Is full frame worth it?

Guys, depth of field does not exist absolutely, or in isolation. It exists only within a defined set of parameters and, to repeat, the one that I think is confusing some folks here is the necessity for a standard print viewed from a standard distance, eg A4 viewed from the same distance as the diagonal, which is roughly 15in.

The moment you start cropping bits from a larger image, you either have to enlarge that area more to re-establish output size, or you have to view it from a closer distance. Both change the perceived DoF.

For example, when you look at something small on the camera's LCD, it usually looks pretty sharp, but when you view it on the computer screen, suddently it doesn't. If you then pixel peep and look at a small section at 100%, it gets even less sharp because you're far too close and you should be viewing from several feet away, or even yards, to re-establish parity.

The concept of DoF is simply addresses this question - what is the smallest detail that can be seen by the naked eye on a standard size print, from a standard distance? That is defined as the maximum Circle of Confusion. Everything is worked back from there, to calculate the fineness of detail that must be imaged on the sensor so that, when everything is subsequently enlarged up to final output size in a print, the CoC size is achieved.

You can check all this at www.dofmaster.com in a few seconds. Input the four parameters - sensor size (camera type, which changes the CoC in the bottom right hand corner), focal length, distance and f/numbe. Then out pops the answer, always bearing in mind that this assumes the standard size print/viewing distance scenario.

When you change any one parameter, DoF changes, and sometimes not in the way you think as has happened above. So what is the best comparison? What best represents parity to most people? I would say (and the logic is hard to argue against) that the best comparsion is to take 'the same' picture - same subject, framed the same, from the same position (to maintain perspective). When you do that with two cameras with different size sensors (and you have to adjust the focal length to maintain framing) you find that the smaller sensor delivers greater depth of field. The difference between full frame and Canon 1.6x crop camera is about one and a quarter stops. The formula is simply f/number x crop factor, so for example f/8 on full frame gives the same DoF as f/5 on the cropper.

For reference, Nikon/Sony crop is 1.5x, MFT is 2x, and most quality compacts are about 5x and correspondingly deliver massive DoF.

Edit: the same concept is used in magazine and newspaper printing, which of course uses dots to fool our eyes into seeing a continuous tone picture. View a full magazine page at normal distance and you can't see the dots, but in smaller images, viewed more closely, you can. Same principle as DoF.
 
Last edited:
I have a 5D2 :)

I sometimes upload photos to Facebook (yes I know... :nono: ) I've noticed that my pics have a distinctly different look to the normal party photos that people put up, and it took me a while to realise why.... depth of field.

Most of my pics have something out-of-focus in the frame - some recent landscape shots I did had out-of-focus foliage in the foreground, whcih I only really noticed when I uploaded the pics. It's an effect that I like and I tend to take for granted, having moved straight from 35mm film photography to a full-frame sensor (5D, then the 5D2). For me, this is the biggest advantage of full-frame - maintaining that classic "35mm" look. The effect becomes less pronounced the smaller the sensor is - my other half's G9 has the ability to manually select aperture, but there is no discernable difference at all in terms of DoF - everything in the shot apart from anything very close to the lens is sharp.

Narrow DoF is a pretty reliable indicator of a decent camera setup - i.e. large sensor size and large aperture lenses.

A.
 
You can lead a horse to water but you can't make it drink.....
 
one thing to maybe consider is pc software and hardware - primarily photoshop or lightroom and a PROPER monitor. I have a friend who has a 5dmk2 and the problem is that when he comes to edit the pictures he has no idea if the colours and contrast are acurate, even with proper callibration. The whites tend to still be blown and colours can just look wrong because the monitor is cheap and just plain bad. I can't emphasise enough how much of a problem this has been for him, he had to redo all his website images on someone else's monitor after he saw how burnt out the images were and how blotchy the processing was.

So the moral of this story is getting a fab camera is great, but if you can't process any images confidently then you're in trouble. Personally it was one of the reasons I went for the 7d and 24-70 lens, because once you factor in software, monitor, cards, batteries and all the other bits you also need, well I just couldn't get the 5D. One day I'll hopefully go full frame but not yet and I'm glad I went the way I did.
 
Last edited:
If you need ultrawides the Canon 16-35mm is not the best performer so choices there may be limited.
 
Ever wonder what the S means in EF-S?

