[Ben_Edict]

This is wrong, as the APS-C sensor is smaller, less than half the light is captured by this sensor compared to 135 film or FF sensor. Light per area unit will be the same for both, but the sensor/film with larger area will collect more light.

But if using a converter to make a FF lens to APS-C lens, so the APS-C sensor can collect all the light from the lens. Then light per area unit will be higher on the APS-C sensor. Shutter speed will then be faster on the APS-C camera if using the same aperture and ISO as on a FF camera.

The captured photon per sensor area is what decides the exposure level. So, there is not the slightrest need for these "equivalency" thoughts. They are not useful for anything. I think, the whole equivalency discussion is based on theoretical thinking about incoming light rays from point sources (and its applicability is also limited to point sources). There is only one application for this (and there this discussion is very old news indeed): astrophotography of star fields (which are nothing but point sources). This is a different world than that of photographing illuminated areas, with which we have to deal.

Your thoughts about concentrating all the incoming light of the FF lens onto the smaller sensor are, in my opinion, strange. What light do you want to concentrate? If you want to capture all of the incoming light of a FF lens ontp the smaller sensor, you would need a kind of tele converter, which at the same time, will increase total focal length and thus reduce the effective aperture. So at the end, you haven't gained anything.

Ben

[Ash]

Ben, I believe the thought stemmed from the possibility of creating a reverse teleconverter - allowing FF lenses to be used on APS-C format, only with the ability to encapsulate the complete FF image circle concentrated onto an APS-C sensor, thereby increasing incident light intensity per sensor area. This effectively makes the FF lens 'faster' than it innately is. I mentioned earlier that I found little practical utility in this as extra glass naturally implied a degradation in IQ - we may as well just have DA lenses for APS-C created with larger apertures (if that is the intention of having such a reverse TC), which means larger lenses (not what Pentax would be wanting to create anyway).

To me, let FF lenses be put on a FF camera (we're waiting, Pentax), and finally we can have a camera that will use the full image circle from the lenses and put an end to this madness!

[Blue]

It's kind of obvious that the lens do not physically change depending camera it is used on, and let the same amount in on all cameras, but the sensor size change how much of the light is used for exposure.

Or do you think that if you put one small and one large bucket out in the rain, that both of them will collect the same amount of water for a given time?

If it is obvious that a FA 77/1.8 lens on 135 is also a FA 77/1.8 lens on aps-c with exception of the FOV, then why to you keep flailing? All that I have been saying is that the focal length of a lens and aperture are physical measurements. Furthermore, rain is measured in inches or millimeters. The entire image from an FA77/1.8 image is not on the aps-c sensor, hence part is essentially cropped out. Go get yourself a roll of 100 ISO 135 film and put your camera on a tripod and take a picture of the same subject at the same distance with an FA 77 lens at F 5.6 and also an a K-5 and you will see that there is more area around the subject in the exposure from the full frame film camera. Where did that left out area go in the aps-c body?

Edit: It also depend on how long you leave those buckets out in the rain. When we have on of our tropical storms here in Florida, those buckets will overflow.

[BrianR]

Or do you think that if you put one small and one large bucket out in the rain, that both of them will collect the same amount of water for a given time?

That depends, are you measuring "amount of water" by the volume of the water in the buckets or by the height of the water in the buckets?

I'm not just being a smart ass here, so many of the disagreements in these kinds of discussions stem from people not being clear about the assumptions they're making and having (unstated) different meanings for "amount of water" or "amount of light".

[Fogel70]

Your thoughts about concentrating all the incoming light of the FF lens onto the smaller sensor are, in my opinion, strange. What light do you want to concentrate? If you want to capture all of the incoming light of a FF lens ontp the smaller sensor, you would need a kind of tele converter, which at the same time, will increase total focal length and thus reduce the effective aperture. So at the end, you haven't gained anything.

Ben

It has to be a "reversed TC" that increase FOV, FI a 0.67x converter that turn a 50/1.4 FF lens into a 33/1.0 APS-C lens. A TC will decrease the concentration of the light so it covers a bigger area.

[Fogel70]

That depends, are you measuring "amount of water" by the volume of the water in the buckets or by the height of the water in the buckets?

I'm not just being a smart ass here, so many of the disagreements in these kinds of discussions stem from people not being clear about the assumptions they're making and having (unstated) different meanings for "amount of water" or "amount of light".

Amount of water is the total amount of rain drops the bucket will collect, just like the amount of light is the number of photons the sensor will capture.
Height of water is not amount of water it's the "intensity" of the rain, and will be the same on both buckets, just "intensity" of light is on different sized sensors.

[Blue]

Amount of water is the total amount of rain drops the bucket will collect, just like the amount of light is the number of photons the sensor will capture.
Height of water is not amount of water it's the "intensity" of the rain, and will be the same on both buckets, just "intensity" of light is on different sized sensors.

Rain is measured in inches or millimeters (or feet and meters). Go check a meteorology site. Otherwise, you are talking about how much water a bucket will hold. A 5 gal bucket will hold 5 gal whereas a 3 gal bucket will hold 3 gal. However, if they are out in the rain for an hour . . . that would be analogous to the exposure. In the camera, exposure is determined by aperture, shutter speed and sensor/film sensitivity. Therefore, I guess an overflowing bucket would by analogous to an over exposure. The examples I have used, I kept the sensitivity or ISO the same, ISO 100 for 135 film and aps-c sensor. Since the buckets are in the same rain even next to each other, the intensity would be closely equal.

