Does wide open tally with the F no. for dof.

Janice

Janice

Messages
11,626
Name
Janice
Edit My Images
Yes
Stupid question.... sorry!! :D

Say I have two lenses, both prime 100mm. One is an F2.8 and one is an F4.

If I pick up the F2.8 and want to take a portrait, I am likely to want to use F4 on it to get a bit more depth of field.

Now if I pick up the F4 lens, and I use it at F4 like I did the first one, will the dof be the same, or will it be really shallow becuase F4 is its widest, and not just because it is F4?

Ie: does F4 at 100mm give the same dof regardless if it is the widest a lens can go to or not.

Does that make sense?
 
Ah thank you for that! Something I have always wondered! :notworthy:
 
Yes, remember that the diameter of the aperture opening is focal length / value.

100mm/4 = 25mm so on both lenses the aperture opening will be 25mm and it's the size of that will define the DoF not the f/ value on it's own.
 
But, what does crop and non crop mean as it seems there are differences on how a 1.4L image looks on FF - I have often thought short fast glass performs better on crop sensor cameras.
 
Janice said:
Ah thank you for that! Something I have always wondered! :notworthy:
Click to expand...

That has just caused me to look at my lenses.
I wondered why you asked that question.
Something is missing and I never realised until now. There is no depth of field scale on the barrels. Wonder when that disappeared from lenses.
 
Do you mean 100mm on FF compared 100mm on a crop?

A thorny subject but if you look at the FF version of

50mm, f/2, 100cm distance on a 1.6x crop

you get

80mm, f/3.2, 100cm distance

and both give the same DoF with a 25mm aperture diameter.

Shoot both at the same aperture and the FF will have less DoF.
 
pxl8 said:
Do you mean 100mm on FF compared 100mm on a crop?

A thorny subject but if you look at the FF version of

50mm, f/2, 100cm distance on a 1.6x crop

you get

80mm, f/3.4, 100cm distance

and both give the same DoF with a 25mm aperture diameter.

Shoot both at the same aperture and the FF will have less DoF.
Click to expand...


Just seems 'different'....
 
Yeah, I know what you mean but I think it's mainly because you don't use a crop and consider the crop factor or aperture shift when shooting.
 
pxl8 said:
Do you mean 100mm on FF compared 100mm on a crop?

A thorny subject but if you look at the FF version of

50mm, f/2, 100cm distance on a 1.6x crop

you get

80mm, f/3.4, 100cm distance

and both give the same DoF with a 25mm aperture diameter.

Shoot both at the same aperture and the FF will have less DOF.
Click to expand...
Incorrect. DOF is controlled by the lens alone. A 50mm lens FF will have a certain DOF the same 50mm lens on a cropped shot camera will have the same size subject on either sensor BUT when you print out, the cropped sensor will have to be enlarged giving the impression that a larger focal length lens was used. Now here is the rub, C of C will also have been enlarged
giving a smaller DOF. Remember Cof C decides the DOF
 
But in my example it's not the same lens, the FF is using an 80mm lens ;)
 
pxl8 said:
But in my example it's not the same lens, the FF is using an 80mm lens ;)
Click to expand...
Your example for the change in aperture value is incorrect, a lens cannot change its aperture value once set. The focal length only seems to increase due to the enlargment of the sensor size when viewing or printing. So you cannot use your calculation.
 
I think we're talking about completely different things here.

a lens cannot change it's aperture value once set.
Click to expand...

Lost me completely.

The objective of the example was to get the same shot with both cameras. The FF uses an 80mm lens so that the subject is framed the same. Now in order to also match the DoF the aperture value must also be changed. On the crop body the aperture diameter is 50/2=25 so 80/25=3.2 is the aperture you'll need on the 80mm.

Put the numbers into a DoF calculator and you'll see both give the same result. The calculation includes the CoC so that the results are correct when both shots are printed the same size.

If you want to use a 50mm lens on the FF body just change the working distance from 100cm to 62.5cm, you'll still need to stop down to f/3.2 though to get the same DoF

You also work out the change in aperture simply by multiplying the f/ on the crop shot by 1.6, f/2*1.6=f/3.2
 
Trying to keep it simple ;) there are just two things that affect depth of field - f/number and image size.