S = Short back, the rear element is closer to the sensor with an EF-S lens, this results in a smaller, lighter lens. The down side is that you get a much larger depth of field when compared to a similar focal length EF lens. This is the reason FF has a shorter DoF than APS-C etc, people beleive it is down to the larger sensor, but its actually down to the EF-S lens geometry.
 
Lostgear said:
Ever wonder what the S means in EF-S?

S = Short back, the rear element is closer to the sensor with an EF-S lens, this results in a smaller, lighter lens. The down side is that you get a much larger depth of field when compared to a similar focal length EF lens. This is the reason FF has a shorter DoF than APS-C etc, people beleive it is down to the larger sensor, but its actually down to the EF-S lens geometry.
Click to expand...

Sorry but that doesn't make sense as if you used an ef lens on a crop instead of ef-s it would then be the same as on ff?
 
Lostgear said:
Ever wonder what the S means in EF-S?

S = Short back, the rear element is closer to the sensor with an EF-S lens, this results in a smaller, lighter lens. The down side is that you get a much larger depth of field when compared to a similar focal length EF lens. This is the reason FF has a shorter DoF than APS-C etc, people beleive it is down to the larger sensor, but its actually down to the EF-S lens geometry.
Click to expand...

No.
 
HoppyUK said:
Alan, and (the other) Andy.

Maybe the bit you are missing is that the concept of DoF requires a print of a given size to be viewed at a distance equal to the diagonal. So if you simply take a small section out of a larger image, you have to view it more closely. That's what changes, and the DoF with it.
Click to expand...

Couldn't resist just harping back to this one last time but by looking more closely you are increasing magnification which isn't keeping all other things the same so you could expect a further change. I think though this is a point of view thing, pardon the pun.

I think fair to say if you take the same picture on a ff and crop you are likely to see less dof in the ff image. Apologies again for my two penny
 
I heard the smaller DOF is actually due to magic only present in full frame camera's. Its got nothing to do with sensor size or distance between the sensor and lens glass. It's magic. And you can't argue with,or properly explain magic. Simple.
 
Morph3ous said:
I heard the smaller DOF is actually due to magic only present in full frame camera's. Its got nothing to do with sensor size or distance between the sensor and lens glass. It's magic. And you can't argue with,or properly explain magic. Simple.
Click to expand...

At last someone has finally explained it properly. I pledge myself to your teachings o great potentate.
 
Lostgear said:
Ever wonder what the S means in EF-S?

S = Short back, the rear element is closer to the sensor with an EF-S lens, this results in a smaller, lighter lens. The down side is that you get a much larger depth of field when compared to a similar focal length EF lens. This is the reason FF has a shorter DoF than APS-C etc, people beleive it is down to the larger sensor, but its actually down to the EF-S lens geometry.
Click to expand...
Can you provide the maths to back this up?

Bob
 
Morph3ous said:
I heard the smaller DOF is actually due to magic only present in full frame camera's. Its got nothing to do with sensor size or distance between the sensor and lens glass. It's magic. And you can't argue with,or properly explain magic. Simple.
Click to expand...

Not magic but imps a FF camera has a better class of imp....so i have been told by people on the internet
 
HoppyUK said:
The concept of DoF is simply addresses this question - what is the smallest detail that can be seen by the naked eye on a standard size print, from a standard distance? That is defined as the maximum Circle of Confusion. Everything is worked back from there, to calculate the fineness of detail that must be imaged on the sensor so that, when everything is subsequently enlarged up to final output size in a print, the CoC size is achieved.

You can check all this at www.dofmaster.com in a few seconds. Input the four parameters - sensor size (camera type, which changes the CoC in the bottom right hand corner), focal length, distance and f/numbe. Then out pops the answer, always bearing in mind that this assumes the standard size print/viewing distance scenario.

When you change any one parameter, DoF changes, and sometimes not in the way you think as has happened above. So what is the best comparison? What best represents parity to most people? I would say (and the logic is hard to argue against) that the best comparsion is to take 'the same' picture - same subject, framed the same, from the same position (to maintain perspective). When you do that with two cameras with different size sensors (and you have to adjust the focal length to maintain framing) you find that the smaller sensor delivers greater depth of field. The difference between full frame and Canon 1.6x crop camera is about one and a quarter stops. The formula is simply f/number x crop factor, so for example f/8 on full frame gives the same DoF as f/5 on the cropper.
Click to expand...

