Larger sensors collect more light? Not really

Rambow

Senior Member
Messages
3,010
Solutions
3
Reaction score
3,157
This is something i come across every now and then, so there is this notion that, for example an apsc sensor with a f3.5 lens collects more light than my 1 inch sensor pocket camera at f1.8 which results in lower noise levels and a lower iso used to take the same shot.

Basically some say that f1.8 on a 1 inch sensor is equivalent to something like f5.6 on a apsc sensor.

In reality though, it's nothing like that. I did my testing. A dslr from 2012 with kit lens really struggles to capture the same low light scene where my 2012 sony rx100 has no issue at all, despite the fact that the two should be at least similar. But no. To get the same image i have to use iso 6400 on the larger camera while the rx100 happily uses iso 3200.

My explanation for this comes down to the lens, an kit lens at f3.5/f4 cannot compete against the faster f1.8 lens on the other camera despite having a much smaller sensor.

If i use a faster lens on the dslr the camera can compete against the rx100 however i lose DoF and noise levels will be similar even though they shouldn't.

What i am saying is that in real life, there is no way the larger sensor "gathers" any more light than the smaller sensor even when using equivalent lens.

Like i said, even if i use f1.8 on both the dslr and the pocket camera, i lose DoF on the larger sensor. This can be a good thing or a bad thing, but generally speaking i prefer more DoF not less.

For the reasons stated above i am quite content with the iso results i get from my RX100 VII even with the slow f2.8 lens.

I realise i started this topic without providing a couple of samples, but i can do this later on.

So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera and less blurry shots than f3.5 on apsc at the same iso and exposure.

I know this because i've been using kit lenses for more than a decade and getting sharp images even with maxing out the iso(6400 in my case) indoors is a real struggle even with IS kit lens.

So for me, faster lenses have helped me out way more than being able to use a higher iso, and this is regardles of the sensor size(before i was a rx100 user i had a Panasonic LX3, it had a f2 lens coupled to a even smaller sensor).

The advantages of fast glass are more relevant in real life than sensor size, and i'm saying this as a current owner or 4 different systems(1 inch sensor, M43 and apsc from Canon and Nikon).

On all these cameras the glass and IS system make the difference, not the sensor size.

Among all these the smaller 1 inch sensor cameras actually have an edge simply because the fast lens comes free with the camera, and the lens are tiny and weigh nothing. You lose DoF of course but depending on your style of shooting you may like or dislike this.
 
Last edited:
Basically it's a BS myth, same as 'lower mp cameras are better than higher mp cameras in low light because they have larger pixels which collect more light'.
 
Period.
So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera and less blurry shots than f3.5 on apsc at the same iso and exposure.
 

"However, if you currently have a crop sensor DSLR, you can improve low light photography by getting a lens with a larger diameter. Changing sensor size (e.g. moving to full frame) will not help low light collection unless you get larger lenses. For example, see Figure 4b on the nightscapes web page (reference 3 below), where the same lens is used on full frame, 1.3x crop and 1.6x crop bodies. The 1.6x crop body (Canon 7D) records more and fainter stars even though it has a reputation for being a poor low light camera. Compare that to the Canon 5D Mark II full frame camera which recorded less stars even though it has a reputation for being a good low light camera. The reason is that` people making the comparisons are not understanding Etendue; they change Etendue and erroneously blame sensors."
 