So in the OP's example, since the f/number and focal length are the same, provided the focusing distance is also the same to maintain identical image size, then depth of field will be the same.

Taking this a step further, it you change the lens focal length but move forwards or backwards so that the subject is still the same size each time, then depth of field remains the same because the image size hasn't changed.

If you only use one camera with the sensor size obviously fixed, then this is all you need to know. Don't read any further! However, if you also use full-frame or a compact camera, then you do need to understand a bit more.

Full-frame gives a larger image, therefore the image is bigger and depth of field is reduced. With full-frame vs crop format, the difference is about 1.3 stops. It's actually the crop factor, so f/4 on a Canon 1.6x crop will give the same depth of field as f/6.4 on full frame (4x1.6=6.4).

The opposite is true of compacts that have tiny sensors, and therefore produce small image size. With these cameras depth of field tends to be huge and shallow depth of field effects are correspondingly hard to create. How much depth of field they have compared to DSLRs is hard to say as sensor formats vary a lot, but they all give greater depth of field to the tune of several stops.

In passing, it is maybe helpful to note that you can use the effect that changing sensor format has on depth of field, especially when you want a bit more. Just move back a bit! Or use a shorter focal length. This will obviously give a smaller image size, but when you enlarge it back up to size in post processing, there will be more depth of field even when the f/number remains the same. I find this very useful in macro, when depth of field is always too small! Of course, you have to be careful that you have enough system resolution so that when you do blow it up a bit bigger in post, image quality is still good.
 
With you until the last paragraph, more on that in a minute.

A better way to understand why smaller sensors have more DoF is to consider magnification. As they're smaller they need less magnification from the lens to fill the frame with a given subject size, eg. 50mm frame on 1.6x crop and 80mm on FF. With compacts the sensor is so small that lenses can have a focal length of just a few mm, the magnification factor is very low which means greater DoF.

The size of the sensor defines how much magnification is needed which in turn defines the DoF (along with aperture).

Back to that final paragraph.

Moving back or reducing the magnification will increase DoF but the more you crop the more the CoC will play a factor. The smaller the sensor the smaller the CoC to start with and the less enlargement it can take. I suspect that the reduced magnification will be offset by the increased enlargement of the CoC (assuming you directly swap one for the other) meaning you won't actually get any more DoF after all.
 
pxl8 said:
With you until the last paragraph, more on that in a minute.

A better way to understand why smaller sensors have more DoF is to consider magnification. As they're smaller they need less magnification from the lens to fill the frame with a given subject size, eg. 50mm frame on 1.6x crop and 80mm on FF. With compacts the sensor is so small that lenses can have a focal length of just a few mm, the magnification factor is very low which means greater DoF.

The size of the sensor defines how much magnification is needed which in turn defines the DoF (along with aperture).

Back to that final paragraph.

Moving back or reducing the magnification will increase DoF but the more you crop the more the CoC will play a factor. The smaller the sensor the smaller the CoC to start with and the less enlargement it can take. I suspect that the reduced magnification will be offset by the increased enlargement of the CoC (assuming you directly swap one for the other) meaning you won't actually get any more DoF after all.
Click to expand...

With you until the last paragraph ;)

You say "I suspect that the reduced magnification will be offset by the increased enlargement." And that is entirely logical thinking, but the effect of format reduction outweighs the opposite effect of increased enlargement.

Honest! The difference is quite dramatic with macro, which is why I mention it. And it's actually very easy to check if you have a macro lens and a ruler. Photograph the ruler at 1:1, then again at 1:2. Check the depth of field by looking at the LCD (it's easy to see even at that small size). Then enlarge the 1:2 pic to the same size and see how much more of the ruler is in focus.

There's even a double benefit to this as with macro at 1:1 light is reduced by 2 stops, but shooting at 1:2 it's only 1 stop, so you can afford to close the f/number down even further :D

I'm digressing here, but I've started so I'll finish.

If you take what's been said above about image size (magnification) and f/number being the only factors that affect depth of field, how come telephotos are said to reduce depth of field, and wide-angles increase it?

Technically they don't, but when you move back with a telephoto or long zoom, the angle of view changes and you get much less backgound in the picture, behind the subject. This does not affect depth of field, but it certainly looks like it and the main subject will stand out much more strongly. The reverse is true of wide-angle photographs.
 
