EF-X Mirrorless concept

Jul 28, 2015
3,369
570
BillB said:
ahsanford said:
BillB said:
So, what would be the aps-c equivalent of a 50mm lens set to f4.0 in terms of framing, light gathering and DOF?

50mm f/4 FF would be the same...

  • Framing and DOF as a 32mm* f/2.8-ish lens on crop (it's not exactly one stop, that's just a rough rule of thumb)
  • Light gathering / exposure as any f/4 lens on any sensor. Sensor size doesn't change how aperture gathers light.

*Canon and Nikon have different APS-C crop, so it varies a bit.


But because light gathering and DOF do not work the same way w.r.t. equivalence, there is no perfectly equivalent situation.

(Someone straighten me out if I've misread that.)

- A

That's what I think too, and that is why I asked Fullstop to explain his notion of equivalence. He posts like the size of the sensor affects the exposure setting (an opinion shared by a surprising number of people), and I don't think it is so. How come a setting of 1/125 and f4.0 taken off my handheld Sekonic works for both my 5DIV and my ancient Rebel? I don't believe that you need a 70-200 F2.8 on an aps-c camera to get "equivalence" with a 70-200 F4 on a full frame camera, at least in any particularly meaningful sense.

Sensor size does not affect exposure setting - but it permits you to go one ISO point higher and/or get images with less noise.
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
BillB said:
I don't believe that you need a 70-200 F2.8 on an aps-c camera to get "equivalence" with a 70-200 F4 on a full frame camera, at least in any particularly meaningful sense.

Same focal length for the two sensors confounds the comparison a bit.

In your example, 70-200 f/2.8 on crop will look resemble 112-320 f/4.5-ish FF output w.r.t. subject isolation / bokeh, but it will gather f/2.8 light from an exposure perspective. In all these equivalence exchanges, one must decouple exposure from DOF. He's harping on DOF and you are focusing on exposure.

To put AvTvM's point in your example, AvTvM is saying 'to get FF f/4 subject isolation on crop, you need to use an f/2.8 lens on that crop body'. That's (roughly, ballpark, broad strokes, etc.) correct.

So, if you want a lens that is quick enough to gather enough light for exposure, you are correct. You don't need that faster lens on crop.

But if you are focused on subjects popping from the background more (or the background transitions looking more pleasing), AvTvM is correct in that the faster lens is 'needed'.

- A
 
Upvote 0
Jul 28, 2015
3,369
570
ahsanford said:
BillB said:
I don't believe that you need a 70-200 F2.8 on an aps-c camera to get "equivalence" with a 70-200 F4 on a full frame camera, at least in any particularly meaningful sense.

Same focal length for the two sensors confounds the comparison a bit.

In your example, 70-200 f/2.8 on crop will look resemble 112-320 f/4.5-ish FF output w.r.t. subject isolation / bokeh, but it will gather f/2.8 light from an exposure perspective. In all these equivalence exchanges, one must decouple exposure from DOF. He's talking about DOF and you are talking about exposure.

To put AvTvM's point in your example, AvTvM is saying 'to get FF f/4 subject isolation on crop, you need to use an f/2.8 lens on that crop body'. That's (roughly, ballpark, broad strokes, etc.) correct.

- A

True.
But that is only if you need the DOF a f2.8 lens gives, and ModeDialAvTvMfullstop keeps telling us he does not need that. And, to be honest, I think relatively few people care about it.
 
Upvote 0
Apr 25, 2011
2,521
1,900
BillB said:
Well, I certainly don't understand your concept of equivalence. Could you please explain?
In order to get exactly the same image on a cropped sensor, with exactly the same resolution and exactly the same illumination per pixel, you need to keep the effective entrance pupil of the lens the same, but move the sensor forward proportionally to its crop factor (and proportionally increase its pixel density), and change the lens focal length so that the sensor is in focus again.

When we can ignore the effects of the magnification factor (say, the magnification factor is less than 1:5), the equivalent lens for a crop will have its f-stop value multiplied by, and its focal length divided by the crop factor. If we cannot ignore the magnification factor, the equations will be a bit more complex.
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
Kit. said:
BillB said:
Well, I certainly don't understand your concept of equivalence. Could you please explain?
In order to get exactly the same image on a cropped sensor, with exactly the same resolution and exactly the same illumination per pixel, you need to keep the effective entrance pupil of the lens the same, but move the sensor forward proportionally to its crop factor (and proportionally increase its pixel density), and change the lens focal length so that the sensor is in focus again.