No, and here's why...

The reason we got into this was because I questioned and opposed the statement that sensor size changes DoF when it's clear to me at least (and many others on the net) that it does not and furthermore I believe that anyone with two cameras with different sized sensors and similar lenses (and an open mind) can easily prove this to themselves.

Of course, the properties of the image are one thing and how you view and perceive the image is something else.

When you set the camera up and press the shutter button the image is set. It is what it is. You can then display it on screen or print it small, medium or large or cut bits out of it and view it with your nose pressed against it or from the back of the garden but these things only change your perception of the image and do not and can not change the actual image itself. That was set when you pushed the shutter button.

Printing a big image will make what we see as DoF / focus (or lack of) obvious. Printing a small image will hide any lack of DoF or focus and make out of focus things look sharper. BUT, these things are just how we perceive the image in that state and under those conditions and are not the actual attributes of the image.

When conducting my little tests what I've tried to do is fairly compare the technical properties of two images and make any DoF / focus differences as obvious as possible. What I've done is take a picture with a full frame camera and a 20mm lens and from the same position take a picture with a MFT camera with a 20mm lens. Both at the same aperture.

The reason I've done this is to make things simple as I've changed only one thing... the format size.

To change other things such as the lens zoom length or the distance between the camera and the subject is to introduce other factors and complicate the issue.

Changing just the format size to prove or disprove that format size does not affect DoF makes sense to me.

When changing just the format size (remember how we got into this? Format size affects DoF? Yes or No?) the resultant images will upon first glance look different because the full frame camera produces an image that has a wider field of view but upon careful examination of the images it is clear that the actual DoF is the same. To make this more obvious I took a heavy crop and equalised the sizes of the images so that the viewer is not distracted by subject size within the frame of field of view. The result is there to be seen. The DoF is the same and the minor differences in the images are just contrast, saturation, sharpness and rendering etc.

Cropping either an on screen or printed image is just taking scissors to it and can not change the things within it like focus. All cropping does is allow is to focus on the subject, the DoF in the image at the same size.

This is how it should be. DoF and focus are properties of distance and aperture and of course lens design but assuming that the lens design is competent that should not be a major factor.

If you use two different format sizes, the same (or very similar lenses) and maintain camera to subject distance I believe..... (drum roll...)... Format Size (in itself and by itself) Does Not Affect DoF and if you do your own test I believe that you'll believe that too :lol:

However, and I've said this before... although format size does not in itself affect DoF it will influence your choices and how the final image will look. For example, because of your format size you may shoot with a particular lens and aperture and these things will affect how you frame the shot and your distance from the subject and these will affect how the final image looks.

With regard to CofC. I think that at normal-ish print and viewing sizes and distances it probably shouldn't be an issue. CofC is the size of disk that can be recorded and viewed and considered to be sharp. Consider a sensor with large pixels and how photons will be collected and the resultant image that is formed. Now consider smaller pixels, is it possible that the CofC's could overlap? What would that do to the image? As the pixel size is reduced further perhaps the disks would no longer overlap? What will that do to the image?

This simple thought process should show that CofC could potentially change with pixel size/density across sensors of the same or different size and that these factors should be considered when calculating the CofC for sensors of different pixel size and density of the same and different format size.

However. IMVHO this whole CofC discussion is secondary and perhaps almost completely irrelevant in a discussion about focus and / DoF differences between format sizes from MFT to FF especially when we consider the technical capabilities of these cameras and that the images produced when conducting a little test like mine or in everyday use and how the images are commonly viewed and / or printed is within the technical capabilities of the cameras or at worst not too far outside. Once it matters it should be pretty obvious that things have been pushed too far.

Those who think that format size or zoom length in themselves and by themselves affect DoF don't need to change what they do because it's the results that matter not what's actually technically correct or true or not as they influence hardware and composition choices and help to create and image in which things such as DoF are more or less obvious.

Obviously I'm of the opinion that it's nice to know what's really going on rather than what simply appears to be going on.

I'll try not to bring this up again but I do hope that I've prompted some people to think and to maybe try a few little tests themselves.
 
But 20mm on a FF sensor isn't the same as 20mm on a MF camera?! Surely the frame composition is completely different. You'd need to move forward/back or change focal length to compare like for like. Same reason that a 50mm on FF is more or less 35mm on a crop sensor.
 
