5D vs. 7D (is there a difference when acquiring video?)

Adrian Sierkowski · January 13, 2011

Yes, in theory, as film is film for 35mm and 16mm, just cut to different sizes, yet at the same time "no," as the size of the photosites does matter too, as David Mentioned, as does their fill factor, but for all intents and purposes you'd just be making a 7D!

Granted, that's just my own theory/understanding of how it would work, you'd need a qualified optical engineer to tell you for sure ;)

rob spence · January 13, 2011

Hi Adrian,

I found this on a website.

• If you use the same lens on a Canon APS-C crop sensor camera and a 35mm full frame body and crop the full frame 35mm image to give the same view as the APS-C crop image, the depth of field is IDENTICAL

rob spence · January 13, 2011

I found this very interesting article

http://www.bobatkins.com/photography/technical/digitaldof.html

that seems to deal with the issues relating to using different format lenses on crop cameras. Looks like I'll be up all night studying!

Chris Millar · January 13, 2011

Hi Adrian,

I found this on a website.

• If you use the same lens on a Canon APS-C crop sensor camera and a 35mm full frame body and crop the full frame 35mm image to give the same view as the APS-C crop image, the depth of field is IDENTICAL

Unless I'm mistaken you're forgetting the blow up factor to make the veiwed image the same size...

As DOF is defined as a limit in human perception we're floating about not getting anywhere without taking into account the final viewing factor.

Relative DOF is always the same per ideal lens focal length when compared with other focal lengths - but a sensor size will determine the relative blow up factor so the imagery fills up the same area on our eventual viewing screen. It is this blow up (or shrink) that allows out of focus imagery to slip in and out of the range of imagery that we determine visually to be in or out of focus, But don't fool yourself conceptually by looking through a lens for a reality check, there are other factors to confuse you also, such as the resolving power of the lens and format acquisition (temporal and spatial). (But certainly there aint no better way than looking through a lens when it comes to actual shooting)

Nothing has made this stuff clearer to me (puns everywhere in photography) than working with large format stills...

And old trick (19th century) to increase DOF was to reduce the sharpness of the real plane of focus... So a reduction in 'quality' gave an impression of better DOF. :blink:

Although the methods have changed this trick is still being used today.

(Perhaps even unknowingly by many ?)

David Mullen ASC · January 14, 2011

Yes, you have to factor degree of enlargement.

Imagine the image produced by the lens being limited to the size of the gate or sensor, so (for the sake of making the math easier) you have one format that is 24mm x 12mm and a larger format that is 48mm x 24mm, twice as large in both directions.

So you put a 50mm lens on the 48mm x 24mm camera and it fills the frame without vignetting. You then put the 50mm lens on the 24mm x 12mm camera. Now it has half the field of view horizontally and vertically, but otherwise, it's the same image as on the 48mm x 24mm camera just cropped in half, so technically the depth of field hasn't changed and it is still a 50mm lens.

However, if you view them side by side, the 48mm x 24mm image is twice as big as the 24mm x 12mm image. Now you blow-up the 24mm x 12mm image to 48mm x 24mm. By enlarging it, you've made the focus twice as critical, so now the 50mm image on the smaller format has less depth of field than the 50mm on the larger format because the smaller image is being enlarged more, so focus is more visible, the Circle of Confusion figure has to be twice as critical.

So a 50mm lens doesn't have the same depth of field no matter what format you put it on because if it did, we wouldn't have separate depth of field charts for 16mm and 35mm, for example.

The only difference between a 50mm designed for a larger format is the coverage, not the field of view, it has to fill the frame without vignetting in the corners. But if you put a 50mm lens designed for a smaller format onto the larger format camera, the depth of field would be the same as the larger 50mm, the field of view would be the same, just that the smaller 50mm lens would vignette in the corners.

But either lens, put onto the smaller camera, would have less depth of field because the degree of enlargement needed to allow you to compare the depth of field to the same lens on the larger format would cause a more critical Circle of Confusion figure to be used. However, since the lens now on the smaller format would have a narrower field of view, more than likely you'd switch to a shorter focal length, which has more depth of field, to compensate.

Chris Millar · January 14, 2011

However, if you view them side by side, the 48mm x 24mm image is twice as big as the 24mm x 12mm image. Now you blow-up the 24mm x 12mm image to 48mm x 24mm. By enlarging it, you've made the focus twice as critical, so now the 50mm image on the smaller format has less depth of field than the 50mm on the larger format because the smaller image is being enlarged more, so focus is more visible, the Circle of Confusion figure has to be twice as critical.

Ack, the terminology used to describe enlargement factors (similar to megapixels vs. 1080, 4K etc...) always gets me in a fuddle.

Confusion alert:

'twice' refers to twice the linear measurements - but actually any given area will increase fourfold, and the distribution of information that allows us to experience DOF is across an area...

Are CoC's termed linearly or by area ?

As an aside I remember a whiles back having the 'ahhhh' moment when this all made sense to me - I'll be giving away my age/experience when I say it was a website that showed what we're talking about with images... Much much better than the words and numbers that we're dealing with here ! I can't find it anymore though

Adrian Sierkowski · January 14, 2011

I'm pretty sure CoC is Logarithmic. And yes we're talking 2x perimeter size, which would be a larger increase in overall area. Welcome to film.. no one ever told us there'd be this much math!

Chris Millar · January 14, 2011

I'm pretty sure CoC is Logarithmic. And yes we're talking 2x perimeter size, which would be a larger increase in overall area. Welcome to film.. no one ever told us there'd be this much math!

I should rephrase the question:

by diameter or by area ?

I went looking for some solid core wire the other day for some lighting stuff I was working on:

sizes:

0.5mm

1mm

1.25mm and so on...

Questions asking if they were referring to diameter (or radius for that matter) or cross sectional area fell on deaf ears - they didn't understand the implication...

Edited January 14, 2011 by Chris Millar

David Mullen ASC · January 14, 2011

Ack, the terminology used to describe enlargement factors (similar to megapixels vs. 1080, 4K etc...) always gets me in a fuddle.

Confusion alert:

'twice' refers to twice the linear measurements - but actually any given area will increase fourfold, and the distribution of information that allows us to experience DOF is across an area...

Are CoC's termed linearly or by area ?

It's the measurement of the diameter of the circle, so it's measured in one direction, not by area. In other words, if the image is enlarged by twice horizontally and twice vertically, then the circle is half the diameter, twice as critical... not 4X.

Hal Smith · January 14, 2011

I went looking for some solid core wire the other day for some lighting stuff I was working on:

0.5mm

1mm

1.25mm and so on...

Questions asking if they were referring to diameter (or radius for that matter) or cross sectional area fell on deaf ears - they didn't understand the implication...

In the US electrical wire size is referred to primarily by AWG number, #10, #14, 000 ("Three Aught"), etc.

There is a secondary system where it is sized by an area based measurement, Circular Mils. Circular mil area size enables describing the current carrying capacity of shapes other than circles like the rectangular buss bars found in distribution panels.