Equivalence thread (including posts moved here from other threads)

[tkbslc]

If you guys want to ignore sensor size effects, then can we throw cell phones into the mix?

 

[pellicle]

If you guys want to ignore sensor size effects, then can we throw cell phones into the mix?

wasn't that in another thread already?

 

[Phocal]

ok, I am done....................unfollowing thread before I get banned.

 

[pellicle]

Yes the size of the hole is larger on a 50mm ƒ1.2 lens then a 25mm ƒ1.2 lens, but it also has to gather and focus that light twice the distance

I think Fred summarizes this well

There's no such thing as light gathering. Light isn't gathered by a sensor. The intensity of the light hitting a photo site is measured. Larger photo sites generally have a better signal to noise ratio.

Depth of field a function f-stop and image size. Period.

For the same field of view and f-stop a "full frame" sensor has twice the image size as m43 and so has less depth of field.

Fred

 

[dalto]

take a look at this comparison.

<snip>

So even though the FF "equivalent" lens gathers 4 times more light, it has to project it onto an image circle 4 times larger as well. So, the amount of light hitting a given unit area on the sensor is the same. FF sensors are "better", because the photosites are bigger and collect more photons, thus yielding a better signal to noise ratio.

So...in practice, comparing an f2.8 lens on MFT to a f5.6 lens on FF offers the same DOF and assuming the same shutter speed requires an appropriately higher ISO on the FF side but should produce a similar image since there is less noise with the larger sensor. That is assuming the FF sensor gives you two stops of noise advantage over MFT. That assumption seems pretty reasonable comparing ISO 400 to 1600 but how reasonable is it comparing ISO 3200 to 12500? I guess it depends on the specific sensors in question.

 

[pdk42]

take a look at this comparison.
MFT: f/1.2 25mm
FF: f/1.2 50mm

The area of a perfect aperture circle.
MFT: Pi * ((25/1.2)/2)^2 = 340.9 mm^2
FF: Pi * ((50/1.2)/2)^2 ~= 1363.5 mm^2

so a FF "equivalent" lens gathers 4 times more light. However now we need to take a look at how the sensor fits into the projected image circle.

the sensor sizes are:
MFT: 17.3 mm x 13mm (4:3 Ratio)
FF: 36mm x 24mm (3:2 Ratio)

The minimum diameter of the circles that will fully inscribe the sensors is as follows.
MFT: 21.64mm
FF: 43.26mm

the area of these circles is:
MFT: 368.8 mm^2
FF: 1470.2 mm^2

So even though the FF "equivalent" lens gathers 4 times more light, it has to project it onto an image circle 4 times larger as well. So, the amount of light hitting a given unit area on the sensor is the same. FF sensors are "better", because the photosites are bigger and collect more photons, thus yielding a better signal to noise ratio.

Great post. Just a tiny elaboration...

FF sensors are "better", because the photosites are bigger and collect more photons, thus yielding a better signal to noise ratio.

To emphasise this point - yes they collect more photons, but the corollary is that each pixel well is deeper meaning that it makes more photons to fill it. This means that highlights blow less easily and that therefore darker areas can collect more photons too leading to better S/N ratio.

 

[Jonathan F/2]

Personally for me, I split my mirrorless kit using M43 for the extreme focal lengths and Sony FF for the medium range focal lengths. No point debating equivalency, just get the right tools for the job!

 

[Nathanael]

stop talking about equivalence, it's boring

 

[Speedliner]

stop talking about equivalence, it's boring

It's equivalent to dull.

 

[EarthQuake]

take a look at this comparison.
MFT: f/1.2 25mm
FF: f/1.2 50mm

The area of a perfect aperture circle.
MFT: Pi * ((25/1.2)/2)^2 = 340.9 mm^2
FF: Pi * ((50/1.2)/2)^2 ~= 1363.5 mm^2

so a FF "equivalent" lens gathers 4 times more light. However now we need to take a look at how the sensor fits into the projected image circle.

the sensor sizes are:
MFT: 17.3 mm x 13mm (4:3 Ratio)
FF: 36mm x 24mm (3:2 Ratio)

The minimum diameter of the circles that will fully inscribe the sensors is as follows.
MFT: 21.64mm
FF: 43.26mm

the area of these circles is:
MFT: 368.8 mm^2
FF: 1470.2 mm^2

So even though the FF "equivalent" lens gathers 4 times more light, it has to project it onto an image circle 4 times larger as well. So, the amount of light hitting a given unit area on the sensor is the same.

