Defining a fast lens - B

From that "Great Bustard" (? spelling of that avian noun?) in the previous filled thread of the same name. - "And the aperture diameter for 100mm f/4 is 100mm / 4 = 25mm, and the aperture diameter for 50mm f/2 is 50mm / 2 = 25mm. In other words, f/2 on 4/3 gives the same aperture diameter as f/4 on FF for the same AOV. "

In both cases as you state, the diameter is 25 mm. Where you go wrong is that you forget or don't know that the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies. This is the same as a guide number for flash, in that double the distance from the light source, and the lens has to be opened up two stops to give correct exposure.

I have both a 35 mm lens and a 350 mm lens. At (for convenience) I set the 350 at f10, the aperture is 350/10 = 35mm aperture. On your faulty logic, it is therefore equivalent in light-gathering power at the sensor of my 35 mm lens at f1.0. The 35 mm is f2.8 maximum.

Please go back to arithmetic classes.

John.

... the egg from which he hatched lacked legal legitimacy. In any case, he does tend to peck at things that aren't moving (much).

EJohnE said:

From that "Great Bustard" (? spelling of that avian noun?) in the previous filled thread of the same name.

Click to expand...

EJohnE said:

From that "Great Bustard" (? spelling of that avian noun?) in the previous filled thread of the same name. - "And the aperture diameter for 100mm f/4 is 100mm / 4 = 25mm, and the aperture diameter for 50mm f/2 is 50mm / 2 = 25mm. In other words, f/2 on 4/3 gives the same aperture diameter as f/4 on FF for the same AOV. "

In both cases as you state, the diameter is 25 mm. Where you go wrong is that you forget or don't know that the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies. This is the same as a guide number for flash, in that double the distance from the light source, and the lens has to be opened up two stops to give correct exposure.

I have both a 35 mm lens and a 350 mm lens. At (for convenience) I set the 350 at f10, the aperture is 350/10 = 35mm aperture. On your faulty logic, it is therefore equivalent in light-gathering power at the sensor of my 35 mm lens at f1.0. The 35 mm is f2.8 maximum.

Please go back to arithmetic classes.

Click to expand...

...that people have to result to insults because they think they don't understand what's going on when they don't have a clue. Anyone else want to sign their name to John K's post so I can know who you all are in one shot? I know at least one other John that's banging on the door.

Here's the answer: a pic of the same scene from the same position with the same framing, the same aperture diameter , and same shutter speed will project the same total amount of light on the sensor (with a small niggle for different aspect ratios, which amounts to but a 4% difference -- 0.06 stops -- in the case of 4:3 and 3:2).

For example, a photo of the same scene from the same position at 50mm f/2 1/100 on 4/3 and 100mm f/4 1/100 on FF will project the exact same amount of light on the sensor (minus the 4% "niggle"). Hence, if the sensors have the same tech, then the noise will also be the same.

Here's an explanation of how it all works:

http://www.josephjamesphotography.com/equivalence/#fratio

And for those that are wondering why I "mysteriously" left out the whole bit about the ISO differential between the systems, well, it's because, for the most part, it's a non-issue:

http://www.josephjamesphotography.com/equivalence/#iso

Anyway, I've been through posturing threads like this before:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=35404091

So John, cut the attitude. If you don't understand something I've said, I'm happy to explain it in more detail.

Great Bustard said:

Here's the answer: a pic of the same scene from the same position with the same framing, the same aperture diameter , and same shutter speed will project the same total amount of light on the sensor (with a small niggle for different aspect ratios, which amounts to but a 4% difference -- 0.06 stops -- in the case of 4:3 and 3:2).

For example, a photo of the same scene from the same position at 50mm f/2 1/100 on 4/3 and 100mm f/4 1/100 on FF will project the exact same amount of light on the sensor (minus the 4% "niggle"). Hence, if the sensors have the same tech, then the noise will also be the same.

Here's an explanation of how it all works:

http://www.josephjamesphotography.com/equivalence/#fratio

And for those that are wondering why I "mysteriously" left out the whole bit about the ISO differential between the systems, well, it's because, for the most part, it's a non-issue:

Click to expand...

Why am I foolishly wading into one of these debates? Must be mad! Anyway, I agree, in the example you give, both sensors will get the same amount of light. Unfortunately (as you hint), given the same base sensitivity, the 35mm sensor will need either four times as much light, or for the signal to be amplified by two stops, to give the same exposure on the final image.

