISO 100 is the same on any camera it's because it makes sens to do so. ISO is like a benchmark that camera have to fit in. Otherwise it will be impossible to measure exposure...

 

[steveo_mcg]

Yes that makes perfect sense. Im thinking into things too much

Probably
That's not to say that manufacturers don't apply different gain to different sensors even batches within the same model but it will be to meet the iso and will be to account for variability in the silicon not the physical size of the sensor.

 

[daralough]

I don't even want to begin to look at the international standard for our cameras ISO numbers, comes under ISO:12233:2006 just incase anyone has an incredible amount of spare time.

But regardless of our points, crop or full frame, digital sensors are pretty amazing at what they do in terms of sensitivity

 

[HoppyUK]

So unfortunately I have been doing some thinking and have drawn the conclusion that crop bodies gather less light. In clear thinking, this would mean the following:

A full frame body at ISO100, 1/60th shutter at f2.8 should have a brighter exposure than a crop body with the exact same parameters (iso100, 1/60th and the same lens at f2.8)

Is this true? Or have I thought to much?

My explanation comes from this fact, say I have a 70-200mm f2.8 that is focussing an image onto a 35mm sensor, that 35mm sensor has an area of just over 870mm squared, now that quantity of light projected from the rear of the lens into the body is covering the sensors full area, now if I put a crop body onto the same lens, the lens projects the same image and the same quantity of ligh, but the crop sensor is only reading a portion of it equal to the crop sensor size. The ff sensor is about 2.5x bigger than the crop sensor, so essentially when an image is projected from that lens, the ff sensor absorbs all the light it projects, whereas the crop sensor will absorb less due to its smaller area.

So this would mean that because the crop sensor is cropping the image projected from the lens, it is also cropping the total light available, hence the exposure would appear darker. In my head the difference will be minimal, but should still be able to be seen. I know the crop factor would not be the way to work this out mathematically (ie, saying that a crop absorbs 1.6x less light is incorrect because this crop factor comes from the diagonal measurement of the sensor which affects the field of view).

So my thinking is that if a ff sensor is roughly 2.5x bigger than the crop in area, would then ff sensor absorb 2.5x more light?

Sorry for the mighty question, hopefully I haven't opened a can of worms and someone can prove my thinking wrong!

This is about 'equivalence' between formats. All cameras/lenses/sensors have to play by the same basic rules of physics, and in theory you should be able to obtain the exact same result by chopping things about and rearranging them. For example, and using a 1.4x crop factor camera* vs full-frame for ease of maths, these two images should be identical in terms of framing, perspective, exposure/noise, and depth-of-field.

Full-frame
70mm lens, f/5.6, 1/125sec, ISO400

1.4x crop camera
50mm lens, f/4, 1/125sec. ISO200

Good article on equivalence at DPReview here http://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care but there's one important point missing from that - sharpness, that is better on FF, and is probably the main reason that we choose FF over croppers. To understand that, you have to look at lens MTF performance. What we call sharpness is a combination of resolution** and contrast** and MTF plots that - those squiggly graphs that lens manufacturers publish.

A fact of phsyics is when resolution demands go up, contrast goes down - and in terms of perceived sharpness, contrast is more significant than resolution. So, to maintain equivalent sharpness on a 1.4x crop format, the lens has to deliver the same standard of contrast at 1.4x the resolution, eg 42 lines-per-mm vs 30 lines-per-mm. That is a very big ask, typically the difference between a very average lens on FF vs state of the art on the cropper. In practise that doesn't happen often, if at all, which is why FF always holds the advantage on image quality.

*1.4x doesn't exist but would have exactly half the sensor area of FF, 1.4 being square root of 2.

**Resolution is fineness of detail, usually quoted in lines-per-mm on a test target. Contrast is how clearly those details are shown, eg black on white lines are much clearer to see than grey on white. When adding sharpening in post-processing, the software darkens the edges around details - basically increasing edge contrast and perceived sharpness.

 

[Terrywoodenpic]

Put a 135mm 2.8 lens on a 5x4 film camera and make an exposure..of an evenly lit wall.... and then cut the film down to 35mm size or aps.
All the various bits of film will have collected the same intensity of light and and had the same exposure.
iso and exposue are independent of film size.

Quality is another issue entirely...a good big one always beats a good small one. That was true of film it is truè for sensors,

however if it is good enough ... there is no reason to go larger......