Everything up to here is perfectly accurate and valid.

FF sensors are "better", because the photosites are bigger and collect more photons, thus yielding a better signal to noise ratio.

This is absolutely incorrect. By this logic, a 43 sensor with larger photosites than a corresponding 35mm sensor would produce an image with less noise, which of course, is unequivocally false.

FF sensors perform better because they have a larger area, which collects more total photons regardless of photosite size. If the FF sensor has a 12MP or 42MP sensor makes very little difference to the quality or noise level of the final image assuming the output is equalized (that is to say, printed and viewed at the same size and distance). See Sony A7s II and A7r II for a real world example, the noise performance between these cameras is very similar outside of extremely high ISO where the A7s has slight advantage.

Another example, the A7r II vs a 12MP 43 sensor, the photosites are close in size but the A7r produces an image with much less noise. This isn't entirely fair, as there is no modern 12MP 43 sensor, but even the recent 16MP 43 sensors only have photosites a bit smaller, but the A7r handily outperforms those sensors.

Per unit (such as mm squared) light gathering, and per-pixel noise performance are largely irrelevant outside of academic discussions for the sake of academics.

 

[dalto]

First of all, I seem to have inadvertently stumbled into some long standing debate about "equivalence" without realizing it. I only frequent one other photography forum so I am out of the loop on that debate as a whole. I probably used the word "equivalent" because we often refer to lenses in terms of their 35mm equivalent focal ranges so it was the word that popped into my mind when I was typing.

The resultant image I get when I use a 25mm lens at any given combination of shutter, ISO and aperture in a MFT camera is not the same as the image I would get with a 50mm lens on my FF camera. The FF camera will have less DOF and less noise. Regardless of the technical reasons for that difference in image quality it is a real difference that shows up in my photos. That it is why I would say that a 25mm f1.4 lens on MFT should not be directly compared to a 50mm f1.4 lens on a different platform.

My FF camera and lenses are also significantly bigger, heavier and more expensive so there is a reasonable trade-off there. I think the question of does the value of MFT decrease as the cost and weight of the lenses increase is an interesting question. For me, since the primary benefit of MFT is the size and weight difference it absolutely would, especially on the wide to normal side of the equation. For others it probably doesn't. After thinking about this more I guess part of the driver would be how much you care about having a single system. 10 years ago I was all about having a single system. Today, I am more of a right tool for the job kind of guy and have multiple systems so adding another or switching back and forth isn't something I really worry about.

 

[DanS]

FF sensors perform better because they have a larger area, which collects more total photons regardless of photosite size. If the FF sensor has a 12MP or 42MP sensor makes very little difference to the quality or noise level of the final image assuming the output is equalized (that is to say, printed and viewed at the same size and distance).

Just because you scaled the noise down doesn't mean it doesn't exist. By all means do all the hand waving you want i know what I'm talking about.

 

[ijm5012]

Just because you scaled the noise down doesn't mean it doesn't exist. By all means do all the hand waving you want i know what I'm talking about.

Yeah, going by his logic an A7, A7S, and A7R all have the same high ISO performance, because they all have the same sensor architecture and size. Of course, we realize that there is a difference in ISO performance, and the A7S is a low-light beast because of how large the photo sites are due to the lower pixel count.

Pixel size/pitch plays a crucial role in determining the iso performance of a sensor, regardless of its area. M43 just performs worse because the pixel pitch is much smaller (a 16MP m43 sensor has the same pixel density as a 64MP FF sensor, something we have yet to see). The closest thing we have is Canon's 5DS/R cameras, and those are only 50MP (equivalent to a 12.5MP m43 sensor in terms of pixel density).

 

[fredlong]

Everything up to here is perfectly accurate and valid.