In theory, that amplified signal will be similar, in terms of noise, to the un-amplified signal of the smaller sensor but that depends very much on which cameras you are comparing. In practice, I like to shoot at the cameras best sensitivity and consider increasing that (for example because I want to use a slower lens or I want to get sufficient depth of field) a major compromise.

It's the theory that lends to endless internet debates! What fun. In practice, I note successive generations of technology seem to have a larger effect than sensor size...
--
Don.

A Land Rover, a camera ... I'm happy!

deep7 said:

Why am I foolishly wading into one of these debates? Must be mad!

Click to expand...

You should see a "qualified psychologist".

deep7 said:

Anyway, I agree, in the example you give, both sensors will get the same amount of light.

Click to expand...

You're off to a great start!

deep7 said:

Unfortunately (as you hint), given the same base sensitivity, the 35mm sensor will need either four times as much light, or for the signal to be amplified by two stops, to give the same exposure on the final image.

Click to expand...

Here's the thing -- the exposure has nothing to do with sensitivity. The sensitivity of a digital sensor is fixed. As the link in my post above states:

http://www.josephjamesphotography.com/equivalence/#iso

Unlike film, the sensitivity of the sensor is fixed. The effect of the ISO is merely to apply an analog gain to the signal that, for some sensors, is a bit more efficient than a digital push (applied either in camera or in post) and results in less noise (at the possible expense of blowing more highlights).

Interestingly, the new Sony sensor in the D7000 and K5 is about to make ISO meaningless:

If higher ISOs were not more efficient than base ISO, then there would be no point in having higher ISOs. The camera would only have base ISO, and we would simply choose the appropriate f-ratio for the DOF and/or sharpness we desire, and the appropriate shutter speed to account for motion blur and/or camera shake. The exposure meter would tell us how over/under exposed we are (less exposure means more noise, more exposure means more blown highlights), at which point the photographer could adjust the balance of the f-ratio and shutter speed with the exposure. We would then push the image digitally to whatever brightness we desired.

So, going back to your statement:

the 35mm sensor will need either four times as much light, or for the signal to be amplified by two stops, to give the same exposure on the final image.

what we see is that exposure is really quite irrelevant when comparing different formats -- what is relevant is the total light collected. In the case of comparing the same format, exposure and total light are effectively the same measure, but very different measures on different formats.

It's not unlike mass and weight -- effectively the same thing in the same acceleration field, but rather different measures in different acceleration fields.

deep7 said:

In theory, that amplified signal will be similar, in terms of noise, to the un-amplified signal of the smaller sensor but that depends very much on which cameras you are comparing.

Click to expand...

Indeed. In the case of the Canon 5D, the difference is huge :

http://www.pbase.com/joemama/pushediso

In the case of the D7000 and K5, the difference is negligible, if even there at all:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=37235114

deep7 said:

In practice, I like to shoot at the cameras best sensitivity and consider increasing that (for example because I want to use a slower lens or I want to get sufficient depth of field) a major compromise.

Click to expand...

We want to gather as much light as possible so as to have the least noise, but to do this, we must either use a longer shutter speed which puts us at greater risk for motion blur and/or camera shake, or use a wider aperture, which results in a more shallow DOF.

For equivalent photos (same perspective, framing, aperture diameter, shutter speed, and display size), all systems will have the same DOF and also gather the same amount of light (thus have the same noise for the same sensor tech).

deep7 said:

It's the theory that lends to endless internet debates! What fun.

Click to expand...

It does at that. However, I am perplexed by why some find the concepts so difficult to understand, and astounded when people actually argue against them.

deep7 said:

In practice, I note successive generations of technology seem to have a larger effect than sensor size...

Click to expand...

That's going to level off, and quickly. The Canon 5D has a QE of 25%, and the D3s has a QE of 57% (one stop better). There's only a stop more improvement to be made until we hit 100% QE (with a Bayer CFA).

However, the pixels on the K5 and D7000 have about the same FWS as the pixels on the 5D, despite only being 40% the size, and have scaled the read noise approximately with that area differential. This is what gives them their massive DR improvement.

So, another stop more to go for QE, but there's no real limit to DR if they can keep increasing the FWS.

Regardless, as always, for cameras with the same sensor tech, the noise will be the same for Equivalent photos. The larger formats will achieve an IQ advantage when they can use a longer shutter speed to gather more light (which presumes neither motion blur nor camera shake are an issue), or use a larger aperture (which presumes that a more shallow DOF is preferable, or, at least, preferable to a more noisy photo with a deeper DOF).

If you're photography does not allow for either of these situations, then the larger format is likely not the best choice.

EJohnE said:

In both cases as you state, the diameter is 25 mm. Where you go wrong is that you forget or don't know that the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies.