This is something i come across every now and then, so there is this notion that, for example an apsc sensor with a f3.5 lens collects more light than my 1 inch sensor pocket camera at f1.8 which results in lower noise levels and a lower iso used to take the same shot.
Perhaps you are misunderstanding what was said. Can you actually give a link to where somebody said that? The usual, correct, statement is that a larger sensor collects more light at a given exposure.
Basically some say that f1.8 on a 1 inch sensor is equivalent to something like f5.6 on a apsc sensor.
Then they are wrong. F/1.8 on a 1" sensor is equivalent to approximately f/3.2 on an APS-C (Sony/Nikon/Fujifilm) sensor or f/3 on a Canon APS-C sensor. It is also equivalent to about f/4.8 on FF.
In reality though, it's nothing like that. I did my testing.
Yeah it will be nothing like that if they get the equivalency wrong by more than a stop.
A dslr from 2012 with kit lens really struggles to capture the same low light scene where my 2012 sony rx100 has no issue at all, despite the fact that the two should be at least similar. But no. To get the same image i have to use iso 6400 on the larger camera while the rx100 happily uses iso 3200.
ISO 6400 on APS-C should be less noisy than ISO 3200 on 1", if the remaining settings give the same lightness.
My explanation for this comes down to the lens, an kit lens at f3.5/f4 cannot compete against the faster f1.8 lens on the other camera despite having a much smaller sensor.
The sensor is slightly less than two stops smaller, but f/4 collects less than two stops lower light than f/1.8, so of course the 1" does better under those conditions.
If i use a faster lens on the dslr the camera can compete against the rx100 however i lose DoF
A larger sensor only gets a noise advantage by using a shallower DOF, (if the shutter and scene luminance are the same).
and noise levels will be similar even though they shouldn't.
Who says they shouldn't be similar? F/2.8 on the APS-C camera will provide only 1/3 more stop of light than f/1.8 on the 1". That's similar noise. You'd probably get similar noise if the APS-C was at f/2.5.
What i am saying is that in real life, there is no way the larger sensor "gathers" any more light than the smaller sensor even when using equivalent lens.
Then what you are saying is flat out wrong. At the same exposure, a larger sensor gathers more light,. It is hardly a novel discovery that reducing the exposure reduces the amount of light gathered.
Like i said, even if i use f1.8 on both the dslr and the pocket camera, i lose DoF on the larger sensor.
Yes. If the larger sensor doesn't use a shallower DOF (assuming same shutter and scene luminance) it has no noise advantage. It gets that shallower DOF at the same f-number or at any f-number between the same one and the smaller sensor's f-number multiplied by the relative crop factor.

I find that, in practice, one is more often constrained WRT only one of DOF or shutter speed in a single shot. So for most shots, the larger sensor can still capture more light., because it will use the same or similar exposure. For those shots were both DOF and shutter speed are constrained, a larger sensor has no noise advantage.

One of the most common shooting situation is a landscape in good light. We'll assume that f/11 gives the right DOF on a Nikon ZII. On your RX100 VII you'd get the same DOF with f/4. In Sunny-16 light, you'd use settings of {1/2,500, f/4, ISO 125} on your RX100 VII. That high shutter speed is a total waste in a typical landscape, but you have to use it to avoid overexposing. On a Z7 II, I'd use {1/160, f/11, ISO 64}. This gives me the same DOF, but I actually have one stop more exposure, so I have captured 4 stops more light (one stop for the exposure difference & three more stops for the sensor size difference. My shadows will be way less noisy than yours.
This can be a good thing or a bad thing, but generally speaking i prefer more DoF not less.
I prefer just enough DOF. More than that gives more diffraction blur and more noise.
For the reasons stated above i am quite content with the iso results i get from my RX100 VII even with the slow f2.8 lens.
I'm happy that you are content.
...

So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera
It is 1/3 stop faster. I doubt you can actually notice that..
and less blurry shots
Blurry? Because of a 1/3 stop difference in shutter speed? Or do you mean noisier? It's 1/3 stop less noisy too, Most peeple won't notice that.
than f3.5 on apsc at the same iso and exposure.
Wait! "At the same ISO and exposure"? That means the APS-C camera is using a slower shutter. But why would you use the same exposure on a different format. That's just silly. At the same exposure, the APS-C camera will gather about 4 times as much light
I know this because i've been using kit lenses for more than a decade and getting sharp images even with maxing out the iso(6400 in my case) indoors is a real struggle even with IS kit lens.

