sensor dictates 'actual' f-stop?

ZoneV wrote in post #14984577
WIth that headline "sensor dictates 'actual' f-stop" I thought about another topic - the microlens problem with the faster lenses. There the different sensors reduce the gathered light from fast lenses due to the angular responivity of the microlenses on the sensor.

Whatever difference the microlenses make from body to body will be a matter of the design of the lens and dependent on the focal length, not the aperture. There won't be a difference between a 50mm f/1.8 and a 50mm f/1.4.

mannetti21 wrote in post #14985218
I guess my overall point is simply that for an individual who is completely new to photography, I can see how they could be easily confused into thinking that, in terms of light gathering, an f4 lens will magically become f2.8 as soon as they attach it to any FF sensor.

I actually see the opposite happening, where people are all excited about lenses for small formats that seem to be better than they are. The best example of this is the Olympus 35-100 1:2 zoom.

People see that f/2 zoom and get all excited. But they ignore the fact that this f/2 zoom can only be used on the 4:3 format sensor.

In terms of what I can do with a 35-100 f/2 zoom on 4:3, it will perform almost exactly like a 70-200 f/4 zoom on a 5D Mark III.

Again, this is because 35-100 gives the same field of view on 4:3 as 70-200 on FF

f/2 on 4:3 gives the same DOF on 4:3 as f/4 does on FF.

And I will certainly be willing to use a 5D Mark III camera at about two stops higher ISO than any same generation 4:3 sensor body.

I know what you are saying about the light transmission through the lens being equal, but that's ignoring the elephant in the room. When I compare equal generation cameras for high ISO noise, the bigger sensors let me go higher.

Phrase it how you like, but thinking in terms of effective focal length and effective aperture is, for me, an easy way to think about lenses. For sure having that Oly 35-100 f/2 is not going to be as awesome as a mythical 70-200 f/2 would be on a FF camera. f/2 on the smaller format just isn't the same.

My direct experience in this BTW came when I moved from my 30D to a 5D several years ago. I found that not only was I shooting longer focal lengths, I was also using smaller apertures.

It was probably my post OP was confused about. The confusion was not because I was wrong, but because I wasn't descriptive enough.

Here is the important assumption:

- Both crop and FF sensor are of the same generation, and have the same number of pixels. If so, crop sensor has roughly the same noise as FF sensor at 1.6^2=2.56x higher ISO value, with the same sensor illumination.

It is true that f/2.8 lens produces as much illumination on crop sensor as it does on FF sensor. But because one can use ~2.6x higher ISO value on FF sensor to get the same noise level, one can have 2.6x lower sensor illumination (with a f/2.8/1.6 = f/4.5 lens) + 2.6x higher ISO on FF camera to get a similar exposure and amount of noise.

So it is not only FoV and DoF equivalence between 17-55 f/2.8 and 24-105 f/4 - as long as the above assumption is (roughly) true, it's also equivalence in terms of noise in the image, for the same exposure.

JeffreyG wrote in post #14985675
I actually see the opposite happening, where people are all excited about lenses for small formats that seem to be better than they are. The best example of this is the Olympus 35-100 1:2 zoom.

People see that f/2 zoom and get all excited. But they ignore the fact that this f/2 zoom can only be used on the 4:3 format sensor.

In terms of what I can do with a 35-100 f/2 zoom on 4:3, it will perform almost exactly like a 70-200 f/4 zoom on a 5D Mark III.

Again, this is because 35-100 gives the same field of view on 4:3 as 70-200 on FF

f/2 on 4:3 gives the same DOF on 4:3 as f/4 does on FF.

And I will certainly be willing to use a 5D Mark III camera at about two stops higher ISO than any same generation 4:3 sensor body.

I know what you are saying about the light transmission through the lens being equal, but that's ignoring the elephant in the room. When I compare equal generation cameras for high ISO noise, the bigger sensors let me go higher.

Phrase it how you like, but thinking in terms of effective focal length and effective aperture is, for me, an easy way to think about lenses. For sure having that Oly 35-100 f/2 is not going to be as awesome as a mythical 70-200 f/2 would be on a FF camera. f/2 on the smaller format just isn't the same.

My direct experience in this BTW came when I moved from my 30D to a 5D several years ago. I found that not only was I shooting longer focal lengths, I was also using smaller apertures.

I believe we are on the same page. At the risk of further complicating this thread, ill say that while you can compensate for light with modern FF sensors, the f4 is still a limitation in comparison to the f2.8. Reason being, I got the 5d3 to use the ISO capabilities to "see in the dark", now, using an f4 I'm only able to see with a night-light on. So I guess I could say I'm crippling my 5D3. At some point, probably years from now, there will be a modern crop sensor that outperforms my FF sensor...then, there is no way to compensate for f4, relative to the new crop with f2.8 .

But regardless, this has essentially become a discussion of semantics.

mannetti21 wrote in post #14985767
I believe we are on the same page. At the risk of further complicating this thread, ill say that while you can compensate for light with modern FF sensors, the f4 is still a limitation in comparison to the f2.8. Reason being, I got the 5d3 to use the ISO capabilities to "see in the dark", now, using an f4 I'm only able to see with a night-light on. So I guess I could say I'm crippling my 5D3. At some point, probably years from now, there will be a modern crop sensor that outperforms my FF sensor...then, there is no way to compensate for f4, relative to the new crop with f2.8 .

But regardless, this has essentially become a discussion of semantics.

You just have to keep upgrading the FF body every few years at $3500 a go and the problem is solved.

SkipD wrote in post #14985367
In fact, there is absolutely no such thing as an "effective focal length". That term suggests that there is some camera format which is the end-all standard against which all other camera formats in the world are measured against and that simply is not so.