For the love of Mike!! I can't believe some people just can't understand this, it's so flipping simple!!

Put a 50mm lens set to f1.8 on a F/F camera, set it to focus on an object 6' from you and your DOF will be 5".
Do the same on a crop camera and compose the shot so the same image is in the viewfinder (i.e move back approx. 3'). The object you're focusing on is now 9' away and your DOF will be greater, at 7". But you will have the same image in both cameras, the crop sensor with more DOF.

Now stop arguing and go take some pictures :)
 
treeman said:
For the love of Mike!! I can't believe some people just can't understand this, it's so flipping simple!!

Put a 50mm lens set to f1.8 on a F/F camera, set it to focus on an object 6' from you and your DOF will be 5".
Do the same on a crop camera and compose the shot so the same image is in the viewfinder (i.e move back approx. 3'). The object you're focusing on is now 9' away and your DOF will be greater, at 7". But you will have the same image in both cameras, the crop sensor with more DOF.

Now stop arguing and go take some pictures :)
Click to expand...

Yep, I think treeman has hit the nail on the head!
 
Lostgear said:
Ever wonder what the S means in EF-S?

S = Short back, the rear element is closer to the sensor with an EF-S lens, this results in a smaller, lighter lens. The down side is that you get a much larger depth of field when compared to a similar focal length EF lens. This is the reason FF has a shorter DoF than APS-C etc, people beleive it is down to the larger sensor, but its actually down to the EF-S lens geometry.
Click to expand...

Biggest load of tosh I have read on here for ages :bang:
 
In answer to the question, yes I thought full frame was worth it. To me it was like someone taking the blinkers off. As for DoF, I'm looking forward to this drawing ;)
 
Lostgear said:
Not with maths, never my good point, but I think a diagram could explain it? Leave it with me for a day or 2 I'll se what I can do.
Click to expand...

Looking forward to it Adam. FWIW, I don't hold with your theory but I'm always ready to look at different viewpoint.

EdBray said:
Biggest load of tosh I have read on here for ages :bang:
Click to expand...
You're on the fence about this one then Eddie?

Bob
 
woof woof said:
No, and here's why...
Click to expand...

There is nothing in what I have said above that you could possibly disagree with.

I'll try not to bring this up again but I do hope that I've prompted some people to think and to maybe try a few little tests themselves.
Click to expand...

With the greatest respect bud, you need to go right back to the basic concept of DoF, its relevance, and how it is applied.

treeman said:
For the love of Mike!! I can't believe some people just can't understand this, it's so flipping simple!!

Put a 50mm lens set to f1.8 on a F/F camera, set it to focus on an object 6' from you and your DOF will be 5".
Do the same on a crop camera and compose the shot so the same image is in the viewfinder (i.e move back approx. 3'). The object you're focusing on is now 9' away and your DOF will be greater, at 7". But you will have the same image in both cameras, the crop sensor with more DOF.

Now stop arguing and go take some pictures :)
Click to expand...

The DoF might be as you describe, but in moving back you have changed perspective so the two images are not the same. To get a like for like comparison, you need to adjust focal length from the same viewpoint.

Lostgear said:
Not with maths, never my good point, but I think a diagram could explain it? Leave it with me for a day or 2 I'll se what I can do.
Click to expand...

No, seriously, don't. You are completely mistaken.

The main difference with EF-S lenses is that the image circle projected is smaller, as it only needs to cover a smaller sensor. That's why crop format lenses are smaller and lighter, amongst other benefits.

The short back focus bit is just a side effect of crop format cameras needing a smaller mirror than full frame, so there is a bit more clearance behind the lens. There are optical benefits* in utilising this extra space (less retro-focus necessary, but it's only a modest advantage here) which Canon has chosen to exploit, though Nikon and third-party makers don't make use of it with their lenses specifically crop format lenses.

*Edit: depth of field has got nothing to do with it.
 
Last edited:
HoppyUK said:
The DoF might be as you describe, but in moving back you have changed perspective so the two images are not the same. To get a like for like comparison, you need to adjust focal length from the same viewpoint.
Click to expand...

Ok, I thought you might say that, so put a 35mm on the crop sensor in order to stay the same distance, and guess what, the DOF is greater that way too :)
 
Seriously, does it really matter? Just use the damn things!
 
Canon Bob said:
Looking forward to it Adam. FWIW, I don't hold with your theory but I'm always ready to look at different viewpoint.


