How can some lenses focus past grates?

[NMGPhotos]

I recently got a film SLR and was playing around with it at the bus stop earlier today. The bus stops shelters where I'm at have, on one side, a coarse grate-like design with circular holes. There is a lot of material between the holes. However, the lens I had was able to focus past it and onto the objects behind it very easily to the point where, at least through the viewfinder, you wouldn't be able to tell it was even there. Granted, it was late at night and perhaps objects beyond it being lit more helped, but I am curious as to the physics of this.

My phone camera was not able to focus past it (although I didn't expect it to be able to).

Edited January 17, 2023 by NMGPhotos

[AJG]

Almost all film SLRs can be focused manually, so you get to choose what is in focus rather than the AI on a chip in your phone.  The other aspect is that phone cameras have very short focal length lenses in comparison with a 35 mm SLR's normal 50 mm lens and the shorter the focal length the more depth of field you will have, which means that objects close to the camera and those far away can be in focus at the same time. That 50 mm lens will probably have a wide open f/stop like f1.8 or f/1.4 which has very little depth of field when used at its widest opening. This means that when you focus on the scene far behind the grate that the grate itself will be so far out of focus that it all but disappears in the picture. You would notice the the grate did cause some softness in a print if you got that far, and if you photographed a similar scene without a grate near the camera to interfere that it will be sharper.

[Alan Marcus]

Tempted to say this is diffraction (path of light rays moving around an obstacle). However, this is a too tiny a path change to explain this phenomenon. More likely: The circular holes you are shooting through are likely acting as a diaphragm. The iris can be forward of the lens or behind or between the lens elements. Think about the tiny aperture opening of the camera lens and the fact that despite its tiny diameter a complete image is realized. 

[NMGPhotos]

1 hour ago, AJG said:

Almost all film SLRs can be focused manually, so you get to choose what is in focus rather than the AI on a chip in your phone.

I understand that. Just for your information though, I did try using a manual focusing app and even focused as far as the app would let me and the grate was still was clearly visible. Again, not particularly surprising, just wanted to add that.

1 hour ago, AJG said:

 The other aspect is that phone cameras have very short focal length lenses in comparison with a 35 mm SLR's normal 50 mm lens and the shorter the focal length the more depth of field you will have, which means that objects close to the camera and those far away can be in focus at the same time. That 50 mm lens will probably have a wide open f/stop like f1.8 or f/1.4 which has very little depth of field when used at its widest opening.

According to spec sheets, the lens on my phone is a fixed ~f/1.7 (which is surprising, as that should lead to a very shallow depth of field?). Could the size of the lens also play a factor?

1 hour ago, AJG said:

This means that when you focus on the scene far behind the grate that the grate itself will be so far out of focus that it all but disappears in the picture.

If it's (at least partially) solid, why would it disappear instead of obstructing the view? At what point does a grate become an obstruction you can't focus past?
 

1 hour ago, AJG said:

You would notice the the grate did cause some softness in a print if you got that far, and if you photographed a similar scene without a grate near the camera to interfere that it will be sharper.

Thanks for this info! That does make sense.

31 minutes ago, alan_marcus2 said:

Tempted to say this is diffraction (path of light rays moving around an obstacle).

That does sound like a plausible explanation.

32 minutes ago, alan_marcus2 said:

More likely: The circular holes you are shooting through are likely acting as a diaphragm. The iris can be forward of the lens or behind or between the lens elements. Think about the tiny aperture opening of the camera lens and the fact that despite its tiny diameter a complete image is realized. 

That is also very interesting. They are small-ish holes. I still just find it fascinating that it's able to disappear like that.

[Gary Naka]

Post a picture of the grating.
That would make evaluation a LOT easier than guessing what it looks like.

[rodeo_joe1]

The physics of this are a bit complicated, but two easily explainable points are:

1. A shorter focal length lens and a small sensor size - such as in a phone-camera - have a much greater depth of field than larger sensors and longer focal length lenses of the same aperture. Therefore your phone-camera is unable to throw the grating out of focus enough to become invisible. 

2. Light radiates away from objects (after reflection) as if the object was a collection of tiny point sources. Such that distant objects radiate light towards you in almost parallel bundles of rays. These parallel bunches of light pass through holes more easily than light that's closer to the holes and will hit the holes 'side on' - if that makes sense. Like a pencil or your finger, that can easily poke straight through a hole, while a pointed cone cannot. Meaning distant light rays can 'poke' through the grating and hit the camera lens much more easily than light radiating from just the other side of the grating. And when the camera lens is focused at a distance it can catch these near-parallel pencils of light and reassemble them into a sharp(ish) image. 

Not a physically perfect analogy, but close enough I think. 