 

[daralough]

Richard, thanks for that link its absolutely brilliant. To quote it:
'ISO is useful, in that it means that the same set of exposures work across all cameras (and frankly, it'd get confusing, otherwise). However, it ends up disguising how much total light each system gets. Since the light intensity is the same (per square mm), the Full Frame camera will receive four times as much light as the Four Thirds camera, during those exposures, because it has four times the sensor area, all experiencing that same intensity'
So in plain english that means that as I said, a full frame sensor will capture 4 time as much light as a Four Thirds, when both systems are mounted with an identical lens.
I think people are misled with ISO numbers, they ensure equivalence across the range of digital cameras. A Four thirds sensor is running two stops of additional sensitivity when compared to a 35mm in the same condition.
'This means that, a Four Thirds camera with a 50mm f/2 lens at ISO100 should produce a JPEG of the same brightness as a Full frame camera with a 100mm f/2 lens at ISO100 and, set to the same F-number and shutter speed, even though its smaller sensor means it is receiving 1/4 as much total light.' - Total light theory right there.
ISO is not a direct measure of what sensitivity our sensors are at, its a measure of exposure of a JPEG image. Across all digital formats, ISO is an equivalence.
When I compare a 5d iii at 1/60th, f2.8 ISO 100 to a Four Thirds running the same settings, the two pictures will be identical in exposure, even though the full frame sensor is capturing 4 times the amount of light due to its area. So why does the four thirds give the same exposure? Simple, when you set your four thirds camera to ISO100, the sensor sensitivity is increased to cope with the fact that it is receiving less light than the full frame, in order to keep everything equivalent. This boost of sensitivity that keeps everything equivalent, is a contributing factor towards the noise created by smaller sensors, they are working harder to capture light so they are induce more noise.
ISO equivalence is very complex, as its not something we have access to as we don’t know exactly how our cameras are programmed.

 

[Lemaildetom]

Am I reading this wrong again, I think you're misleading and that I'm right! (but at the end of the day if I'm wrong I'm ok with being told that I am so please do)

A FF sensor receives 4 times the quantity of light because it has a four time bigger sensor yes.
So why do the get the same exposure?
Just because they get the same light per unit of surface.
The total light is irrelevant, a pixel count the photon that fall on it the total light has no matter to it.

If you take the sensor of the 5d mark iii and chop it down to the size of a micro 4/3 sensor and put in a camera. Do you thing you will need to apply more gain to the sensor to get the same exposure? You can realised that experimentation in an easier way by cropping a picture too!

the sensor sensitivity is increased to cope with the fact that it is receiving less light than the full frame

It's not the total light that matter, neither the size of the sensor that matter but the size of the pixel will determine how much gain is required.

Small sensor camera have a large mega-pixel count and therefore a much bigger pixel density. This involve that each pixel don't get has much light, this as nothing to do with the total light but with the light per pixel received. So smaller pixel camera have to apply more gain and are more noisy because of their higher pixel density.

A good exemple of this is the Sony full frame A7s which with only 12mp (very low for a full frame nowadays) has some amazing high iso capability which are much better that the sony full frame A7r with 36mp. They do have the same total light but surely with 36mp you need to boost more each pixel to reach the same exposure.

I might be wrong but I don't see how the total light influence the results, I think the light per pixel is the important bit when it came to high iso, and also likely that the generation of the camera with recent camera being better at copying with gain.

 

[dinners]

the Full Frame camera will receive four times as much light as the Four Thirds camera, during those exposures, because it has four times the sensor area, all experiencing that same intensity'
So in plain english that means that as I said, a full frame sensor will capture 4 time as much light as a Four Thirds, when both systems are mounted with an identical lens.

That's what everybody has been explaining since the start of the thread Dara.

'More' light rather than 'brighter' light is falling on the full frame sensor so exposure is equal.