So for me, faster lenses have helped me out way more than being able to use a higher iso, and this is regardles of the sensor size(before i was a rx100 user i had a Panasonic LX3, it had a f2 lens coupled to a even smaller sensor).
Faster lenses allow higher exposure and higher exposure is always preferable to raising ISO.
The advantages of fast glass are more relevant in real life than sensor size,
That's not true. If you have a 3-stop difference in sensor size, you'd need a lens that was more than three stops faster to get a better result from the faster lens than from the larger sensor.
and i'm saying this as a current owner or 4 different systems(1 inch sensor, M43 and apsc from Canon and Nikon).

On all these cameras the glass and IS system make the difference, not the sensor size.
On all of these systems, both the glass and the sensor size make a difference.
 
This is something i come across every now and then, so there is this notion that, for example an apsc sensor with a f3.5 lens collects more light than my 1 inch sensor pocket camera at f1.8 which results in lower noise levels and a lower iso used to take the same shot.
Perhaps you are misunderstanding what was said. Can you actually give a link to where somebody said that? The usual, correct, statement is that a larger sensor collects more light at a given exposure.
Basically some say that f1.8 on a 1 inch sensor is equivalent to something like f5.6 on a apsc sensor.
Then they are wrong. F/1.8 on a 1" sensor is equivalent to approximately f/3.2 on an APS-C (Sony/Nikon/Fujifilm) sensor or f/3 on a Canon APS-C sensor. It is also equivalent to about f/4.8 on FF.
In reality though, it's nothing like that. I did my testing.
Yeah it will be nothing like that if they get the equivalency wrong by more than a stop.
A dslr from 2012 with kit lens really struggles to capture the same low light scene where my 2012 sony rx100 has no issue at all, despite the fact that the two should be at least similar. But no. To get the same image i have to use iso 6400 on the larger camera while the rx100 happily uses iso 3200.
ISO 6400 on APS-C should be less noisy than ISO 3200 on 1", if the remaining settings give the same lightness.
My explanation for this comes down to the lens, an kit lens at f3.5/f4 cannot compete against the faster f1.8 lens on the other camera despite having a much smaller sensor.
The sensor is slightly less than two stops smaller, but f/4 collects less than two stops lower light than f/1.8, so of course the 1" does better under those conditions.
If i use a faster lens on the dslr the camera can compete against the rx100 however i lose DoF
A larger sensor only gets a noise advantage by using a shallower DOF, (if the shutter and scene luminance are the same).
and noise levels will be similar even though they shouldn't.
Who says they shouldn't be similar? F/2.8 on the APS-C camera will provide only 1/3 more stop of light than f/1.8 on the 1". That's similar noise. You'd probably get similar noise if the APS-C was at f/2.5.
What i am saying is that in real life, there is no way the larger sensor "gathers" any more light than the smaller sensor even when using equivalent lens.
Then what you are saying is flat out wrong. At the same exposure, a larger sensor gathers more light,. It is hardly a novel discovery that reducing the exposure reduces the amount of light gathered.
Like i said, even if i use f1.8 on both the dslr and the pocket camera, i lose DoF on the larger sensor.
Yes. If the larger sensor doesn't use a shallower DOF (assuming same shutter and scene luminance) it has no noise advantage. It gets that shallower DOF at the same f-number or at any f-number between the same one and the smaller sensor's f-number multiplied by the relative crop factor.

I find that, in practice, one is more often constrained WRT only one of DOF or shutter speed in a single shot. So for most shots, the larger sensor can still capture more light., because it will use the same or similar exposure. For those shots were both DOF and shutter speed are constrained, a larger sensor has no noise advantage.

One of the most common shooting situation is a landscape in good light. We'll assume that f/11 gives the right DOF on a Nikon ZII. On your RX100 VII you'd get the same DOF with f/4. In Sunny-16 light, you'd use settings of {1/2,500, f/4, ISO 125} on your RX100 VII. That high shutter speed is a total waste in a typical landscape, but you have to use it to avoid overexposing. On a Z7 II, I'd use {1/160, f/11, ISO 64}. This gives me the same DOF, but I actually have one stop more exposure, so I have captured 4 stops more light (one stop for the exposure difference & three more stops for the sensor size difference. My shadows will be way less noisy than yours.
This can be a good thing or a bad thing, but generally speaking i prefer more DoF not less.
I prefer just enough DOF. More than that gives more diffraction blur and more noise.
For the reasons stated above i am quite content with the iso results i get from my RX100 VII even with the slow f2.8 lens.
I'm happy that you are content.
...