I've not read this whole thread but I think I get the gist..............surely the only reason the DoF is a smaller when a 50mm lens at f2 is used on a FF body is because when the lens is used on a crop camera the subject would need to be positioned further away to get the same field of view - hence a greater DoF.
 
HoppyUK said:
Honest! The difference is quite dramatic with macro, which is why I mention it. And it's actually very easy to check if you have a macro lens and a ruler. Photograph the ruler at 1:1, then again at 1:2. Check the depth of field by looking at the LCD (it's easy to see even at that small size). Then enlarge the 1:2 pic to the same size and see how much more of the ruler is in focus.
Click to expand...

DoF for 1:1, f/11 is 1.5mm
DoF for 1:2, f/11 is 4.5mm

Clearly quite a difference but in order to end up with the same shot you need you crop the 1:2 shot to 25% of the original, that means the CoC will be enlarged by a factor of 4 on the cropped version. 4.5/4 = 1.125mm. So on paper it seems that shooting at 1:2 and then cropping won't win you any DoF but I'd be interested to see the ruler shots you've tested it with.
 
Gofer said:
I've not read this whole thread but I think I get the gist..............surely the only reason the DoF is a smaller when a 50mm lens at f2 is used on a FF body is because when the lens is used on a crop camera the subject would need to be positioned further away to get the same field of view - hence a greater DoF.
Click to expand...

To get the same framing with a crop camera, yes, you need to move further away to get the same field of view (also known as angle of view). And obviously on a crop camera the image on the sensor is much smaller, so depth of field is increased.

That is a characteristic of small vs larger formats, and as stated above the difference between full frame and crop cameras, is about 1.3 stops (rule of thumb for Nikon 1.5x and Canon 1.6x formats). Just multiply the f/number by the crop factor to get the equivalent depth of field.
 
pxl8 said:
DoF for 1:1, f/11 is 1.5mm
DoF for 1:2, f/11 is 4.5mm

Clearly quite a difference but in order to end up with the same shot you need you crop the 1:2 shot to 25% of the original, that means the CoC will be enlarged by a factor of 4 on the cropped version. 4.5/4 = 1.125mm. So on paper it seems that shooting at 1:2 and then cropping won't win you any DoF but I'd be interested to see the ruler shots you've tested it with.
Click to expand...

Doing the maths is dangerous with macro! Confuses the heck out of me :D Maybe you're getting mixed up between linear and area enlargement? Anyway, if what you say is true, then depth of field would remain the same regardless of format, and we surely agree that it is not!

Anyway, just try it. The difference is as absolutely clear. The difference between depth of field at 1:1 and 1:2 is about double in the finished print. That's the rule of thumb I use, and it's a very handy trick with macro. You can't push it too far though, especially with crop format cameras, as you are giving the image a big pull up in post. I would much rather get some more light in there and stop down further to get the depth right in the first place, but sometimes that's not possible.

One of the reasons I chose crop format over full frame is because you get fundamentally more depth of field. However, I must say that I'm attracted to the latest round of very hi-res full frame camera, not becuase they have massively more resolution that I could possible ever need, but because when you crop them down to, say, the same as crop cameras, you still have stacks of image quaility to play with.

This is great for sport or wildlife or any action photography where accurate framing is difficult. I find the extra depth of filed on enlargements usually a bonus rather than a negative, and I also use cropping as a way of sorting out converging verticals in-camera - it's just like using rising front. When we see the kind of sensor developments that Fuji has recently shown in compacts, such as pairing up pixels to give substantially extended dynamic range, or enhanced low light performance - not to mention all the other amazing digital imaging enhancements we almost take for granted already, then the 20mp-plus camera makes real sense :)
 
HoppyUK said:
Doing the maths is dangerous with macro! Confuses the heck out of me :D Maybe you're getting mixed up between linear and area enlargement? Anyway, if what you say is true, then depth of field would remain the same regardless of format, and we surely agree that it is not!
Click to expand...

Area enlargement as it should be. If the 1:1 image was 100x100px then the area would be 1000 pixels. A 1:2 shot would need to be cropped to 50x50px which has an area of 250px - a quarter of the original.

