Help With DOF

[KimThomas]

Hi, I'm a beginner to photography and i have come across some problems understanding the basics of DOF and what it even is. And how to change it. So as you can see i'm in need of some enlightening. Your help would be most appreciated.

 

[alpina]

Hi, I'm a beginner to photography and i have come across some problems understanding the basics of DOF and what it even is. And how to change it. So as you can see i'm in need of some enlightening. Your help would be most appreciated.

Look here

www.cambridgeincolour.com

 

[Sparkles33]

Depth of field, in blunt terms... shallow depth of field is focusing in on something and making either the background or foreground blurry. This is done by having a wide aperture (low number aperture) it lets more light into the camera.

http://www.shortcourses.com/use/using1-9.html is a great link, it really helped me understand DOF and aperture.

 

[HoppyUK]

This tutorial was written with folks like you in mind. There's a section on depth of field post #9 http://www.talkphotography.co.uk/forums/showthread.php?t=99841

 

[mrtoad]

This tutorial was written with folks like you in mind. There's a section on depth of field post #9 http://www.talkphotography.co.uk/forums/showthread.php?t=99841

yes better to read a well thought out explanation

not so confusing circularly speaking

 

[AliB]

I simply remember that for apertures, a small number = a big hole = small depth of field. So f2.8 means a stonking great hole to let the light in and a very shallow depth of field. Simples

 

[mrtoad]

I simply remember that for apertures, a small number = a big hole = small depth of field. So f2.8 means a stonking great hole to let the light in and a very shallow depth of field. Simples

its good to understand whats causing these phenomenon...and the relationship with focal lengths as well

if you can tight-rope niagra you can walk to the co-op blindfolded

:bonk:

 

[AliB]

I can just about make it to the co-op without the blindfold, the zebra crossing is lethal!

I know there is more to it than that but that's exactly what my starting point was, one step at a time on that tightrope

 

[KimThomas]

Ok, thanks for all your replies, ill take a look at some of the tutorials.

 

[KimThomas]

So as i've read. DOF is basically the depth of the picture which is in focus. The lower the f/ number the smaller the section of the image that is in focus, and the higher f/ number the larger the section of the image that is in focus.

Am i correct?

 

[mrtoad]

I can just about make it to the co-op without the blindfold, the zebra crossing is lethal!

I know there is more to it than that but that's exactly what my starting point was, one step at a time on that tightrope

absolutely...a march of a thousand miles starts with one step...mao tse tung

 

[mrtoad]

So as i've read. DOF is basically the depth of the picture which is in focus. The lower the f/ number the smaller the section of the image that is in focus, and the higher f/ number the larger the section of the image that is in focus.

Am i correct?

only one point is in focus...to use a term which the eye relates to
there are other areas not so in focus but the human eye cant actually differentiate these too well
there comes a point..or points where the eye says..out of focus
this occurs somewhere in front and behind the actual focal point..
those distances vary depending on focal length of the lens and aperture
and are loosely termed....depth of field...the field of vision in focus or acceptable focus when viewed with the human eye..sensibly
all very mathematically able to be understood
the phenomenon is/are called the circle..or circles of confusion
when these circles get too big your eye cant reassemble them into an image
out of focus..confused
the circles are in some way similar to pixel size on the image
the finer and smaller the more sharp the image appears

thats my take on it..

the wiki link explains it more in line with optical jargon

 

[Forbiddenbiker]

So as i've read. DOF is basically the depth of the picture which is in focus. The lower the f/ number the smaller the section of the image that is in focus, and the higher f/ number the larger the section of the image that is in focus.

Am i correct?

Yes on the first bit ...near on the second.

As an alternate view point expanding on MC's post, try thinking of the aperture as a filter restricting the quantities of spread light rays against parallel light rays for a given exposure.

Then remember that when a lens is focused on an object it is in fact adjusting the light rays from that subject to hit the sensor as near parallel light rays as possible to render the subject in sharp focus.

Parallel light rays mean the tiny points of light called Circles of confusion will appear in focus at the sensor surface.
Less parallel light rays mean the CoC will start to appear out of focus.
totally spread light rays mean the CoC is also spread and will appear totally out of focus.

Then remember that the distance to the object being focused is also going to affect the spread rays against parallel light ray quantities.

Close focus with a wide aperture (low number, say f2.8 around 30mm's ) can cause areas away from the focal point to contain lots of spread CoC. (A nice short DOF) Correspondingly using the same wide aperture f2.8 and focusing on a subject 100 feet away will greatly decrease the amount of spread CoC and will render much of the scene in focus especially beyond the focal point all the way to infinity ...simply because all of the scene beyond the focus point now also contains parallel light rays causing in focus tight CoC at the focal plane/ the sensors surface.

Its also noted that using very closed apertures (high numbers ) can cause the parallel rays of light to spill or diffract around the aperture blades causing the in focus tight CoC to diffract into miss focus. …This is called diffraction thankfully, unlike the out of focus Coc which is called bokeh, lol …yes that's spelled correctly.

I do hope that makes some sence, its not easy to put into words ...

 

[DiddyDave]

its not easy to put into words ...

Obviously







Soz - couldn't help myself

Oh - and he's right too

DD

 

[Forbiddenbiker]

Morning to you too...