So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera
It is 1/3 stop faster. I doubt you can actually notice that..
and less blurry shots
Blurry? Because of a 1/3 stop difference in shutter speed? Or do you mean noisier? It's 1/3 stop less noisy too, Most peeple won't notice that.
than f3.5 on apsc at the same iso and exposure.
Wait! "At the same ISO and exposure"? That means the APS-C camera is using a slower shutter. But why would you use the same exposure on a different format. That's just silly. At the same exposure, the APS-C camera will gather about 4 times as much light
I know this because i've been using kit lenses for more than a decade and getting sharp images even with maxing out the iso(6400 in my case) indoors is a real struggle even with IS kit lens.

So for me, faster lenses have helped me out way more than being able to use a higher iso, and this is regardles of the sensor size(before i was a rx100 user i had a Panasonic LX3, it had a f2 lens coupled to a even smaller sensor).
Faster lenses allow higher exposure and higher exposure is always preferable to raising ISO.
The advantages of fast glass are more relevant in real life than sensor size,
That's not true. If you have a 3-stop difference in sensor size, you'd need a lens that was more than three stops faster to get a better result from the faster lens than from the larger sensor.
and i'm saying this as a current owner or 4 different systems(1 inch sensor, M43 and apsc from Canon and Nikon).

On all these cameras the glass and IS system make the difference, not the sensor size.
On all of these systems, both the glass and the sensor size make a difference.
"The sensor is slightly less than two stops smaller, but f/4 collects less than two stops lower light than f/1.8, so of course the 1" does better under those conditions."

No, it collects about the same light, if the diameters are similar.
 
On all these cameras the glass and IS system make the difference, not the sensor size.
That's very wrong indeed (as well as some other statements you've made).

Think in terms of exposure. Given the same exposure and the same field of view, a larger sensor will collect more total light than a smaller sensor.

Bigger sensor area has greater capacity in terms of light collection (photon counting). In order to fill this capacity, you need bigger lenses (wider aperture glass) or longer exposure time (or both).

You can also think in terms of the total light gathered by a sensor. When a small sensor gets saturated (highlights become blown), a larger sensor can take the same amount of light being distributed over its larger area, so its pixels won't be saturated.

So of course a larger sensor is able to collect more light if you give it more light.
 
I have to disagree with this.

Although I have not tested a Full Frame camera with say a 1 inch sensor camera, I have tested an APSC, 1 inch sensor and 1/2.3 inch sensors superzooms.

Both the APSC and 1 inch is able to focus in dark setting, but the 1/2.3 inch camera fails or results in an image with heavy noise. Even using a small sensor camera with max aperture f2.8 like the Panasonic FZ300 was no match for the APSC or 1 inch sensor.
 
This is something i come across every now and then, so there is this notion that, for example an apsc sensor with a f3.5 lens collects more light than my 1 inch sensor pocket camera at f1.8 which results in lower noise levels and a lower iso used to take the same shot.
The amount of light is not specified by aperture alone but by a combination of shutter speed & aperture. In an equivalent setting, there is no benefit of the larger sensor. However, if you can keep the equivalent aperture but increase the shutter speed (give the camera more light), a larger sensor will collect more light. You cannot use equivalent settings to get the most out of a larger sensor.
Basically some say that f1.8 on a 1 inch sensor is equivalent to something like f5.6 on a apsc sensor.

In reality though, it's nothing like that. I did my testing. A dslr from 2012 with kit lens really struggles to capture the same low light scene where my 2012 sony rx100 has no issue at all, despite the fact that the two should be at least similar. But no. To get the same image i have to use iso 6400 on the larger camera while the rx100 happily uses iso 3200.

My explanation for this comes down to the lens, an kit lens at f3.5/f4 cannot compete against the faster f1.8 lens on the other camera despite having a much smaller sensor.