Edited January 17, 2023 by rodeo_joe1

[Tony Parsons]

The same principle applies when photographing captive animals - the closer you can get (safely !) to the bars, netting or whatever protects the animals from the humans, and the wider the aperture you use, the better chance you have of not recording the obstruction on film / sensor. 

 

 

 

[AJG]

You mentioned that your phone camera lens was f/1.7.  Did the specs say what the focal length was? Most phone lenses are very short in focal length because the chips on which they record their images are very small and because most phones (the ones with only one lens) take a fairly wide field of view.  Wide angle lenses on any format are always a shorter focal length than normal lenses and thus will have more depth of field. With very short focal lengths diffraction can be a problem so the ability to stop down is frequently mechanically limited to prevent the issue from leading to soft results. Not having a changeable f/stop will also make the camera cheaper to manufacture. 

[NMGPhotos]

6 hours ago, AJG said:

You mentioned that your phone camera lens was f/1.7.  Did the specs say what the focal length was?

Looking at specs it appears to be 27mm (full-frame equivalent). This does still seem a bit strange, as a shallow depth of field should let me focus past the grate? I would assume it's perhaps the size of the lens in relation to the size of the holes.

18 hours ago, Gary Naka said:

Post a picture of the grating.
That would make evaluation a LOT easier than guessing what it looks like.

Apologies! Attached is a picture I took with my phone camera. As mentioned, my SLR lens could easily focus past it and to the bus behind it.

[glen_h]

I suspect it you look carefully, you will see some effect on the image.

But yes, if you are close enough, it is completely out of focus.  I have tried this

with window screens, not a grate as coarse as yours.

 

With AF cameras, you have to be careful, as the AF system can focus on the grate!

 

I have a Panasonic camcorder that I had out one time.  When I wasn't shooting,

I noticed a strange look on the viewing screen.  (That is, fold out LCD.)

With the zoom all the way to wide angle, there is enough depth of field

that it could (almost) focus on the dust on the skylight filter!

 

But for the same reason as your grate, in ordinary shooting you won't

notice dust on the lens of filter.  (Or, in the case of the Canon

Pellix, on the mirror.). Though you want the aperture close to wide open.

[NMGPhotos]

I am still curious as to why my phone camera was unable to focus past the grate (even when focused manually). Not that I expect it to, just wondering about possible reasons.

[AJG]

27 mm equivalent focal length is probably something more like 4 or 5 mm actual focal length or even less. With a focal length that short you could still have enough depth of field for the grate to be in focus even though you have focused on a distant scene behind it. As great as phone cameras can be, if you want control of the final image you are better off starting with a more sophisticated camera that allows full manual operation of focus, exposure, etc.

[Ken Katz]

I have never used a smart phone app to manually focus the camera, so I can't speak about why that feature did not seem to work.  The 27mm FF equivalent main camera has a really short actual focal length, as AJG describes above, so that even at f1.7, DOF is still quite significant and would result in a quite visible grate as per your example.  Frankly, using almost any camera and lens, such a dense grate would likely be visible in the image, even if you manually focused on the objects beyond the grate.  

I do take photos at zoos (Bronx and San Diego), and deal with various types of barriers between me and the subjects.  Some have little noticible interference, but many still result in a noticible barrier in the files.  Dirty and scratched glass/plexi can also be problematic.  I use a 150mm lens at f5.6 on m43 format, which is not ideal from a blurred OOF foreground/backround, but has far, far, less DOF than a cell phone camera.

FYI, at this time smart phone cameras all have fixed lens openings.  With the main WA lens, smaller apertures are generally not needed.

Edited January 18, 2023 by Ken Katz

[NMGPhotos]

15 hours ago, AJG said:

27 mm equivalent focal length is probably something more like 4 or 5 mm actual focal length or even less.

What are you basing this calculation on? Just curious.

15 hours ago, AJG said:

With a focal length that short you could still have enough depth of field for the grate to be in focus even though you have focused on a distant scene behind it.

It was my understanding that the f-number has effect on DOF, but can focal length have effect as well?

15 hours ago, AJG said:

As great as phone cameras can be, if you want control of the final image you are better off starting with a more sophisticated camera that allows full manual operation of focus, exposure, etc.

Completely understand and completely agree. Was just curious about the differences.

 

23 minutes ago, Ken Katz said:

Frankly, using almost any camera and lens, such a dense grate would likely be visible in the image, even if you manually focused on the objects beyond the grate.

That is interesting. Looking through the viewfinder, I was unable to see the grate. I don't know if it would have shown up on film as I didn't have any at the time.

[petrochemist]

There are multiple factors that affect how visible the mesh is.