When we can ignore the effects of the magnification factor (say, the magnification factor is less than 1:5), the equivalent lens for a crop will have its f-stop value multiplied by, and its focal length divided by the crop factor. If we cannot ignore the magnification factor, the equations will be a bit more complex.

And, not picking on you -- I am guilty here -- this is why certain topics: equivalence, batteries, to filter or not filter, etc. tend to derail the original topic of the thread.

We can talk about the necessity of large apertures in mirrorless all day, but if this thread continues down an equivalence details rabbit hole, I will lock it. (No disrespect intended! :) )

- A
 
Upvote 0
May 11, 2017
1,365
635
ahsanford said:
Kit. said:
BillB said:
Well, I certainly don't understand your concept of equivalence. Could you please explain?
In order to get exactly the same image on a cropped sensor, with exactly the same resolution and exactly the same illumination per pixel, you need to keep the effective entrance pupil of the lens the same, but move the sensor forward proportionally to its crop factor (and proportionally increase its pixel density), and change the lens focal length so that the sensor is in focus again.

When we can ignore the effects of the magnification factor (say, the magnification factor is less than 1:5), the equivalent lens for a crop will have its f-stop value multiplied by, and its focal length divided by the crop factor. If we cannot ignore the magnification factor, the equations will be a bit more complex.

And, not picking on you -- I am guilty here -- this is why certain topics: equivalence, batteries, to filter or not filter, etc. tend to derail the original topic of the thread.

We can talk about the necessity of large apertures in mirrorless all day, but if this thread continues down an equivalence details rabbit hole, I will lock it. (No disrespect intended! :) )

- A

And thank you for your help in bringing this into something approaching an equilibrium state
 
Upvote 0

Don Haines

Beware of cats with laser eyes!
Jun 4, 2012
8,246
1,939
Canada
There is no such thing as equivalence.

You have a bunch of related parameters, and as you change one, you affect the others.

There is the “framing” equivalence..... if I have a 100mm lens on a crop camera, then if standing at the same spot, I need a 160mm lens on a FF camera to get the same framing, but I have now lost equivalence on depth of field, but I keep perspective.

If I want to keep the same lens on both cameras, then I zoom with my feet to keep the same framing, but I loose perspective.....

If I want to keep DOF, I shoot at F4 on my FF and F2.8 on my crop.... but now I have different shutter speeds....

The thing is, you can adjust settings to get equivalence for one parameter, but in doing so you loose equivalence in others.... And remember, it is an EQUIVALENCE! When you move a 100mm F2.8 lens from a FF camera to a crop camera, it remains a 100mm F2.8 lens...... it does not magically change into a 160mm F4 lens! The photons passing through that lens have no idea what the sensor is.......
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
Again, on the original topic of the mount, I am skeptical Canon will go for some transformer/telescoping mount. I'm also skeptical Canon will pursue the EF-X concept listed here as I honestly believe they will turn off the 'keep it small' crowd with that move.

Like it or not, pragmatism be damned, I think Canon is going thin mount and small with one of its first offerings to court the 'keep it small' crowd.

- A
 
Upvote 0

Don Haines

Beware of cats with laser eyes!
Jun 4, 2012
8,246
1,939
Canada
ahsanford said:
Again, on the original topic of the mount, I am skeptical Canon will go for some transformer/telescoping mount. I'm also skeptical Canon will pursue the EF-X concept listed here as I honestly believe they will turn off the 'keep it small' crowd with that move.

Like it or not, pragmatism be damned, I think Canon is going thin mount and small with one of its first offerings to court the 'keep it small' crowd.

- A

I agree with you about the telescoping Mount... imagine what happens with a big white attached? Also, how do you seal it? What about the contacts? This is most definitely not an elegant solution
 
Upvote 0
May 11, 2017
1,365
635
ahsanford said:
Again, on the original topic of the mount, I am skeptical Canon will go for some transformer/telescoping mount. I'm also skeptical Canon will pursue the EF-X concept listed here as I honestly believe they will turn off the 'keep it small' crowd with that move.

Like it or not, pragmatism be damned, I think Canon is going thin mount and small with one of its first offerings to court the 'keep it small' crowd.