Click to expand...

Before accusing anyone of going wrong, you should put some thought into (a) what lies behind the inverse square law and (b) why you think some of the photons that enter through the entrance pupil won't reach the exit pupil just because the focal length was increased.
--
Rikke

EJohnE said:

From that "Great Bustard" (? spelling of that avian noun?) in the previous filled thread of the same name. - "And the aperture diameter for 100mm f/4 is 100mm / 4 = 25mm, and the aperture diameter for 50mm f/2 is 50mm / 2 = 25mm. In other words, f/2 on 4/3 gives the same aperture diameter as f/4 on FF for the same AOV. "

In both cases as you state, the diameter is 25 mm. Where you go wrong is that you forget or don't know that the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies. This is the same as a guide number for flash, in that double the distance from the light source, and the lens has to be opened up two stops to give correct exposure.

Click to expand...

As you might know, I quite like thought experiments. Here's a little one for you. An EC20 2x Teleconverter is 41mm long. If you put this behind a 300/2.8 lens you get a 600/5.6 lens. So, if I'm understanding you correctly, in that 41mm the light travels an extra 300mm. Is that it?
--
Bob

I assume that most here have no or little experience of cameras without auto aids for exposure, focus etc. The digital camera can take care of most of the technical aspects. It is possible that anything written with apparent authority will be taken uncritically as gospel by the digital only, auto-all photographer. I consider it unethical for a person to quote “authoritative, expert, independent” articles purporting to confirm in the trusting reader his pontifications, without clearly showing that he is the author of both.

I suggest that you examine a hand-held photographic lightmeter. You would then notice that the only needed variable is the adjustment of the ISO/ASA speed to match the sensor/film setting at which you wish to shoot. Most meters will then give a full range of shutter/aperture combinations each giving the same amount of light at the capture device – whether sensor or film, when the needles are aligned, or whatever method is used to set the reading. How can that meter in incident light mode (whereby one measures the light falling on the subject, rather than reflected from it), give an exposure correct for any format size? As denied by the more bird-brained: f2 is f2 is f2. If the camera can been set to auto-ISO, and the desired shutter speed to e.g. 1/100 sec., then the two stops difference can be nullified by the ISO being reset by the camera to account for that. e.g. 1/100 @f2 – 100 ISO, 1/100 @ f4 – 400 ISO. The exif details would show that. A not really possible light reading from each lens 50mm from the diaphragm should be near-equal, but only the 50mm image will be in focus. If the small 4/3rd format sensor requires f2.0 at 50mm, whereas the 35 mm sensor requires only f4.0 at 100mm (the same aperture diameter of 25mm in each case) at the same shutter speed and ISO setting to allow the same amount of light to fall on their respective sensors, why is there no adjustment on the lightmeter for differing sensor sizes to account for this “anomaly”?

ISO stands for International Standards Organisation. There are ISO standards for just about anything. Standard repeatable testing is adhered to by any manufacturing member of the ISO, so that one can be assured of reasonable consistency within any standard. Film does have a fixed ISO rating, as does a sensor. Both may be modified by less/more exposure, and compensating processing. Developing for film, signal amplification for sensor. For this great genius to write “ We want to gather as much light as possible so as to have the least noise.....” is nonsense. We want to gather the correct amount of light for that photograph. A great amount of light could be gathered if exposed at f2 and 30 seconds at 6400 ISO on a bright day. Is that what is meant?

bobbin2 asks if I imagine his converter stretches the lens physically to 600 from 300. No, but he answers his own question by describing the EC20 teleconverter as 2x. That is an optical magnifier of the 300's image size. By design, all telephoto lenses are shorter than their focal length: all retrofocus lenses are longer than their focal length. One practical reason for the end-magnification of the teles is to make them more compact. My 135mm SLR lens is about 100mm from nodal point area to the film/sensor. The 7-14mm lens would be impossible if the 7mm setting was actually 7mm from the sensor. It has corrected reducing components or whatever clever solution is required to produce that angle of view from a greater actual lens distance.

Rikki Rask wants to know why I think some photons wont reach the exit pupil. They are outside the sensor's frame area. Where do photons go when after looking at something, I shut my eyes? Perhaps as an old song asks, do they go where the flies go in the winter?

EJohnE said:

I assume that most here have no or little experience of cameras without auto aids for exposure, focus etc. The digital camera can take care of most of the technical aspects. It is possible that anything written with apparent authority will be taken uncritically as gospel by the digital only, auto-all photographer. I consider it unethical for a person to quote “authoritative, expert, independent” articles purporting to confirm in the trusting reader his pontifications, without clearly showing that he is the author of both.