You're on the fence about this one then Eddie?

Bob
Click to expand...

:lol: Err, No!

I am willing to have a large bet that a 17-40L has the same DOF as a 17-55 IS or even a kit lens at comparable focal lengths on the same crop body focused on the same spot.

Its nothing to do with the architecture of the lens, a Carl Zeiss 150mm is a 150mm whether it is a Sonnar or a Tessar and they will both have similar characteristics focused on the same distance and at the same aperture on the same camera.
 
treeman said:
Ok, I thought you might say that, so put a 35mm on the crop sensor in order to stay the same distance, and guess what, the DOF is greater that way too :)
Click to expand...

Yes!
 
treeman said:
For the love of Mike!! I can't believe some people just can't understand this, it's so flipping simple!!

Put a 50mm lens set to f1.8 on a F/F camera, set it to focus on an object 6' from you and your DOF will be 5".
Do the same on a crop camera and compose the shot so the same image is in the viewfinder (i.e move back approx. 3'). The object you're focusing on is now 9' away and your DOF will be greater, at 7". But you will have the same image in both cameras, the crop sensor with more DOF.

Now stop arguing and go take some pictures :)
Click to expand...

Nice succinct explanation treeman. As you've clarified, once you normalise the field of view (which is essential to composition) the DoF changes.
 
scottthehat said:
if you have an (set at f2.8)
50mm on a 1.5x crop it will give you 75mm in 35mm temrs, so if you focus on something 10 foot away the dof will be = 42.195cm

and full frame
50mm on a full frame is 50mm, same again focus on something 10 foot away
the dof will be = 63.636cm so more dof then then the crop.

But to get the same field of view as the crop camera on the full frame use a 75mm lens and focus again at 10 foot away and the dof will be = 28.056cm which is 12cm less dof then the crop camera with the same field of view.
Click to expand...
 
Mark does have a point!
 
Whilst we're on the subject ;) think on this. A lens only has a single 2D plane which is precisely in focus. Anything in front or behind that plane will be out of focus. The amount it is out of focus will depend on how far from that plane the object is (and aperture and lens focal length as well). The TRUE depth of field of a lens is actually infinitesimally small - it's just whether you see that when you print it. That's why sensor size comes into it - for the same image, you have to enlarge it less with the full frame sensor compared to a smaller sensor to get to the same sized print. It really is that simple....
 
arad85 said:
Whilst we're on the subject ;) think on this. A lens only has a single 2D plane which is precisely in focus. Anything in front or behind that plane will be out of focus. The amount it is out of focus will depend on how far from that plane the object is (and aperture and lens focal length as well). The TRUE depth of field of a lens is actually infinitesimally small - it's just whether you see that when you print it. That's why sensor size comes into it - for the same image, you have to enlarge it less with the full frame sensor compared to a smaller sensor to get to the same sized print. It really is that simple....
Click to expand...

Yes :)

I don't want to drag this out unnecessarily, but I hope this is helpful. Depth of field is an optical illusion, not quite magic, but certainly something of a visual trick.

It is simply based on the fact that the human eye can only detect a certain level of detail in an image (determined by the diameter of the circle of confusion) when viewed from a 'normal' distance. Same principle as the dots that make up a printed magazine page, or the dots on a TV screen or the monitor you're looking at now.

Anything smaller can't be detected, and so falls within the range of depth of field. If the viewing distance is changed, or the size of the print is changed, or any other apsect of magnification is changed (sensor size, focal length, subject distance, and f/number*) then the DoF changes with it.

* f/number is also technically an aspect of magnification, because it directly affects the size of the CoC formed by the lens.
 
There are some good reasons for buying a FF camera. Good IQ is, for me the best reason. If that's important to you go and buy a FF camera and take great pictures with it. For god's sake you can deal with the DOF. Go and take pictures. I only read about six posts on this thread before I wanted to stab someone in the eye and then take two high IQ pictures of them on a FF camera. One with shallow DOF and another with deep DOF achieved by twiddling the knobby things on the camera.
 
chrisgeary said:
*sigh*
Click to expand...

You and me both Chris - personally I don't give a ****** - I just like to take photos of things that interest me and if the pic I have needs more bokeh to isolate the subject I just reach for the blur brush.

Simples.

.
 
You must log in or register to reply here.
Back
Top