If i use a faster lens on the dslr the camera can compete against the rx100 however i lose DoF and noise levels will be similar even though they shouldn't.

What i am saying is that in real life, there is no way the larger sensor "gathers" any more light than the smaller sensor even when using equivalent lens.

Like i said, even if i use f1.8 on both the dslr and the pocket camera, i lose DoF on the larger sensor. This can be a good thing or a bad thing, but generally speaking i prefer more DoF not less.

For the reasons stated above i am quite content with the iso results i get from my RX100 VII even with the slow f2.8 lens.

I realise i started this topic without providing a couple of samples, but i can do this later on.

So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera and less blurry shots than f3.5 on apsc at the same iso and exposure.

I know this because i've been using kit lenses for more than a decade and getting sharp images even with maxing out the iso(6400 in my case) indoors is a real struggle even with IS kit lens.

So for me, faster lenses have helped me out way more than being able to use a higher iso, and this is regardles of the sensor size(before i was a rx100 user i had a Panasonic LX3, it had a f2 lens coupled to a even smaller sensor).

The advantages of fast glass are more relevant in real life than sensor size, and i'm saying this as a current owner or 4 different systems(1 inch sensor, M43 and apsc from Canon and Nikon).

On all these cameras the glass and IS system make the difference, not the sensor size.

Among all these the smaller 1 inch sensor cameras actually have an edge simply because the fast lens comes free with the camera, and the lens are tiny and weigh nothing. You lose DoF of course but depending on your style of shooting you may like or dislike this.
 
Just get a tripod, use base ISO, and a longer shutter duration on the larger sensor.
 
Your comments suggest that you don't understand equivalence.
That would require a three year degree course.
Reading and understanding this should be sufficient:

http://www.josephjamesphotography.com/equivalence/
Try this, and read all the pages:

https://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care

Then you can probably confirm it yourself by shooting a properly staged comparison.
 
Just get a tripod, use base ISO, and a longer shutter duration on the larger sensor.
Two tripods, one for each of the two cameras being compared. And bracket the shutter times, to see which time gives the same lightness. Set the aperture and shutter values manually.

If you use Sony cameras, you can put the same lens on the two cameras, taking care to transfer it rapidly in case the light changes.

Don
 
Look, it's quite possible to say one gets acceptable results from smaller sensor even in marginal conditions, like lower light. That's a matter of judgment.

But the amount of noise in shadows is higher with a smaller sensor than it would be with a bigger sensor, contrary to the first post, all else being equivalent. And sometimes it can't be equivalent. That's just science, and unlike a judgment call re aesthetics it's a fact. Really.

And light gathering is a thing. There are solid physical reasons why telescopes have large diameters. We could have saved a lot of money sending up a smartphone camera instead of the JWST (extra wide frame? :) ), but it wouldn't work well. I often see it explained like having an umbrella upside down catching rain; the bigger, the more water. Same with our cameras.

So enjoy the smaller cameras. I do. I enjoy their many advantages, but I don't kid myself about their capabilities.
 
This is something i come across every now and then, so there is this notion that, for example an apsc sensor with a f3.5 lens collects more light than my 1 inch sensor pocket camera at f1.8 which results in lower noise levels and a lower iso used to take the same shot.

Basically some say that f1.8 on a 1 inch sensor is equivalent to something like f5.6 on a apsc sensor.
I'm not sure where you get those f-numbers.

1 inch sensor has a diagonal of about 15.9mm, an APS-C sensor about 28.3mm, give or take a bit in each case. Thus the relative size ratio is about 28.3/15.9=1.78. This means that the f-number of the larger format should be about 1.78 times larger to to create an equivalent output - similar amount of light collected, thus similar noise, similar depth of field, similar diffraction.

Thus f/3.2 should provide similar performance on APS-C that a f/1.8 provides for 1 incher, in this context.
In reality though, it's nothing like that. I did my testing. A dslr from 2012 with kit lens really struggles to capture the same low light scene where my 2012 sony rx100 has no issue at all, despite the fact that the two should be at least similar. But no. To get the same image i have to use iso 6400 on the larger camera while the rx100 happily uses iso 3200.