- A

Fair enough. Nice hedge--"one of the first" is hard to bet against. Some of it depends on how small they can get while still using the EF mount. How much market space is there between the smallest possible EF fullframe and the M5? And how would the smallest possible Canon EF FF stack up against the A7 III or the Nikon whatever in the smallness wars? Canon trusts its market research whatever the internet buzz. If they do go "Super M" with a Fullframe EF-M mount camera, my guess is there will be a 24-70 f4, a prime or two, maybe a spiffed up adapter, and then maybe a 16-xx f4 down the road a bit, and not much more. Late thought--are we sure that the 32mm EF-M is aps-c? :)
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
BillB said:
Late thought--are we sure that the 32mm EF-M is aps-c? :)

Yes, for three reasons:

1) 32mm screams crop as it's 'Canon crop speak' for a 50 prime. Why would they avoid the classic 24, 35, 50 increments on that one for FF?

2) The 35L II already exists.

3) The non-L 35 f/2 IS already exists.

- A
 
Upvote 0
May 11, 2017
1,365
635
ahsanford said:
BillB said:
Late thought--are we sure that the 32mm EF-M is aps-c? :)

Yes, for three reasons:

1) 32mm screams crop as it's 'Canon crop speak' for a 50 prime. Why would they avoid the classic 24, 35, 50 increments on that one for FF?

2) The 35L II already exists.

3) The non-L 35 f/2 IS already exists.

- A

But it would be so cool to roll out a "Super M" FF EF-M mount camera with a lens that could be used as a 32mm on FF Super M and as a 50mm equivalent on plain old fashioned aps-c M cameras. A natural followon would be a 50mm EF-M usable on FF Super M that could be used as 80mm equivalent on aps-c. What if it turned out that the 22mm EF-M has had a FF image circle all along!

(And either of the EF 35's would need an adapter, which would annoy the small is beautiful crowd.)
 
Upvote 0
Mar 2, 2012
3,188
543
Don Haines said:
ahsanford said:
Again, on the original topic of the mount, I am skeptical Canon will go for some transformer/telescoping mount. I'm also skeptical Canon will pursue the EF-X concept listed here as I honestly believe they will turn off the 'keep it small' crowd with that move.

Like it or not, pragmatism be damned, I think Canon is going thin mount and small with one of its first offerings to court the 'keep it small' crowd.

- A

I agree with you about the telescoping Mount... imagine what happens with a big white attached? Also, how do you seal it? What about the contacts? This is most definitely not an elegant solution

You could do it with a deploying wire harness, but it would be needlessly complicated. Sounds like a cool patent but I’ll be shocked to see it go anywhere.
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
BillB said:
But it would be so cool to roll out a "Super M" FF EF-M mount camera with a lens that could be used as a 32mm on FF Super M and as a 50mm equivalent on plain old fashioned aps-c M cameras. A natural followon would be a 50mm EF-M usable on FF Super M that could be used as 80mm equivalent on aps-c. What if it turned out that the 22mm EF-M has had a FF image circle all along!

(And either of the EF 35's would need an adapter, which would annoy the small is beautiful crowd.)

Lenses that will cover FF image circles -- at those speeds especially -- will be lead pipe pickle jars. I'm not sure what the appetite for this (see pic) would be for the EOS M crowd. Your hands would have zero finger room (probably worse than A7 + GM lenses) and you'd be packing a ton more weight than you need to.

That said: I'm sure the IQ from the center of that image circle would be something ridiculous.

- A
 

Attachments

  • Screen Shot 2018-07-09 at 3.02.58 PM.png
    Screen Shot 2018-07-09 at 3.02.58 PM.png
    130.9 KB · Views: 308
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
3kramd5 said:
Sounds like a cool patent but I’ll be shocked to see it go anywhere.

In my line of work, it's the very definition of my management saying "Wow. That's... wow. Go patent that concept immediately exactly as it is today -- unsolved problems, warts and all -- and please promptly move on to more practical solutions."