Click to expand...

Consider as you please. The same amount of light falls on the sensor for the same scene, perspective, framing, aperture diameter, and shutter speed (ISO has nothing to do with it).

For example, 50mm f/2 1/200 on 4/3 puts the same amount of light on the sensor as 100mm f/4 1/200 on FF (minus the 4% -- 0.06 stop -- "niggle" due to the slight difference in framing between 4:3 and 3:2).

EJohnE said:

I suggest that you examine a hand-held photographic lightmeter. You would then notice that the only needed variable is the adjustment of the ISO/ASA speed to match the sensor/film setting at which you wish to shoot. Most meters will then give a full range of shutter/aperture combinations each giving the same amount of light at the capture device – whether sensor or film, when the needles are aligned, or whatever method is used to set the reading. How can that meter in incident light mode (whereby one measures the light falling on the subject, rather than reflected from it), give an exposure correct for any format size? As denied by the more bird-brained: f2 is f2 is f2.

Click to expand...

It would be sad to be less intelligent than a bird and fail to make the distinction between the density of the light (exposure) and the total amount of light. The noise in the photo is a function of the total amount of light and the sensor efficiency -- the density of the light (exposure) has nothing to do with it, except inasmuch as the total amount of light can be calculated as the product of the exposure and the sensor area.

EJohnE said:

If the camera can been set to auto-ISO, and the desired shutter speed to e.g. 1/100 sec., then the two stops difference can be nullified by the ISO being reset by the camera to account for that. e.g. 1/100 @f2 – 100 ISO, 1/100 @ f4 – 400 ISO. The exif details would show that. A not really possible light reading from each lens 50mm from the diaphragm should be near-equal, but only the 50mm image will be in focus. If the small 4/3rd format sensor requires f2.0 at 50mm, whereas the 35 mm sensor requires only f4.0 at 100mm (the same aperture diameter of 25mm in each case) at the same shutter speed and ISO setting to allow the same amount of light to fall on their respective sensors, why is there no adjustment on the lightmeter for differing sensor sizes to account for this “anomaly”?

Click to expand...

Because you don't understand what's going on. To get a noise advantage with a larger sensor system, it must gather more light. The only way to do that is to use a larger aperture diameter (which will result in a more shallow DOF) or a longer shutter speed (which will put the photo at greater risk for motion blur and/or camera shake).

EJohnE said:

ISO stands for International Standards Organisation. There are ISO standards for just about anything. Standard repeatable testing is adhered to by any manufacturing member of the ISO, so that one can be assured of reasonable consistency within any standard. Film does have a fixed ISO rating, as does a sensor. Both may be modified by less/more exposure, and compensating processing. Developing for film, signal amplification for sensor. For this great genius to write “ We want to gather as much light as possible so as to have the least noise.....” is nonsense. We want to gather the correct amount of light for that photograph. A great amount of light could be gathered if exposed at f2 and 30 seconds at 6400 ISO on a bright day. Is that what is meant?

Click to expand...

ISO has nothing to do with it. Changing the ISO merely applies an analog gain to the signal which, for some sensors, results in greater efficiency (at the risk of more blown highlights). In the case of the new Sony sensor in the K5 and D7000, the efficiency is the same regardless of the ISO, paving the way for an ISO-less camera.

EJohnE said:

bobbin2 asks if I imagine his converter stretches the lens physically to 600 from 300. No, but he answers his own question by describing the EC20 teleconverter as 2x. That is an optical magnifier of the 300's image size. By design, all telephoto lenses are shorter than their focal length: all retrofocus lenses are longer than their focal length. One practical reason for the end-magnification of the teles is to make them more compact. My 135mm SLR lens is about 100mm from nodal point area to the film/sensor. The 7-14mm lens would be impossible if the 7mm setting was actually 7mm from the sensor. It has corrected reducing components or whatever clever solution is required to produce that angle of view from a greater actual lens distance.

Rikki Rask wants to know why I think some photons wont reach the exit pupil. They are outside the sensor's frame area. Where do photons go when after looking at something, I shut my eyes? Perhaps as an old song asks, do they go where the flies go in the winter?

Click to expand...

All besides the point. The simple fact of the matter is that the same number of photons fall on the sensor for the same scene, perspective, framing, aperture diameter, and shutter speed.