The DoF figures came from a great article about macro and DoF here:
http://xoomer.virgilio.it/ripolini/Close_up.pdf

The question of DoF and format size is an interesting one. The CoC for a 1.6x crop is 1.6x smaller than a FF. This means details need to be 1.6x smaller on the crop to have the same sharpness as the FF if both are printed to the same size. But if you keep the print sizes proportional to the sensor size then the CoC would be the same - So a 10" print from FF and a 6.25" print from a 1.6x crop would have the same DoF for the same lens and aperture. Both have been enlarged by the same amount.

The CoC on crop cameras is smaller simply because it's assumed the image will be enlarged more to start with therefore you need smaller details (CoC) to withstand the enlargement.

The argument for pixel density on a crop body works but only if all the pixels are in focus to start with, if they're not then you've got extra density in an OOF are that has no detail and so has no benefit.
 
HoppyUK said:
Trying to keep it simple ;) there are just two things that affect depth of field - f/number and image size.
Click to expand...

So in the OP's example, since the f/number and focal length are the same, provided the focusing distance is also the same to maintain identical image size, then depth of field will be the same.
[/QUOTE]

There are actualy three things that affect DOF

Aperture
Focal length
Distance to Subject (focus point)
 
EOS_JD said:
There are actualy three things that affect DOF

Aperture
Focal length
Distance to Subject (focus point)
Click to expand...

Yes, but you can merge Focal length and Distance into one thing, and call it Image size or Magnification. And I think it makes things clearer and simpler when you do, as it reveals the truth behind the common misconception that long lenses produce less depth of field. They don't - they increase image magnification which reduces depth of field, just as moving closer increases magnification. The controlling factor is magnification (or image size, whatever).

I have also pointed out that this only holds true for a given sensor format, eg full frame or crop format. And I have further explained the effect that angle of view has on perceived depth of field.

pxl8, do you accept that smaller sensors produce more depth of field on a given print size, eg with a compact? Surely you must. Then what I have suggested above, by effectively reducing fomat and using 'digital zoom' to restore the print size, is doing exactly the same thing. If you have a macro lens, it will take you two minutes to see it.
 
HoppyUK said:
pxl8, do you accept that smaller sensors produce more depth of field on a given print size, eg with a compact? Surely you must. Then what I have suggested above, by effectively reducing fomat and using 'digital zoom' to restore the print size, is doing exactly the same thing. If you have a macro lens, it will take you two minutes to see it.
Click to expand...

Well, maybe. It depends how and what you compare it to.

1.6x, 50mm, f/2, 100cm = 2.89cm DoF
FF, 50mm, f/2, 100cm = 4.56cm DoF

In this example the image on the 1.6x has more magnification compared to the FF image.

1.6x, 50mm, f/2, 160cm = 7.54cm DoF
FF, 50mm, f/2, 100cm = 4.56 DoF

Now the subject has the same apparent size on both sensors but effectively has less magnification on the crop sensor, lower mag = greater DoF.

The interesting thing here is that if you took the first example but printed so that the subject was the same physical size, eg. print the FF at 10" and the crop at 6.25", then the DoF would be the same for both prints, .

So does crop have more, less, or the same DoF really depends on what you're comparing as it can be all 3 ;)

Magnification (assuming aperture diameter is constant) rules the day, the less you magnify the image on the sensor or print the greater the DoF will be.

Which brings me back to the point about macro. If you magnify half as much at the sensor stage but twice as much at the print stage they cancel each other out. Pixel density has no effect on DoF because, as we've seen, it's magnification that matters.
 
pxl8 said:
Well, maybe. It depends how and what you compare it to.
Click to expand...

Not really. If you reduce the image size on the sensor, but then restore it to the final size you want by enlarging it in post processing, then depth of field is always increased.

The format used, whether crop or full frame, is irrelevant, so 1.6x or whatever doesn't matter as it has no bearing here, and neither does focal length or focusing distance. Only the size of the image on the sensor matters, however that is acheived, and nothing else.