The 35mm film format (24mm by 36mm) is not, never was, and never will be such a universal standard. It is simply the format of the cameras that are the approximate physical size that today's DSLRs were modeled after so that the existing lenses for the 35mm film cameras could be used on them.

Thus, using the terms "effective focal length" and "effective aperture" are totally meaningless unless the comparison details are explained along with the terms.

I agree with what you are saying in theory, however I believe it has been well stated in this thread that we are talking about full frame digital bodies and 1.6 digital bodies only.

SkipD wrote in post #14985367
Further, the "effective aperture" thing is quite silly in my opinion, because that term does not indicate that the only factor being considered in the "effective" part is depth of field. There is absolutely no difference in exposure settings when a lens is moved from one format camera to another.

Actually the DOF is not the only factor Jeffrey and now Pulsar123 are talking about. We also have to consider the noise capabilities of the sensor and the physical size of the pixels. Larger pixels see a larger flux of protons in a given exposure than a higher density crop sensor and thus noise has a smaller effect on the output of each pixel. Also since the larger pixels in a same generation full frame camera will have to be enlarged less for the same size output print than a 1.6 sensor, the noise will be enlarged less and be less visible.

In order to make this statement, we need to be clear, to Mannetti's point, on exactly which bodies we are comparing and that is one thing that people, myself included, should be better about doing. Clearly, we can't compare a 30D to a 5D2 in this argument because they have the size pixel size. Also with the new Nikon D800 with a relatively high density sensor compared to their 1.5 crop offerings of the same generation, even stating the same generation is getting to be too large of a generalization.

pulsar123 wrote in post #14985704
- Both crop and FF sensor are of the same generation, and have the same number of pixels. If so, crop sensor has roughly the same noise as FF sensor at 1.6^2=2.56x higher ISO value, with the same sensor illumination.

One thing to add here though is that this is in theory because there is no crop and FF sensor of the same generation with the same pixel count at least in Canon land. For this calulation to be real world, we would need to correct the numbers to reflect the 18mp and 21mp (5d2) and 22mp (5D3)

You're using the term "semantics" incorrectly, as though it were a synonym for "irrelevancies." It's not. A semantical distiction is important, like the difference between carbon dioxide and carbon monoxide or the difference between a bull and a cow or the difference between an iron and a wedge.

And this is a case of the point. It's important to know what the lens controls and what it does not control, and as well what each other component controls so that one knows the effects of changing. This is especially true when the same lens might be used on bodies that differ vastly in their own image recording and processing capabilities. You have to know what the lens itself really does.

Unless this is a discussion among the shallowest of dilittantes, of course. A kid from Manhattan probably doesn't need to know the difference between a cow and a bull.

Well, it can be mounted on a 24x36mm camera, and it will perform precisely the same...but only within the format of 4:3. The point being: The lens does what it does regardless of what's mounted behind it. It never handles light any differently.

pulsar123 wrote in post #14985704
It was probably my post OP was confused about. The confusion was not because I was wrong, but because I wasn't descriptive enough.

Here is the important assumption:

- Both crop and FF sensor are of the same generation, and have the same number of pixels. If so, crop sensor has roughly the same noise as FF sensor at 1.6^2=2.56x higher ISO value, with the same sensor illumination.

It is true that f/2.8 lens produces as much illumination on crop sensor as it does on FF sensor. But because one can use ~2.6x higher ISO value on FF sensor to get the same noise level, one can have 2.6x lower sensor illumination (with a f/2.8/1.6 = f/4.5 lens) + 2.6x higher ISO on FF camera to get a similar exposure and amount of noise.

So it is not only FoV and DoF equivalence between 17-55 f/2.8 and 24-105 f/4 - as long as the above assumption is (roughly) true, it's also equivalence in terms of noise in the image, for the same exposure.

I think this is an incorrect assumption, in most of these type of discussions. There are many, many, user's on this forum, using perfectly excellent 5Dc's(and even some still using 35mm film), and so.. this makes a 24-105 f/4L IS a 24-105 f/1.4L IS(simply due to the 5D3 being so many [whatever it is.. 2/3/4?] stops cleaner at high ISO)?

No! This is a huge generalization!

pulsar123 wrote in post #14985704
It was probably my post OP was confused about. The confusion was not because I was wrong, but because I wasn't descriptive enough.

Here is the important assumption:

- Both crop and FF sensor are of the same generation, and have the same number of pixels. If so, crop sensor has roughly the same noise as FF sensor at 1.6^2=2.56x higher ISO value, with the same sensor illumination.

It is true that f/2.8 lens produces as much illumination on crop sensor as it does on FF sensor. But because one can use ~2.6x higher ISO value on FF sensor to get the same noise level, one can have 2.6x lower sensor illumination (with a f/2.8/1.6 = f/4.5 lens) + 2.6x higher ISO on FF camera to get a similar exposure and amount of noise.

So it is not only FoV and DoF equivalence between 17-55 f/2.8 and 24-105 f/4 - as long as the above assumption is (roughly) true, it's also equivalence in terms of noise in the image, for the same exposure.

1Tanker wrote in post #14987415
I think this is an incorrect assumption, in most of these type of discussions. There are many, many, user's on this forum, using perfectly excellent 5Dc's(and even some still using 35mm film), and so.. this makes a 24-105 f/4L IS a 24-105 f/1.4L IS(simply due to the 5D3 being so many [whatever it is.. 2/3/4?] stops cleaner at high ISO)?

No! This is a huge generalization!

Beyond that, Tanker, it's simply a load of bovine excrement.