More reading for you over the Holiday:

http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/

http://forums.dpreview.com/forums/read.asp?forum=1018&message=32064270

Basically what you're saying is that ISO on digital cameras is sort of a gimmick because the only thing that's actually changing is a software algorithm that merely achieves the same effect as changing film speed. With film, the ISO (or ASA as my OM10 calls it) is related to the size of photosensitive crystals on the film. Without looking anything up, bigger crystals mean higher ISO because they accept more light, and the resulting "noise" or "grain" are the individual crystals and the gaps between them.
--
Brevity is the soul of wit.

rustler2x4 said:

Basically what you're saying is that ISO on digital cameras is sort of a gimmick because the only thing that's actually changing is a software algorithm that merely achieves the same effect as changing film speed. With film, the ISO (or ASA as my OM10 calls it) is related to the size of photosensitive crystals on the film. Without looking anything up, bigger crystals mean higher ISO because they accept more light, and the resulting "noise" or "grain" are the individual crystals and the gaps between them.

Click to expand...

No, ISO is not a gimmick at all on many cameras, since some sensors are significantly more efficient at higher ISOs -- see here for an example of this:

http://www.pbase.com/joemama/pushediso

Thus, the photographer must intelligently balance noise vs blown highlights by selecting the "proper" ISO.

However, for some sensors, and most notably the new Sony sensor in the K5 and D7000, yes, ISO is nothing more than a "gimmick", and all that using a higher ISO gets you is more blown highlights:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=37235114

Much better to set the f-ratio for the DOF / sharpness that you want, and the shutter speed for motion blur / camera shake that you can accept, and adjust the brightness of the image to taste in post.

The exposure meter will show you how many stops you are from the "recommended" exposure in the metering mode you are using, and you can then decide how much you want to compromise on the f-ratio and shutter speed to get less noise.

Of course, note that I said, "exposure meter" and not "total light meter". If we are not comparing between formats, we don't need to distinguish between exposure and total light, just as when we are not talking about different acceleration fields, we don't need to distinguish between mass and weight.

One of the huge differences between film and digital is that the digital sensor has a linear response to light, whereas film did not:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=37240582

EJohnE said:

Rikki Rask wants to know why I think some photons wont reach the exit pupil. They are outside the sensor's frame area.

Click to expand...

Try harder, EJohnE, and re-read the question. Maybe you will come to realize how utterly wrong your reference to the inverse square law is.
--
Rikke

EJohnE said:

ISO stands for International Standards Organisation.

Click to expand...

Actually, it doesn't. It just stands for 'ISO'. The organisation's full name is the 'International Organisation for Standardisation'. Don't ask me why that's the case - presumably someone at ISO knows.

EJohnE said:

There are ISO standards for just about anything. Standard repeatable testing is adhered to by any manufacturing member of the ISO, so that one can be assured of reasonable consistency within any standard.
Film does have a fixed ISO rating,

Click to expand...

it doesn't - only in conjunction with some reference development process - development can have a significant effect on the exposure index of a film

EJohnE said:

as does a sensor.

Click to expand...

And a sensor doesn't have an ISO rating at all. The relevant standards are ISO 5800:1987, ISO 6:1993 ISO, 2240:2003 for colour negative, monochrome negative and colour reversal films and ISO 12232:2006 for digital. It may be profitable to read those before telling people what they do and don't say.

EJohnE said:

Both may be modified by less/more exposure,

Click to expand...

exactly not. The Exposure Index defines a target exposure.

EJohnE said:

and compensating processing.

Click to expand...

the processing determines what's black and what's white.

EJohnE said:

Developing for film, signal amplification for sensor. For this great genius to write “ We want to gather as much light as possible so as to have the least noise.....” is nonsense.

Click to expand...

Before mocking you might want to check that you know what you're talking about. In this case you don't - since it is exactly correct that the major component on noise in a digital image, the photon shot noise, is determined directly by the amount of light in the image.

EJohnE said:

We want to gather the correct amount of light for that photograph. A great amount of light could be gathered if exposed at f2 and 30 seconds at 6400 ISO on a bright day. Is that what is meant?

Click to expand...

The sensor will saturate at some point and clip the highlights. That's probably the biggest exposure its wise to give.

EJohnE said:

bobbin2 asks if I imagine his converter stretches the lens physically to 600 from 300. No, but he answers his own question by describing the EC20 teleconverter as 2x. That is an optical magnifier of the 300's image size. By design, all telephoto lenses are shorter than their focal length: all retrofocus lenses are longer than their focal length. One practical reason for the end-magnification of the teles is to make them more compact. My 135mm SLR lens is about 100mm from nodal point area to the film/sensor. The 7-14mm lens would be impossible if the 7mm setting was actually 7mm from the sensor. It has corrected reducing components or whatever clever solution is required to produce that angle of view from a greater actual lens distance.