This is absolutely incorrect. By this logic, a 43 sensor with larger photosites than a corresponding 35mm sensor would produce an image with less noise, which of course, is unequivocally false.

FF sensors perform better because they have a larger area, which collects more total photons regardless of photosite size. If the FF sensor has a 12MP or 42MP sensor makes very little difference to the quality or noise level of the final image assuming the output is equalized (that is to say, printed and viewed at the same size and distance). See Sony A7s II and A7r II for a real world example, the noise performance between these cameras is very similar outside of extremely high ISO where the A7s has slight advantage.

Another example, the A7r II vs a 12MP 43 sensor, the photosites are close in size but the A7r produces an image with much less noise. This isn't entirely fair, as there is no modern 12MP 43 sensor, but even the recent 16MP 43 sensors only have photosites a bit smaller, but the A7r handily outperforms those sensors.

Per unit (such as mm squared) light gathering, and per-pixel noise performance are largely irrelevant outside of academic discussions for the sake of academics.

If I cut a 4/3 sensor out of the middle of a "full frame" sensor does it get noisier because the total number of photons has been reduced?

Fred

 

[pellicle]

If I cut a 4/3 sensor out of the middle of a "full frame" sensor does it get noisier because the total number of photons has been reduced?

no but it would if you reconfigured that to be a 16Megapixel segment ... (which is the point many miss when comparing FF sensors and 43rds)

I thought this was informative

Canon EOS 5D Sensor Info & Specs
12.7Mpix - 1.49 MP/cm²
Canon EOS 5D Mark II Sensor Info & Specs
21Mpix - 2.43 MP/cm²
Olympus OM-D E-M5 Sensor Info & Specs
16Mpix - 7.15 MP/cm²

 

[EarthQuake]

If I cut a 4/3 sensor out of the middle of a "full frame" sensor does it get noisier because the total number of photons has been reduced?

Fred

If you take a photo of the same subject, print it at the same size, and compare it to the full 35mm segment of the same sensor, then yes, obvious the image would be nosier.

The per-pixel noise does not change, but the total noise level of the image certainly does, which is much more important to overall image quality.

 

[EarthQuake]

Yeah, going by his logic an A7, A7S, and A7R all have the same high ISO performance, because they all have the same sensor architecture and size. Of course, we realize that there is a difference in ISO performance, and the A7S is a low-light beast because of how large the photo sites are due to the lower pixel count.

Certainly not the exact same performance, but the variance in performance is much smaller than say, the difference in performance between any of those sensors compared to a M43 sensor. You simply can't beat physics. More photons = less noise. Full stop.

 

[Turbofrog]

Certainly not the exact same performance, but the variance in performance is much smaller than say, the difference in performance between any of those sensors compared to a M43 sensor. You simply can't beat physics. More photons = less noise. Full stop.

I dunno, the A7s is about a full stop better in low light than an A7, which is only about 1 stop better than M4/3 (I.e ISO 12800 really didn't look any better than M4/3 at 6400).

Even the A7r is at least 1/2 EV better than the A7.

 

[fredlong]

no but it would if you reconfigured that to be a 16Megapixel segment ... (which is the point many miss when comparing FF sensors and 43rds)

I thought this was informative

Canon EOS 5D Sensor Info & Specs
12.7Mpix - 1.49 MP/cm²
Canon EOS 5D Mark II Sensor Info & Specs
21Mpix - 2.43 MP/cm²
Olympus OM-D E-M5 Sensor Info & Specs
16Mpix - 7.15 MP/cm²

I'm not sure what you mean by reconfiguring. I was responding to EarthQauke's idea that photo site size doesn't matter, but sensor size does. By "cutting out" I mean a smaller sensor with the same photo site size and same underlying technology, both hardware and software. I believe, and I could be wrong, that the signal to noise ratio would be the same.

Fred

 

[fredlong]

If you take a photo of the same subject, print it at the same size, and compare it to the full 35mm segment of the same sensor, then yes, obvious the image would be nosier.

The per-pixel noise does not change, but the total noise level of the image certainly does, which is much more important to overall image quality.

So cropping increases noise?

Fred