Confused about the depth of field

Hey guys, I just bought the Samyang XP 14mm F2.4 lens. I have the EOS 5D Mark IV body.

 

A few questions:

 

a) The aperture shown on the body is 2.5. Does this mean the actual aperture is 2.5? or internally, the aperture of 2.4 is used?

 

b) The hyperfocal distance for 14mm/f2.5 on the full frame body is around 2.61m. From what I understand, if I focus at that distance, everything from 1.3 m in front to inf will be in focus. However, I just did an experiment in which I changed the point of focus to see from what point everything becomes sharp, and the sharpness point was well beyond the hyperfocal distance! To better give an example, consider these two scenarios

 

- focus at 3m. Theoretically everything behind that point should be sharp

 

- focus at 6m. Everything behind that point should be sharp.

 

My car was parked at 6m (the 6m focus was on the car). The license should have been sharp at both 3m and 6m focus point (from the theory of hyperfocal distance). However, the 6m focus is the sharpest. I can see a clear difference for my car's license between 3m and 6m. What am I missing? Am I misunderstanding the concept of HFD?

 

The reason for this experiment is that I am trying to do night photography, and am trying to figure out what's the closest distance such that everything from that point (maybe slightly to the front of it ) to inf is in focus. Isn't that theoretically the HFD?

 

Thank you.

Well, see Hyperfocal distance - Wikipedia for details. Also "circle of confusion" (Circle of confusion - Wikipedia)

 

As for the actual wide-open aperture, you are into what is "real" and what is "nominal"

When selling lenses (or other equipment for that matter) the manufacturer sets a value, usually within some margin of error, for whatever the physical measurement with a micrometer, for example, might be. The more "obscure" the manufacturer or seller may be, the more the tendency to stretch to a more 'sales-attractive' value.

 

I don't know about Samyang, but even major Japanese makers have done this - so your lens could be anything from a 15mm f/2.5 lens to a 13mm f/2.3 lens, just for an example.

 

Also - "sharp" is ambiguous in terms of focus. The human eye is not precise and it will accept images as being sharp even when they are not actually and precisely so.

 

Someone else can take it from here, if they want.

Regarding hyperfocal distances and sharpness...I like to think of it as a bell curve with the sharpest point being the actual point of focus and image sharpness tailing off ahead and behind that point. The key thing to keep in mind is that the extreme distances in a hyperfocal example are supposed to be "acceptably" sharp, neither razor sharp nor unrecognizably soft. In spite of what charts, lens markings, and manufacturer charts tell you, you really need to experiment a little with your own gear to see what "acceptably sharp" means to you and how your gear produces that.

After I calculate what the DOF I need should be, I stop down one additional stop for safe measure.

My car was parked at 6m (the 6m focus was on the car). The license should have been sharp at both 3m and 6m focus point (from the theory of hyperfocal distance). However, the 6m focus is the sharpest. I can see a clear difference for my car's license between 3m and 6m. What am I missing? Am I misunderstanding the concept of HFD?

 

If the car license plate was at 6 meters, and you focused at 6 meters, and the result was sharp, that's exactly how it should be.

 

If the car license plate was at 6 meters, and you focused at 3 meters, and the result was not as sharp, that's exactly how it should be.

 

To paraphrase SCL, think of DOF as a curve, in which your focus plane will be sharp, and objects in front or behind will be gradually less sharp the farther they are from where you focused. Depth of field simply gives you an area of "acceptable sharpness," but it's up to you to figure out and decide what is acceptable. Relying on hyperfocal distance will often result in photos with areas which you wish had been sharper. It was frequently used by photographers who knew they might not have time to focus and simply wanted to get everything sharp enough. Nowadays, with increasingly fast and effective autofocus, and focusing aids for when you want to control focus yourself, there is much less reason for using it. To think of it another way, the ultimate in the use of hyperfocal distance is a pinhole camera, which produces photos in which everything is rather soft. The effect can be lovely, but I don't think that is what you are seeking here.

"The hyperfocal distance for 14mm/f2.5 on the full frame body is around 2.61m. From what I understand, if I focus at that distance, everything from 1.3 m in front to inf will be in focus."

 

Well, strictly speaking this is a physical impossibility. It maybe in sufficient focus for a human eye to skip over the differences - but only ONE distance will give you a true focus.

the bell curve analogy above is good: there's still one focus point but the slope is the difference. In practical terms, with an aperture of f/8 and a 14mm you can get pretty huge DOF. However, that doesn't necessarily translate to "sharpness" because that is a function of the lens itself, the design, aberrations, etc...

The key to understand DOF is that everything inside of it can't be in "perfect" focus. Everything in the DOF will be in "acceptable" focus. Who decides what's "acceptable"? Well . . . The truth is that scientists in a lab made the determination of just how far out of focus something could be and still be considered "in focus". The trouble is that we very rarely photograph in a lab.