Click to expand...

You completely miss the point. You stated that 'the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies.' So, my question was that we have doubled the focal length of a lens, adding 300mm, yet its physical length is 41mm longer, so, where, in your view, does the extra 259mm, that the light has to travel for the inverse square law to apply, come from. I fully understand about telephoto and retrofocus lenses, but that knowledge completely undermines your case. If a lens' physical length doesn't match its focal length, where is this distance that the lens has to travel for the inverse square law to apply?

--
Bob

(I have lately been more concerned with the ongoing flooding around here than with less threatening haggles in DPR).

bobn2 in triumphant “Gotcha!” mode again asks me if the light through the 300mm lens plus 2x converter has to travel 600mm. His initial “thought experiment” was : If you put this (EC20 2x Teleconverter) behind a 300/2.8 lens you get a 600mm/5.6 lens. No you don't. Where did I claim that? I refuted it by stating that the 2x converter was a (twice-times) magnifier. The focal length remains at 300mm, but the image is optically twice magnified, and the area fitting within the sensor's frame is captured. The great master of equivalence might care to comment on why, with the unaltered size of diaphram aperture, f2.8 changes to f5.6.

As an aside, some have advocated changing lens nomenclature from focal lengths to the particular format's angles of coverage, as few modern lenses can be externally seen to physically agree with their optical focal length. I do not think it will happen.

The business about the same number of photons or whatever going to the sensor if the aperture diaphram is the same size is largely meaningless for practical purposes. Four thirds sensor is near enough to ¼ the size of the 35mm sensor. f2, 50mm, infinity focus, 1/100 sec., 100 ISO on 4/3rds quadruples the image density/brightness compared with 100mm f4 etc., on the “full frame”sensor.

Deep7 had the correct take on this. Don't bother to bring noise into this. A good big one will always beat a good little one. If Olympus can show no visible noise at 1600 ISO, a larger sensor can be be designed to outdo it.

As a little parable: If someone less mean than I am gives a child $10 pocket money, and to a group of four children also $10, they will complain that it is not fair. The equivalence gurus would tell them to stop grizzling, as the amount given to and received by each group was exactly the same.

I am a bit sick of this. I suggest you call DPR's resident polymath, Joseph Wisniewski into your equivalence net. If not, then find the absent Andzej Wrotniak.

EJohnE said:

(I have lately been more concerned with the ongoing flooding around here than with less threatening haggles in DPR).

bobn2 in triumphant “Gotcha!” mode again

Click to expand...

Not so triumphant, it's just that I have 'gotcha'. There is a fundamental inconsistency in your argument, and I'm asking you to consider it. You don't seem willing to do so, but that's not my problem.

EJohnE said:

asks me if the light through the 300mm lens plus 2x converter has to travel 600mm. His initial “thought experiment” was : If you put this (EC20 2x Teleconverter) behind a 300/2.8 lens you get a 600mm/5.6 lens. No you don't.

Click to expand...

What do you have then? How is the 300+TC different from a 600mm lens?

EJohnE said:

Where did I claim that? I refuted it by stating that the 2x converter was a (twice-times) magnifier. The focal length remains at 300mm, but the image is optically twice magnified, and the area fitting within the sensor's frame is captured.

Click to expand...

And not due to the inverse square law? Good. Now you raised in your last post the issue of a telephoto lens. A telephoto is a lens shorter than the final FL with a negative lens behind it. A TC lengthens the FL by placing a negative lens behind a lens of shorter FL than the final FL. What, then is the difference?

EJohnE said:

The great master of equivalence might care to comment on why, with the unaltered size of diaphram aperture, f2.8 changes to f5.6.

Click to expand...

Don't need him. The answer is simple. f-number is the aperture diameter divided by the focal length. If the aperture diameter stays the same and the focal length doubles, then the f-number will halve. Simple arithmetic.

EJohnE said:

As an aside, some have advocated changing lens nomenclature from focal lengths to the particular format's angles of coverage, as few modern lenses can be externally seen to physically agree with their optical focal length. I do not think it will happen.

Click to expand...

Nor do I - not that its at all relevant to the current discussion. Just to bring you back to the original point. As you say, most long lenses are shorter than their FL (because they are telephotos, a short lens with a built in TC), so given what you said in your OP - 'Where you go wrong is that you forget or don't know that the "focal length" means that the light has to travel twice as far (50 to 100 mm) to reach the sensor, and that the inverse square law applies.' given that the lens is not as long as it focal length, exactly where is this 'travelling' happening? It has to be somewhere.