If you try to work it through using conventional DoF calcs with focal length and focusing distance, you can get in a pickle. Maybe if you input only magnifcation will get the right answer. Just a guess.

pxl8 said:
Which brings me back to the point about macro. If you magnify half as much at the sensor stage but twice as much at the print stage they cancel each other out.
Click to expand...

But they don't, which is why you should try it. This is how crop format delivers greater depth of field than full-frame, and compacts with tiny sensors even more. That is exactly the same principle as I am sugesting here.

Reducing the image size on the sensor dramatically increases depth of field, whereas increased enlargement of the finished print has much less effect, so the two do not cancel out. I know it sounds weird, but that's how it is.

If you want the maths to add up, which of course it does, then work it back from this assumption and you'll find it's ok. Sorry, but I haven't checked your calcs to see where you're at.

Andy, I see from your profile that you don't list a macro lens, which is a shame given that lovely bag of lenses you've got :) Maybe you have some tubes that will give you both a 1:1 image, and a 1:2 image for comparison.

Of course what I am saying here applies to all photography, and not only macro - it's just easier to see it with the reduced depth of field, and dramatic fall-off you automatically get with macro. But to see the effect in practise I'm sure you can cobble something together - just shoot a subject where it is easy to see the depth of field and get as close as you can for a frame-filler, then move back to double the distance, take another pic and compare the depth of field. It will have shot up.

Of course, the image will also have degraded so I am not suggesting this as an everyday technique. It's handy when all else fails, and works well if you don't push it too far. You obviously need a high-res sensor and very sharp lenses to hold everything together as much as possible for a good print.
 
So here's your test.


lifesize.jpg

1:1 or thereabouts

Then again but reduced mag to around 1:2 and then cropped and enlarged to match.

cropped.jpg


The DoF looks about the same to me, I know the angle has changed slightly as well as the centre point but I think it's pretty clear the 1:2 shot hasn't magically produced any extra DoF
 
HoppyUK said:
Yes, but you can merge Focal length and Distance into one thing, and call it Image size or Magnification. And I think it makes things clearer and simpler when you do, as it reveals the truth behind the common misconception that long lenses produce less depth of field. They don't - they increase image magnification which reduces depth of field, just as moving closer increases magnification. The controlling factor is magnification (or image size, whatever).

I have also pointed out that this only holds true for a given sensor format, eg full frame or crop format. And I have further explained the effect that angle of view has on perceived depth of field.

pxl8, do you accept that smaller sensors produce more depth of field on a given print size, eg with a compact? Surely you must. Then what I have suggested above, by effectively reducing fomat and using 'digital zoom' to restore the print size, is doing exactly the same thing. If you have a macro lens, it will take you two minutes to see it.
Click to expand...

Not sure that would be correct.... Magnification has nothing to do with distance to subject.

Long lenses do increase magnfication but I can use a 200mm lens at 4' and a 200mm lens at 40ft and get very different results.

All three parts need to be separated. Aperture, focal length (also called, as you say, magnification) and distance to subject.
 
EOS_JD said:
Not sure that would be correct.... Magnification has nothing to do with distance to subject.

Long lenses do increase magnfication but I can use a 200mm lens at 4' and a 200mm lens at 40ft and get very different results.

All three parts need to be separated. Aperture, focal length (also called, as you say, magnification) and distance to subject.
Click to expand...

I don't understand what you're saying. If you move closer to the subject, it gets bigger. That's magnification.

The sensor doesn't care if that increased magnification is achieved by moving closer, or by using a longer lens. It doesn't change depth of field.

You do not have to separate focal length and distance in order to calculate depth of field, and I think it makes the concept easier to understand if you don't. If you separate out focal length in this way, it implies that focal length alone per se affects depth of field. It doesn't, and if you fit a long lens but then move back to maintain image size (magnification) then depth of field remains unaltered.
 
pxl8 said:
So here's your test...

...The DoF looks about the same to me...
Click to expand...

That's not a good test angle at all! So yes, of course it looks the same! You can't see depth of field change when you are shooting an almost flat plane subject! You need to reshoot the test. We are talking about a couple of mms here, and it's not easy to measure it unless you deliberately devise a clear test.

That shot is looking down, and the rule is at 45 degs. The depth of field is moving under the ruler and across it, not along it. To see the effect clearly, the rule needs to be directly along the plane of depth of field, not going across it, and not from almost directly above. You need to shoot low, as close to the plane of the rule as possible, and with the rule going directly away from you.
 