My explanation for this comes down to the lens, an kit lens at f3.5/f4 cannot compete against the faster f1.8 lens on the other camera despite having a much smaller sensor.
Well, an APS-C needs f/3.2 to compete with f/1.8 on 1 incher in this context.
If i use a faster lens on the dslr the camera can compete against the rx100 however i lose DoF and noise levels will be similar even though they shouldn't.
The thing with depth of field and noise is that you make a tradeoff - you reduce depth of field to collect more light. In other words, you reduce depth of field to reduce noise.
What i am saying is that in real life, there is no way the larger sensor "gathers" any more light than the smaller sensor even when using equivalent lens.
The size of the hole dictates how much light goes through (assuming same framing and exposure time). If the f-nuber is the same, the hole is larger on larger formats - on the other hand, if the depth of field is the same, then the hole will of of the same size and the same amount of light will flow through.
Like i said, even if i use f1.8 on both the dslr and the pocket camera, i lose DoF on the larger sensor. This can be a good thing or a bad thing, but generally speaking i prefer more DoF not less.

For the reasons stated above i am quite content with the iso results i get from my RX100 VII even with the slow f2.8 lens.

I realise i started this topic without providing a couple of samples, but i can do this later on.

So my conclusion is, F1.8 on a 1 inch sensor results in a noticeably faster camera and less blurry shots than f3.5 on apsc at the same iso and exposure.
Well, the difference - ignoring vignetting and image sensor performances, JPG engine performance etc. - is very modest, the same what difference between f/3.2 and f/3.5 is, hardly noticeable. Thus it is likely that any major differences you might see are down to some other reason, maybe one or many of the ones I mentioned above.
I know this because i've been using kit lenses for more than a decade and getting sharp images even with maxing out the iso(6400 in my case) indoors is a real struggle even with IS kit lens.

So for me, faster lenses have helped me out way more than being able to use a higher iso, and this is regardles of the sensor size(before i was a rx100 user i had a Panasonic LX3, it had a f2 lens coupled to a even smaller sensor).
If we compare formats, then f/3.2 on APS-C and f/1.8 on 1" are of equal "speed". Speed is not a synonym of f-number, but is a function of aperture diameter.
The advantages of fast glass are more relevant in real life than sensor size, and i'm saying this as a current owner or 4 different systems(1 inch sensor, M43 and apsc from Canon and Nikon).

On all these cameras the glass and IS system make the difference, not the sensor size.

Among all these the smaller 1 inch sensor cameras actually have an edge simply because the fast lens comes free with the camera, and the lens are tiny and weigh nothing.
It's just that f/1.8 on one inch is not faster than f/3.2 on APS-C.
You lose DoF of course but depending on your style of shooting you may like or dislike this.
DoF and noise go hand in hand - increase one and you increase the other as well. This is format independent.

Btw, it would be a good idea to use syntax f/2.8 and not f2.8 as the aperture diameter is focal length (f) divided by a number, not multiplied, nor is the aperture size a unitless number. This would also make it easier to see that if focal length of 50mm on one format and focal legth of 100mm on another create equal framing (field of view), the aperture diameter on one format would be twice/half of the other if the same f-number is used for some reason.
 
Basically it's a BS myth, same as 'lower mp cameras are better than higher mp cameras in low light because they have larger pixels which collect more light'.
used to be true

sensor got better, and the lens in front of each sensor cell evolved to capture more of the light that used to be lost due to inter-sensor cell gaps

also, look at how amazing modern sensors are in low light!! They are much more efficient, have much less noise, in using every photon that hits the sensor

iso 800 used to be marginally useful. Sometimes took a lot of pp and advanced (for their day) NR (noise reduction). NR programs used to be sold "stand alone" and everybody would constantly compare which gave the best NR. Every few months another great (for their day) advanced NR program would be released

lenses got much better also
 

Keyboard shortcuts

Back
Top