[BrianR]

Amount of water is the total amount of rain drops the bucket will collect, just like the amount of light is the number of photons the sensor will capture.
Height of water is not amount of water it's the "intensity" of the rain, and will be the same on both buckets, just "intensity" of light is on different sized sensors.

How much rain did you get last May in Stockholm? Do you measure it in trillions of Liters or in mm?

"Amount if light" or "Amount of water" is vague, and is open to interpretation depending how the reader wants to measure "Amount" in either case and an assumption that your version is somehow the only 'correct' way is very optimistic and bound to lead to misunderstandings. Taking the time to write a couple more words can clear these ambiguities. This is especially important on an international forum filled with people from many different backgrounds. A similar thing happens in DoF discussions all the time, look at posts #12, and #14 of this thread, both were absolutely right but based on different conditions that were cleared up in posts #17 and #18.

[Ikarus]

The behavior is the opposite of what you suggested in your earlier post:
It reaches "Excellent" in each case, using Photozone's criteria for comparison.

They say that "The Nikon AF-S 28/1.8 struggles a bit on the high density CX sensor of the V1 test camera. The resolution in the image center is only good wide open", whereas on the large sensor, it says "Sharpness is very good across the frame wide open already". I'm not sure how they can still both end up rated ''excellent" in the center.

[ElJamoquio]

The captured photon per sensor area is what decides the exposure level.

Yes.

So, there is not the slightrest need for these "equivalency" thoughts. They are not useful for anything.

Equivalency discussions are useful only when taking pictures of things.

[ElJamoquio]

If it is obvious that a FA 77/1.8 lens on 135 is also a FA 77/1.8 lens on aps-c with exception of the FOV, then why to you keep flailing? All that I have been saying is that the focal length of a lens and aperture are physical measurements.

No one is arguing against that. 'f/2.8' is always 'f/2.8' in terms of exposure.

'f/2.8' requires sensor size to figure out DOF.

Saying 'f/2.8 is always f/2.8' is strictly true but leaves out pertinent information to the uninitiated.

It's also perhaps a bit confusing when the aperture is the same size, but the focal length of a lens changes (as with a T/C). I don't care what vocabulary is used but not everyone is using the same vocabulary, it seems.

[Lowell Goudge]

Ben, I believe the thought stemmed from the possibility of creating a reverse teleconverter - allowing FF lenses to be used on APS-C format, only with the ability to encapsulate the complete FF image circle concentrated onto an APS-C sensor, thereby increasing incident light intensity per sensor area. This effectively makes the FF lens 'faster' than it innately is. I mentioned earlier that I found little practical utility in this as extra glass naturally implied a degradation in IQ - we may as well just have DA lenses for APS-C created with larger apertures (if that is the intention of having such a reverse TC), which means larger lenses (not what Pentax would be wanting to create anyway).

To me, let FF lenses be put on a FF camera (we're waiting, Pentax), and finally we can have a camera that will use the full image circle from the lenses and put an end to this madness!

Unfortunately what you describe is not possible. In its simplest form aperture is not related to image circle at all, no extra light that comes into a lens can be recovered and used to increase brightness of image. Aperture especially on a tele lens is the focal length divided by front element diameter. Just look at a K300/4 and the DA300/4. Both use 77mm filters. The bigger the front element the more light gathered. That's all. To put elements behind the lens, and put more light on the image would require a retro focus element / group, reducing the focal length.

You can also achieve this in a limited manner with a close up lens. These effectively shorten the focal length of the lens, and since the diameter is not changing the F-number gets smaller (bigger aperture)

[Blue]

No one is arguing against that. 'f/2.8' is always 'f/2.8' in terms of exposure.

'f/2.8' requires sensor size to figure out DOF.

Saying 'f/2.8 is always f/2.8' is strictly true but leaves out pertinent information to the uninitiated.

It's also perhaps a bit confusing when the aperture is the same size, but the focal length of a lens changes (as with a T/C). I don't care what vocabulary is used but not everyone is using the same vocabulary, it seems.

Fogel hasn't been debating depth of field with me. I have been clear from the beginning that I have been talking about exposure which comes down to shutter speed, aperture and film/sensor sensitivity. Furthermore, I also the FOV and angle of view difference is due to the sensor size. That is basic photography. Where things get murky is when people want to start using equivalency in terms of aperture relative to depth of field which misleads folks into think it effects exposure.

[lytrytyr]

They say that "The Nikon AF-S 28/1.8 struggles a bit on the high density CX sensor of the V1 test camera. The resolution in the image center is only good wide open", whereas on the large sensor, it says "Sharpness is very good across the frame wide open already". I'm not sure how they can still both end up rated ''excellent" in the center.

In the MTF50 chart on CX, center performance is "excellent" from f/4 to f/8.
In the MTF50 chart on FF, center performance is "excellent" from f/2.8 to f/8.

The criterion for "excellence" is 1740 lw/ph on CX and 3460 lw/ph on FF.

[ElJamoquio]

'Excellent' is a sliding scale based on the sensor size. Lenses can only do so much, after all. The lens resolves the same on a lp/mm basis, but the mm is so much higher on the FF sensor, that on a lp per picture height basis, excellent on a CX is on par with 'poor' or so on FF.