EJohnE said:

The business about the same number of photons or whatever going to the sensor if the aperture diaphram is the same size is largely meaningless for practical purposes. Four thirds sensor is near enough to ¼ the size of the 35mm sensor. f2, 50mm, infinity focus, 1/100 sec., 100 ISO on 4/3rds quadruples the image density/brightness compared with 100mm f4 etc., on the “full frame”sensor.

Click to expand...

Exactly so, so both images contain the same number of photons, have the same shot noise, so we might say that 50/2 on Four Thirds is 'equivalent' to 100/4 on full frame.

EJohnE said:

Deep7 had the correct take on this. Don't bother to bring noise into this. A good big one will always beat a good little one. If Olympus can show no visible noise at 1600 ISO, a larger sensor can be be designed to outdo it.

Click to expand...

Exactly so - except that it's all about noise - that is the mechanism by which the good big one beats the good little one. You seem to be taking both sides of the argument at once. The nice thing about the equivalence concept though, is it tells you exactly wehen the big one and little one produce the same results. If you always are within that envelope, then there's no point paying the price and size premium for the big one.

EJohnE said:

As a little parable: If someone less mean than I am gives a child $10 pocket money, and to a group of four children also $10, they will complain that it is not fair. The equivalence gurus would tell them to stop grizzling, as the amount given to and received by each group was exactly the same.

Click to expand...

Not really a good metaphor at all. A better one would be if someone in one country was given $10 and someone else in another country where the cost of living was one quarter was given $2.50. Although $10=$10=$10, the person in the second country can do the same job with $2.50 as can the one in the first country with $10.

EJohnE said:

I am a bit sick of this.

Click to expand...

Well, you started it. If you hadn't decided to start a whole thread devoted to your ill-informed and scientifically illiterate rant against equivalence, you wouldn't have been called upon to find arguments to support your bloopers.

EJohnE said:

I suggest you call DPR's resident polymath, Joseph Wisniewski into your equivalence net.

Click to expand...

He's already here:

'This is going to come as a complete shock to you, but Joseph James is correct in nearly everything he says. I was teaching what he calls "equivalence" 15 years ago, with examples from 35mm, 6x7, and 4x5. Including the use of different film speeds on the different formats to demonstrate the same concept as we have with different sensor ISO ratings.' - Joseph Wisniewski.
http://forums.dpreview.com/forums/readflat.asp?forum=1021&message=35821548

EJohnE said:

If not, then find the absent Andzej Wrotniak.

Click to expand...

No need, the Joes are quite sufficient.
--
Bob

Much as I hate to agree with one of these non 4/3 owners who seem to be infecting this forum so much lately... A 100 F4 FF lens has the same aperture diamater as a 4/3 50 F2.

What's missing is the image circle. The FF lens's image circle is four times as large, so the light coming through that 25mm aperture covers four times the area. Yes, it will be fainter on a per pixel basis.

Often, people unfamiliar with 4/3 don't realize that the ZD glass has an image circle optimized for a 4/3 sensor, whereas a lot of APS glass produces an FF image circle, with a lot of light being thrown away. (hence the 'crop' designation. ZD on 4/3 is not a crop) As well, the more square 4/3 ratio makes more efficient use of the image circle than the oblong 3:2 sensor found in APS and FF. I doubt that makes a substantial difference, but it is a more efficient use of the available image circle.

A few years back, I tested my old Nikkor 200 3.5 against my ZD 50-200 at 200 3.5, both on my E1. Theoretically, the same focal length, same aperture, but the 50-200 yielded about 1.5 stops better performance, because it was focusing the same amount of light into a smaller image circle, and it was making better use of that image circle. Keep in mind that a bit of wiggle room is left in the 4/3 image circle to allow for IBIS, so that's likely why I didn't see a 2 stop advantage. What I did notice is that the ZD zoom handily outresolved a most respectable EDIF Nikkor prime. I was impressed, as that Nikkor had served me well over the years.

Of course, if you actually used both systems instead of using one and theorizing in an agenda driven manner about the other, you'd know this.

bobn2 said:

What do you have then? How is the 300+TC different from a 600mm lens?

Click to expand...

What manufacturer makes a contemporary 600mm f/5.6 that performs adequately on digital?

(contemporary = acceptable AF, acceptable sharpness, acceptable CA/PF, acceptable micro-contrast, given current lenses)

Cheers,
--
Tim
'I haven't been everywhere, but it's on my list.'
E3/7-14/12-60/35-100/150/25/EC14/EC20
http://www.flickr.com/photos/timskis6/

Timskis6 said:

Timskis6 said:

What do you have then? How is the 300+TC different from a 600mm lens?