Translation: we should bank this innovation so we never get scooped on the idea by a competitor, but there's no way in hell I'm letting you take the time to climb Mt. Awful to sort out all the major issues I can immediately see. I'd like to see this new product before I retire, and I'm not sure this one will fit that timeline. ;D

- A
 
Upvote 0
May 11, 2017
1,365
635
ahsanford said:
BillB said:
But it would be so cool to roll out a "Super M" FF EF-M mount camera with a lens that could be used as a 32mm on FF Super M and as a 50mm equivalent on plain old fashioned aps-c M cameras. A natural followon would be a 50mm EF-M usable on FF Super M that could be used as 80mm equivalent on aps-c. What if it turned out that the 22mm EF-M has had a FF image circle all along!

(And either of the EF 35's would need an adapter, which would annoy the small is beautiful crowd.)

Lenses that will cover FF image circles -- at those speeds especially -- will be lead pipe pickle jars. I'm not sure what the appetite for this (see pic) would be for the EOS M crowd. Your hands would have zero finger room (probably worse than A7 + GM lenses) and you'd be packing a ton more weight than you need to.

- A

Well, they could use STM instead of USM for starters, and you wouldn't need to use retrofocus... :)
 
Upvote 0
Mar 2, 2012
3,188
543
ahsanford said:
BillB said:
So, what would be the aps-c equivalent of a 50mm lens set to f4.0 in terms of framing, light gathering and DOF?

50mm f/4 FF would be the same...

  • Framing and DOF as a 32mm* f/2.8-ish lens on crop (it's not exactly one stop, that's just a rough rule of thumb)
  • Light gathering / exposure as any f/4 lens on any sensor. Sensor size doesn't change how aperture gathers light.

*Canon and Nikon have different APS-C crop, so it varies a bit.


But because light gathering and DOF do not work the same way w.r.t. equivalence, there is no perfectly equivalent situation.

(Someone straighten me out if I've misread that.)

- A

Sensor size doesn’t affect how the aperture gathers light, naturally.

But f/4 on a 50mm lens is not the same as f/4 on a 32mm lens. The former has more than double the physical aperture diameter, and will transmit less light. You need to stop the shorter lens up to about f/2.6 to get that ~120mm2 opening, which with the same FOV will transmit the same light volume.
 
Upvote 0

ahsanford

Particular Member
Aug 16, 2012
8,620
1,651
3kramd5 said:
Sensor size doesn’t affect how the aperture gathers light, naturally.

But f/4 on a 50mm lens is not the same as f/4 on a 32mm lens. The former has more than double the physical aperture diameter, and will transmit less light. You need to stop the shorter lens up to about f/2.6 to get that ~120mm2 opening, which with the same FOV will transmit the same light volume.

Which is why all discussions on equivalence lead to thread plague. ;D

Point taken, but let's please move back on topic.

- A
 
Upvote 0

docsmith

CR Pro
Sep 17, 2010
1,243
1,198
ahsanford said:
Again, on the original topic of the mount, I am skeptical Canon will go for some transformer/telescoping mount. I'm also skeptical Canon will pursue the EF-X concept listed here as I honestly believe they will turn off the 'keep it small' crowd with that move.

Like it or not, pragmatism be damned, I think Canon is going thin mount and small with one of its first offerings to court the 'keep it small' crowd.

- A


Agreed....if we go thin (flange distance less than 44 mm) then "EF-X" still is needed, as we will need a FF image circle coming from a lens with a shorter flange distance.
So, what I think will happen is that Canon will not want to limit their future camera body sizes to fit lenses designed for a mirror box that is no longer needed. I expect 3-4 "EF-x" lenses to be introduced with a FF mirrorless camera.


But...let me get your opinion on this as a "sexy" and "elegant" solution. The Canon FF mirrorless entry has a shorter flange distance, but has a built in lens element in the camera body that takes the EF image circle and broadens it to fit a FF image sensor in a distance of less than 44 mm.


So, say a 25-30 mm flange distance, so still more compact, but enough room for that lens element. We would no longer remove a lens and see the sensor, rather we'd remove the lens and see some glass.


Has this already been suggested? Thoughts? Could get odd when the focal point occurs within this new lens element, but maybe it is possible. Also that whole "physics," "science," and "reality" may be issues as the same element would need to work with light coming in at different angles. I do like how the listed focal lengths on the lenses would still be the equivalent focal lengths, even though they would technically be a little different, FOV would be the same. This only happens if you have a 44 mm flange distance either natively or with an adapter or if we are bending light with a lens element and a shorter flange distance, as suggested.
 
Upvote 0