The closer the lens is to the mesh the better

The further the subject is from the mesh the better

The longer the (actual) focal length the better

The wider the aperture the better.

With a reasonably long telephoto wide open & touching chicken wire I shot this:

 

The mesh behind the bird is visible but I can't see anything in front.

At motorsports there is frequently safety fencing that you can't get close to (or it would provide less protection). Broadly similar settings but around 5 foot from the mesh gave:

 

The mesh doesn't detract too much but it can be seen

Edited January 18, 2023 by petrochemist
lack of BBcode support

[AJG]

8 hours ago, NMGPhotos said:

What are you basing this calculation on? Just curious.

It was my understanding that the f-number has effect on DOF, but can focal length have effect as well?

Completely understand and completely agree. Was just curious about the differences.

 

That is interesting. Looking through the viewfinder, I was unable to see the grate. I don't know if it would have shown up on film as I didn't have any at the time.

Focal length is also crucial in determining depth of field.  When your specs said 27 mm equivalent on full frame, that is a wide angle lens on 24 x 36 mm digital chip cameras, the same dimensions as classic Leica film cameras and 35 mm SLR cameras.  The chip in your phone is a lot smaller--4x6 mm(?) maybe, so to take in an equivalent angle the focal length of the phone camera lens must be much shorter than 27 mm.  Other things being equal, shorter focal lengths will take in a wider angle and have more depth of field at a given f/stop than longer focal lengths.

[rodeo_joe1]

A phone camera is about 1/10th of the size of a 'proper' camera. Everything scales down, (or up in the case of depth-of-field) and according to my depth-of-field spreadsheet, the D-o-F of a 5mm lens at f/1.7 is about 18m with a 4m focus; while a 50mm lens @ f/1.7 has only 0.6m focus-depth with the same subject distance. 

In short, a phone camera has almost no selective focus ability, and no way to throw the grating out-of-focus. 

OTOH, the lens in the phone camera is probably small enough to 'see' completely through one of those grating holes if poked close enough to it. Whereas the bigger SLR lens will always be 'looking' at some of the surrounding metal. 

There are online depth-of-field calculators where you can see the effect of different focal-lengths and apertures. It'll be instructive to plug different numbers into one and see how the D-o-F is affected. 

This is the first such calculator that Google threw at me. You need to select a specific camera rather than having a free choice of format size, but it does the job. 

Edited January 19, 2023 by rodeo_joe1

[NMGPhotos]

12 hours ago, rodeo_joe1 said:

according to my depth-of-field spreadsheet, the D-o-F of a 5mm lens at f/1.7 is about 18m with a 4m focus;

What is the 4mm focus in reference to, and also how is D-o-F calculated like this?

[rodeo_joe1]

On 1/19/2023 at 10:16 PM, NMGPhotos said:

What is the 4mm focus in reference to, and also how is D-o-F calculated like this?

It's 4 meters, not millimeters. That's the subject distance. 

The calculation of depth-of-field is a complicated formula that involves calculating the interception of the film or sensor plane with the cone of light rays projected by the lens. Where the cone diameter equals something called a 'circle of confusion' - honest, no kidding - is deemed to be the limit of acceptable focus. 

This 'circle of confusion' is dependent on the film or sensor frame size. E.G. For the 36mm x 24mm 'Full-frame" size it's usually set at 0.029mm, but it's a fairly arbitrary choice, since some people will think a larger circle of blur is acceptable, and others that it's too much. 

The angle of the cone of rays varies with the aperture diameter and the focal-length of the lens (plus any lens extension due to focussing). So you can see that a lot of variables have to be taken into account by any accurate D-o-F formula. Which is why I leave it to my spreadsheet to calculate. 

It all starts with the basic formula for conjugate-focii (subject-to-lens, and lens-to-image distances) 

The formula is:

1/f = 1/v + 1/u

Where f is the lens focal length, v is the lens to image distance, and u is the distance from lens to subject; all expressed in the same units. With millimetres being the most preferred unit of measurement. 

Starting with that basic formula and a bit of trigonometry, you can work out the out-of-focus light cone intersection with the image-plane and calculate the corresponding subject distances for near and far limits of focus. Easy! (Not) 🤔

 

Edited January 21, 2023 by rodeo_joe1

[glen_h]

On 1/19/2023 at 1:38 AM, rodeo_joe1 said:

 

(snip)

In short, a phone camera has almost no selective focus ability, and no way to throw the grating out-of-focus. 

(snip)

But there is another way, which is to view through just one hole.  (If you are close enough.)

 

Some years ago, there was a device which would allow nearsighted people to see

(near objects and for a short time).

It was a metal rectangle with holes in it. 