Click to expand...

What manufacturer makes a contemporary 600mm f/5.6 that performs adequately on digital?

(contemporary = acceptable AF, acceptable sharpness, acceptable CA/PF, acceptable micro-contrast, given current lenses)

Click to expand...

Sure, there is no current 600/5.6 - the question I was asking was how a 300+TC is distinguishable from a 600 assuming that there was one (except, of course, it comes apart in the middle).

--
Bob

TrapperJohn wrote about non-4/3rds users butting in on the Olympus DSLR forum. If I am meant, then he is mistaken. I have 3 E-x cameras, with 7 lenses ranging from the 7-14 to the 70-300, plus FL36, and small accessories. I also own a Canon 5D-II for which I have only bothered to buy the cheap, but optically rather good 50mm f1.8. Both camera types were bought so that I could use my Leica R lenses on them. I have little interest in more lenses for the Canon, but I was surprised at the quality of the Olympus kit lenses. So much so, that I bought more.

I also have 10 Leica R lenses with several adapters for those lenses to fit on the E and the Canon bodies. The cheap chipped adapters are handy, but I prefer the quality and precision of the Cameraquest and Novoflex ones which accurately focus to infinity. Amongst the Leica lenses, are 50mm f2 and 100 Macro f4. Very convenient if I feel the need to prove anything. Oh, I also have three Leica R cameras. None of that qualifies me as an expert. It qualifies me only as one who can afford rather a lot of gear.

To bobn2 who is still rabbiting on about what he wrongly assumes to be my opinion that a 2x converter becomes another 300mm of light travel. Dear, oh dear. If I look at a postage stamp through a magnifier I see more detail, but have in no way changed its physical size As to why a lens plus converter is different from a telephoto lens: the lens can be used on its own, the lens/converter can be used together, the converter on its own will give nothing but unfocused illumination. If the (negative, is it?) elements are removed from the telephoto lens, what you have is an expensive repair job. To save embarrassing him I will refrain from commenting on his version of that parable.
Once the holiday season is over, he should find it easier to obtain Prozac.

As for my cluelessness re the inverse square law: that relationship was discovered, not devised by, say, Fred Flinstone. It applies to all sorts of things. An image falling on a lens does not go as we see it straight through to the film plane. The top of a building's image goes to the bottom of the sensor/film. It goes to a crossover point within the lens, and then the whole captured image expands to fill the format's frame. That nodal point may be considered to be a point source. For a 50mm lens, a light reading 25mm behind that point would be four times higher than at the 50mm distant sensor.

The more serious matter of Joseph Wisniewski's imputed agreement with a 4/3rds f2 50mm lens being only equivalent to a “Full frame” (a sloppy term) f4 100mm lens. If that refers only to the same lens diameter allowing the same number of photons or whatever to reach each sensor size, I cannot understand why such a statement of the obvious is regarded as such a “Eureka” moment. The fact that the same quantity of light passes through each lens is not in dispute. What is, is that each sensor then has the same image density. It demonstrably doesn't. Just try it on the ubiquitous brick wall. The same subject, same distance, same shutter speed, same ISO from a 4/3rds canera at 50mm f2 will show two stops greater exposure than a 5D-II with 100mm at f4. Just look at the histograms. For the histograms to align properly, the 100mm f4 would need to be exposed for 4 times as long.

I have the gear to do such an unnecessary experiment, but I will leave it to others who may be able to show the results in DPR. I have no interest in putting photographs on Flikr etc, and DPR seems to require such a source to download pictures. The Leica Users' Forum can use photos directly from a member's computer, so I occasionally (as EJohnE) have contributed some there, but not for several months.

EJohnE said:

The more serious matter of Joseph Wisniewski's imputed agreement with a 4/3rds f2 50mm lens being only equivalent to a “Full frame” (a sloppy term) f4 100mm lens. If that refers only to the same lens diameter allowing the same number of photons or whatever to reach each sensor size, I cannot understand why such a statement of the obvious is regarded as such a “Eureka” moment. The fact that the same quantity of light passes through each lens is not in dispute. What is, is that each sensor then has the same image density. It demonstrably doesn't.

Click to expand...

It's all explained, in detail , here:

http://www.josephjamesphotography.com/equivalence/#exposure

If the details are too much of a "bother" for you, here it is on the short: exposure is meaningless in cross format comparisons -- the relevant measure is total light .