HoppyUK said:
That's not a good test angle at all! So yes, of course it looks the same! You can't see depth of field change when you are shooting an almost flat plane subject! You need to reshoot the test. We are talking about a couple of mms here, and it's not easy to measure it unless you deliberately devise a clear test.
Click to expand...

Er, ok... but I was expecting that as...

The difference is quite dramatic with macro
Click to expand...

..I would see a dramatic difference. If it doesn't show up on that simple test then, even if there is a not so dramatic difference, I can't see what possible use it would be when shooting macro :shrug:

That's the best angle I can get with tubes, the MFD is just too short and I need to introduce the angle in order to get anything in focus.

You've got a couple of macro lenses so why not post up some examples as I suggested earlier?
 
pxl8 said:
You've got a couple of macro lenses so why not post up some examples as I suggested earlier?
Click to expand...

A cynic might think that you were trying to prove me wrong ;)

But I can assure you it works, so purely in the interests of science I'll try and post a couple of pics later.

I don't have any software loaded on this PC so I'll have to see what the wonders of Microsoft and PhotoBucket can do.

Good morning!
 
pxl8 said:
Er, ok... but I was expecting that as...



..I would see a dramatic difference. If it doesn't show up on that simple test then, even if there is a not so dramatic difference, I can't see what possible use it would be when shooting macro :shrug:

That's the best angle I can get with tubes, the MFD is just too short and I need to introduce the angle in order to get anything in focus.

You've got a couple of macro lenses so why not post up some examples as I suggested earlier?
Click to expand...

OH! Janice, look what you have started :lol:
 
I think Janice - and ALL 'normal' peeps for that matter ;) - would have been happy with post #2

:lol:

While the rest of this thread/argument is interesting I'm guessing it has little relevance to the next shot you take

Keep it going though lads, looking forward to seeing a 'winner'; presently both camps are entrenched and holding their ground well, could the next move be decisive :eek:

:D

DD
 
pxl8 said:
Well, maybe. It depends how and what you compare it to.

1.6x, 50mm, f/2, 100cm = 2.89cm DoF
FF, 50mm, f/2, 100cm = 4.56cm DoF

In this example the image on the 1.6x has more magnification compared to the FF image.

1.6x, 50mm, f/2, 160cm = 7.54cm DoF
FF, 50mm, f/2, 100cm = 4.56 DoF

Now the subject has the same apparent size on both sensors but effectively has less magnification on the crop sensor, lower mag = greater DoF.

The interesting thing here is that if you took the first example but printed so that the subject was the same physical size, eg. print the FF at 10" and the crop at 6.25", then the DoF would be the same for both prints, .

So does crop have more, less, or the same DoF really depends on what you're comparing as it can be all 3 ;)

Magnification (assuming aperture diameter is constant) rules the day, the less you magnify the image on the sensor or print the greater the DoF will be.

Which brings me back to the point about macro. If you magnify half as much at the sensor stage but twice as much at the print stage they cancel each other out. Pixel density has no effect on DoF because, as we've seen, it's magnification that matters.
Click to expand...

I vowed never to enter this debate, but I cringe when I read some of the posts. Here is a start for your macro.
DOF for macro is
2 * A* C of C * (M+1)/ M^2
Here is a fact. No matter what you use, FF or CROP the image of subject on either sensor will be the same size. :clap:
 
HoppyUK said:
A cynic might think that you were trying to prove me wrong ;)
Click to expand...

I'm just asking you to demonstrate the effect you've described, I tried but didn't shoot from the right angle so seeing a shot from the right angle would be useful for me to understand how the effect works.

I'm quite happy for you to prove that you're right and I'm wrong, clearly one of us has to be. Seeing the effect means I can experiment and learn from it, it's a bit hard to do that if I just rely on your assertion.

Oldfella, clearly you know something on the subject so instead of looking down your nose why not enlighten us?
 
I am not looking down my nose, I think its good fun. I am not allowed to add attachments to my posts, which makes it a bit difficult.
 
Janice's fault - she thinks about stuff far too much. :D
 
You must log in or register to reply here.
Back
Top