4/3 vs Full Frame - DOF and f-stops

[Boatman]

I've seen some of the controversy over Tony Northrup's videos, so I had to go watch them for myself. I understood what he was saying but I was still confused over possible differences between a m43 lens and a full frame lens. Rather than argue about it, I went and took some photographs to compare them.

Using an f1.4/50 Takumar and my Panasonic 14-140 lens at 50mm, I shot two photos. Both are shot at f5.6, 1/25 of a second at ISO 160. In both images I manually focused on the 4x4 post in the middle of the frame. Guess what? The two images are essentially the same: same exposure, same depth of field. There are coloration differences, which is to be expected. The images can be downloaded from http://1drv.ms/1jXsdxT if you want to examine them.

The upshot of this is that f-x is f-x. A given depth of field for a certain lens at a certain aperture opening is always the same. It is only when you start to compare the formats with comments like x-lens equals y-lens in some different format that things get complicated. This also means that if you have an old prime from a 35mm camera, the DOF guide is still going to be accurate when used on a m43 camera.

 

[bassman]

All this "equivalence" talk confuses people. Focal length = focal length, Fstop = Fstop. The lens knows not what sensor lives in the camera it's mounted on.

If you were to mount your 50/1.4 on a FF camera and take the same shot from the same position with the same settings, and then crop the resulting image to the same view as you got with a 50mm on m43, you would again see the same result: same exposure, same DoF. Because that's why we call m43 "cropped": the sensor is smaller than FF.

When people say m43 has deeper DoF than FF, what they are describing is that you choose a 25mm lens on m43 to get the same angle of view as you would get using 50mm on FF. And we all know that a shorter FL has greater DoF at a given Fstop.

 

[SkiHound]

I agree that a lens is a lens is a lens, but...

If you compare a 25mm f1.4 on m43 versus a 50mm f/1.4 on FF the field of view is about the same (with a little approximation for different aspect ratios), and the working exposure value is the same, but the the 50mm on FF will have considerably shallower depth of field than the 25mm on m43. So I agree, that a lens, is a lens, is a lens, but the effective field of view depends on sensor size and thus there are real differences in terms of controlling depth of field. Of course, sometimes deeper depth of field is an advantage and sometimes narrower depth of field is an advantage.

 

[carpandean]

I always like to think about this issue working backwards from the image that I want. Let's say that I have an image in my head that I want to capture. This can be broken down to a desired field of view, depth of field, perspective and exposure level (also, I might have a desired level of motion and acceptable level of noise.) If I know what it would take to achieve this on m4/3, then to achieve the same exact image (ignoring aspect ratio) on a FF camera, I would stand in the same spot (to control perspective) and use a lens with double the focal length and half the aperture (seems like double, but f/4 is half of f/2). Also, assuming similar generation of bodies, I could use four times the ISO to maintain the same shutter speed (to control motion) for the desired exposure at the smaller aperture and end up with the same approximate amount of noise.

So, yes, the physical characteristics of a lens do not depend on the format that it is used on, but there is a practical difference on how it will be used on each format. Loosely speaking, on similar generations of bodies, a 25/1.4 on m4/3 is "equivalent" to a 50/2.8 on FF in terms of creating final images.

 

[lightmonkey]

There's an undue level of obsession with equivalence on this internet thing.

What's the point in comparing with what someone else have - or , what you *dont* have?

Worry about your camera bodies and your lenses and how they render your scenes with your equipment.

Can you imagine a bunch of racing drivers on Pikes Peak in Colorado saying "well, if this mountain was dropped to sea level in Houston I would lap it this xx:xx fast instead!!!"????

The biggest relevance to equivalence is that 135mm / 24x36 have been popularly shot for 50 years so you'd like to that when people say "50mm prime provides the most useful everyday coverage" that it means the 25mm in Olympus and Pansonic world. That's all you really need to know. Why so much obsession over blur equivalence and not something like diffraction.

 

[bassman]

There's an undue level of obsession with equivalence on this internet thing.

What's the point in comparing with what someone else have - or , what you *dont* have?

Worry about your camera bodies and your lenses and how they render your scenes with your equipment.