If you hold it close to your eye, you see through one hole.

Small enough for lots of depth of field, large enough to see through.

And as your eye moves, it looks through a different hole.

 

But usually, I make a small hole with my finger, and look through that.

[rodeo_joe1]

On 1/21/2023 at 11:11 AM, glen_h said:

But there is another way, which is to view through just one hole.  (If you are close enough.)

 

On 1/19/2023 at 9:38 AM, rodeo_joe1 said:

OTOH, the lens in the phone camera is probably small enough to 'see' completely through one of those grating holes if poked close enough to it.

 

[David_Cavan]

Shooting at race tracks sometimes puts you behind the fences, often these are chain-link if they aren't in a dangerous location. I've found that very often I can shoot "through" those fences, especially if panning, so that the fencing magically disappears. That's especially better if you can get real close, and the fence itself is not lit up too much.  It does not always work, but when it does the results are surprising.

 

[NMGPhotos]

On 1/21/2023 at 5:04 AM, rodeo_joe1 said:

Starting with that basic formula and a bit of trigonometry, you can work out the out-of-focus light cone intersection with the image-plane and calculate the corresponding subject distances for near and far limits of focus. Easy! (Not) 🤔

I'm pretty good with math, lol! do you know of any comprehensive guides (perhaps maybe with visuals) that show how this works?

[rodeo_joe1]

6 hours ago, NMGPhotos said:

I'm pretty good with math, lol! do you know of any comprehensive guides (perhaps maybe with visuals) that show how this works?

OK, you asked for it! 

First we have to look at how lenses focus, and where the conjugate focii formula comes from. 

For the sake of depth-of-field it's always assumed that focussing is done by changing the distance between lens and film, or lens and sensor - the 'focal plane'. Conventionally, the distance between lens and focal plane is designated the letter 'v', and the distance between lens and subject is called 'u'. The focal-length of the lens is assigned the letter 'f'. 

In order for very distant ('infinite) subjects to be in focus the lens needs to be nearer to the focal-plane than for closer subjects, and vice-versa. That is; to be in sharp focus, near subjects require the lens to be further away from the focal-plane than for distant subjects. v is always => f, the focal-length of the lens. 

Now, (and here's where the trigonometry comes in), we can imagine the light projected by the lens from any point on a subject as a perfect cone, i.e. an isosceles triangle rotated through its vertex. And conversely we can treat the cone of light as a 2 dimensional isosceles triangle. 

So if we zoom in on the focal plane shown in the diagram above, and with the lens focused on an intermediate distance subject, we get this 2 dimensional model - 

We already know the physical size of the lens aperture from f/n, where f is the lens focal-length and n is the relative aperture (f-number). While the circle-of-confusion diameter is got from a look-up table. The aperture diameter divided by the C-O-C diameter gives us the values of delta-v, and hence v1 and v2, by similar triangles

The distance of the focal-plane from the lens, v, is got from using the conjugate focii formula and the subject distance. So, that's all the information we need to work out delta-v, and from there the values of v1 and v2.

When v1 and v2 are plugged back into the conjugate focii formula we get u1 and u2, which are the limits of far and near 'sharp' focus. 

Some of the maths cancels out and there are a few shortcuts that make the calculation(s) easier. One of which is to know that the subject 'magnification' is given by v/u.

Well, my head's spinning now, going over all that again.

Good luck getting your own head around it! 

Edited January 24, 2023 by rodeo_joe1

[ozzey]

On 1/17/2023 at 9:56 AM, rodeo_joe1 said:

The physics of this are a bit complicated, but two easily explainable points are:

1. A shorter focal length lens and a small sensor size - such as in a phone-camera - have a much greater depth of field than larger sensors and longer focal length lenses of the same aperture. Therefore your phone-camera is unable to throw the grating out of focus enough to become invisible. 

2. Light radiates away from objects (after reflection) as if the object was a collection of tiny point sources. Such that distant objects radiate light towards you in almost parallel bundles of rays. These parallel bunches of light pass through holes more easily than light that's closer to the holes and will hit the holes 'side on' - if that makes sense. Like a pencil or your finger, that can easily poke straight through a hole, while a pointed cone cannot. Meaning distant light rays can 'poke' through the grating and hit the camera lens much more easily than light radiating from just the other side of the grating. And when the camera lens is focused at a distance it can catch these near-parallel pencils of light and reassemble them into a sharp(ish) image. 

Not a physically perfect analogy, but close enough I think. 

well explained:) the reason why your film SLR camera was able to focus past the grating while your phone camera was not is due to the difference in aperture size between the two cameras, the SLR camera lens has a larger aperture size which allows more light in and focus better in low light conditions and has a shallower depth of field.