 

[chris malcolm]

If we were to conduct this argument in terms of numbers and equations rather than words it would be interesting to see how much less text would be required

 

[Phil V]

Richard, thanks for that link its absolutely brilliant. To quote it:
'ISO is useful, in that it means that the same set of exposures work across all cameras (and frankly, it'd get confusing, otherwise). However, it ends up disguising how much total light each system gets. Since the light intensity is the same (per square mm), the Full Frame camera will receive four times as much light as the Four Thirds camera, during those exposures, because it has four times the sensor area, all experiencing that same intensity'
So in plain english that means that as I said, a full frame sensor will capture 4 time as much light as a Four Thirds, when both systems are mounted with an identical lens.
I think people are misled with ISO numbers, they ensure equivalence across the range of digital cameras. A Four thirds sensor is running two stops of additional sensitivity when compared to a 35mm in the same condition.
'This means that, a Four Thirds camera with a 50mm f/2 lens at ISO100 should produce a JPEG of the same brightness as a Full frame camera with a 100mm f/2 lens at ISO100 and, set to the same F-number and shutter speed, even though its smaller sensor means it is receiving 1/4 as much total light.' - Total light theory right there.
ISO is not a direct measure of what sensitivity our sensors are at, its a measure of exposure of a JPEG image. Across all digital formats, ISO is an equivalence.
When I compare a 5d iii at 1/60th, f2.8 ISO 100 to a Four Thirds running the same settings, the two pictures will be identical in exposure, even though the full frame sensor is capturing 4 times the amount of light due to its area. So why does the four thirds give the same exposure? Simple, when you set your four thirds camera to ISO100, the sensor sensitivity is increased to cope with the fact that it is receiving less light than the full frame, in order to keep everything equivalent. This boost of sensitivity that keeps everything equivalent, is a contributing factor towards the noise created by smaller sensors, they are working harder to capture light so they are induce more noise.
ISO equivalence is very complex, as its not something we have access to as we don’t know exactly how our cameras are programmed.

Nope. You're conflating again. The same amount of light is hitting each point of the sensor, but the bigger sensor is spreading them out more, there's no increased 'light sensitivity' (measured in ISO), but there's a greater ability to measure that light which we would see as increased Dynamic Range.
Keep thinking back to the window light on the wall. No matter how much of it you're looking at, the same intensity of light is hitting it

 

[daralough]

Nope. You're conflating again. The same amount of light is hitting each point of the sensor, but the bigger sensor is spreading them out more, there's no increased 'light sensitivity' (measured in ISO), but there's a greater ability to measure that light which we would see as increased Dynamic Range.
Keep thinking back to the window light on the wall. No matter how much of it you're looking at, the same intensity of light is hitting it

Ok thinking back to the wall theory, imagine the pictures are solar panels, one panel will generate more output. The same way when I thought about the bucket theory in terms of volume of water.

 

[Lemaildetom]

Ok thinking back to the wall theory, imagine the pictures are solar panels, one panel will generate more output. The same way when I thought about the bucket theory in terms of volume of water.

Yes, of course and that a really good analogy. Their is an advantage at having more that one solar panel (or a larger sensor) but that doesn't change that the solar panel isn't more efficient because it's place next to another one.

 

[Terrywoodenpic]

It is not the size of the sensor that changes the potential dynamic range, but the size and density of the pixels. and also the bit depth of the system.

The larger the physical size of a pixel element the more photons can be captured at any given light intensity. (Rather like the bigger the net the more fish you can capture in one go.)

 

[DizMatt]

so having not read the whole thread (sorry) here is my tuppence.
If you crop into an image, the centre is often darker than the surroundings, so therefore taking a meter reading from that section might be read as being darker.
The centre of a portrait outside, might be darker than a whole portrait that includes the sky.
To test this, if you have a matrix metering effect, is the metering different when you do a partial metering? well sometimes, the answer will be yes.

However, what's not taken into account (has someone said this yet?) that if you have a cropped sensor, then you will reframe the subject anyway. Less angle of view surely means that you might move your feet.
So if the scene viewed is the same, then the reflected light is the same. certainly not 2.5x different....

 

[Terrywoodenpic]

so having not read the whole thread (sorry) here is my tuppence.
If you crop into an image, the centre is often darker than the surroundings, so therefore taking a meter reading from that section might be read as being darker.
The centre of a portrait outside, might be darker than a whole portrait that includes the sky.
To test this, if you have a matrix metering effect, is the metering different when you do a partial metering? well sometimes, the answer will be yes.

However, what's not taken into account (has someone said this yet?) that if you have a cropped sensor, then you will reframe the subject anyway. Less angle of view surely means that you might move your feet.
So if the scene viewed is the same, then the reflected light is the same. certainly not 2.5x different....

if you use the more accurate Incident light meter system for exposure estimation, You will see that neither the subject nor sensor size has any bearing on the exposure measurement.


I have rarely read a more confused thread than this.
Back to basics..... seems to be the order of the day for many of the contributors.

I hope people are just having a laugh................