Can you imagine a bunch of racing drivers on Pikes Peak in Colorado saying "well, if this mountain was dropped to sea level in Houston I would lap it this xx:xx fast instead!!!"????

The biggest relevance to equivalence is that 135mm / 24x36 have been popularly shot for 50 years so you'd like to that when people say "50mm prime provides the most useful everyday coverage" that it means the 25mm in Olympus and Pansonic world. That's all you really need to know. Why so much obsession over blur equivalence and not something like diffraction.

As you properly point out, another problem with using "equivalence" is that it's not really equivalent. Perhaps one aspect is: FOV, or perhaps, DoF, or perhaps total light gathering. But overall, it's not. So I'd also rather talk about the appropriate kit for the job: on my m43, 25mm gives me the proper FOV and f/2.8 gives me the proper DoF for the shot I'm taking. It's quite clear that a 25/1.4 on m43 is not really equivalent to a 50/2.8 on FF in all aspects. You have to keep addressing other aspects of your settings (ISO, SS) in an attempt to make the setting(s) you care about right now sort of equivalent.

And why bother? Again, I agree with you: I choose the kit to match the shot, not to match some other arbitrary kit. Need shallow DoF? Go for the biggest sensor, longst FL and widest aperture I can manage - given my other constraints such as desired and possible perspectives, equipment inventory, available light, etc.

Help stamp out equivalence !

 

[Rasmus]

Can you imagine a bunch of racing drivers on Pikes Peak in Colorado saying "well, if this mountain was dropped to sea level in Houston I would lap it this xx:xx fast instead!!!"????

I DO remember a rally driver in the eighties who said that if one were to put a trained monkey behind the steering wheel of an Audi Quattro, that monkey would still beat all the best drivers in the world if they had two wheel drive cars. So everyone compares gear.

 

[bikerhiker]

I've seen some of the controversy over Tony Northrup's videos, so I had to go watch them for myself. I understood what he was saying but I was still confused over possible differences between a m43 lens and a full frame lens. Rather than argue about it, I went and took some photographs to compare them.

Using an f1.4/50 Takumar and my Panasonic 14-140 lens at 50mm, I shot two photos. Both are shot at f5.6, 1/25 of a second at ISO 160. In both images I manually focused on the 4x4 post in the middle of the frame. Guess what? The two images are essentially the same: same exposure, same depth of field. There are coloration differences, which is to be expected. The images can be downloaded from http://1drv.ms/1jXsdxT if you want to examine them.

The upshot of this is that f-x is f-x. A given depth of field for a certain lens at a certain aperture opening is always the same. It is only when you start to compare the formats with comments like x-lens equals y-lens in some different format that things get complicated. This also means that if you have an old prime from a 35mm camera, the DOF guide is still going to be accurate when used on a m43 camera.

All lenses have the same depth of field. DOF is dictated by 3 things.
1, Subject distance
2, Aperture used
3, Angle of View

Focal length has no relevance in DOF. Angle of view is dicated by the size of sensor. An 80mm lens is considered telephoto with 35mm film or full frame sensor, but the same 80mm is considered wide angle on a 4x5 view camera.

In photography, we know that there is ONLY one plane that is truly in focus, or sharp. Everything else is just blur and aperture is based used to describe the area of in-focus or the area of out-focus when stopped down to say f/22. All lenses have a sweet working aperture that provides the best image quality.

A lot of confusion hinges upon the bokeh quality of the lens, where people believe that bokeh relates to the extreme shallow depth of field look. In reality, bokeh is the quality of blur the lens produced and you can actually predict this from the lens MTF chart. The proximity of the dotted and dash lines of the Sagittal and Meridonial lines of low spatial frequency vs high spatial frequency dictates how harsh and how pleasing the bokeh is rendered. And since very few lenses have the same MTF performance due to lens design, the bokeh quality between 2 lenses are rarely equalled. So aperture equivalency is just a way to describe what to expect in regards to how much or less blur you'll get with a lens compared to full frame. But full frame is not the de-facto measurement in regards to DOF. Full frame is just what's popularly compared to.