 

[Suvv]

Then remember that when a lens is focused on an object it is in fact adjusting the light rays from that subject to hit the sensor as near parallel light rays as possible to render the subject in sharp focus.

Parallel light rays mean the tiny points of light called Circles of confusion will appear in focus at the sensor surface.
Less parallel light rays mean the CoC will start to appear out of focus.
totally spread light rays mean the CoC is also spread and will appear totally out of focus.

Sorry but I do not think that this is correct.
The lens bends light rays so that rays from a single point on the subject, which spread out and hit all parts of the lens, are bent by different amounts depending on where on the lens the ray hits and therefore all rays coming from a single point on the subject meet at the same point on the sensor - they are focused. They are not parallel.
Rays from point closer than the subject are focused before the sensor and they spread out again to form circles.
Rays from points further than the subject are focused after the sensor but they hit the sensor before they can be focused and once again they form circles.
This article, and particularly the diagram, explains it well.
http://en.wikipedia.org/wiki/Circle_of_confusion

 

[Forbiddenbiker]

No prob, well its possibly badly formed but being as you've not given my words any credibility at all, cos your also right in my eyes, I have to assume you don't understand the effect fully yourself, soz.

That Wiki article explains it very badly I've always thought.

Let me reason with you if I can, why near parallel makes for good long DOF .. with a question.

You've just focused your light rays accurately on your focal plane, your using f16 on a landscape, you chose a tree not far away for foreground interest and that’s your focal point. Your focus is bang on and back at the sensors surface the points of light / COC is spot on 'in focus'... the results are perfect, back home everything’s in focus lovely, prints out massive no problem loads of DOF very sharp.

Considering you only focused on the tree why is the grass nearby and the mountain in the background also in pin sharp focus. Becasue the focused light that you made meet on the sensor by adjusting the lens was only from the tree?

 

[Forbiddenbiker]

Sorry but I didn't mean to start a questioning but I could think of I way too answer you...

so I came up with that.

:shrug:

 

[Suvv]

No prob, well its possibly badly formed but being as you've not given my words any credibility at all, cos your also right in my eyes, I have to assume you don't understand the effect fully yourself, soz.

That Wiki article explains it very badly I've always thought.

Let me reason with you if I can, why near parallel makes for good long DOF .. with a question.

You've just focused your light rays accurately on your focal plane, your using f16 on a landscape, you chose a tree not far away for foreground interest and that’s your focal point. Your focus is bang on and back at the sensors surface the points of light / COC is spot on 'in focus'... the results are perfect, back home everything’s in focus lovely, prints out massive no problem loads of DOF very sharp.

Considering you only focused on the tree why is the grass nearby and the mountain in the background also in pin sharp focus. Becasue the focused light that you made meet on the sensor by adjusting the lens was only from the tree?

I agree about the Wiki article - the top RH diagram is OK, as for the rest . In fact the 2nd diagram is wrong.


OK in answer to your question - The grass nearby and the background are not actually in pin sharp focus. The small circles (Circles of Confusion) that they are made up of are very, very, very small though and to your eye they appear to be pin sharp.

The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens.

Is that what you meant by parallel, that the light rays allowed through by the stopped down aperture will all be a similar angle to the centre axis of the lens?

 

[Suvv]

No prob, well its possibly badly formed but being as you've not given my words any credibility at all, cos your also right in my eyes, I have to assume you don't understand the effect fully yourself, soz.

That Wiki article explains it very badly I've always thought.

Let me reason with you if I can, why near parallel makes for good long DOF .. with a question.

You've just focused your light rays accurately on your focal plane, your using f16 on a landscape, you chose a tree not far away for foreground interest and that’s your focal point. Your focus is bang on and back at the sensors surface the points of light / COC is spot on 'in focus'... the results are perfect, back home everything’s in focus lovely, prints out massive no problem loads of DOF very sharp.

Considering you only focused on the tree why is the grass nearby and the mountain in the background also in pin sharp focus. Becasue the focused light that you made meet on the sensor by adjusting the lens was only from the tree?

I agree about the Wiki article - the top RH diagram is OK, as for the rest . In fact the 2nd diagram is wrong.


OK in answer to your question - The grass nearby and the background are not actually in pin sharp focus. The small circles (Circles of Confusion) that they are made up of are very, very, very small though and to your eye they appear to be pin sharp.

The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens.

Is that what you meant by parallel, that the light rays allowed through by the stopped down aperture will all be a similar angle to the centre axis of the lens?

 

[Suvv]

Duplicate

 

[mrtoad]

I agree about the Wiki article - the top RH diagram is OK, as for the rest . In fact the 2nd diagram is wrong.


OK in answer to your question - The grass nearby and the background are not actually in pin sharp focus. The small circles (Circles of Confusion) that they are made up of are very, very, very small though and to your eye they appear to be pin sharp.

The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only :shake:..is this right...light comes from all places equally...lens wise otherwise it could be sad

Is that what you meant by parallel, that the light rays allowed through by the stopped down aperture will all be a similar angle to the centre axis of the lens?

 

[mrtoad]

I agree about the Wiki article - the top RH diagram is OK, as for the rest . In fact the 2nd diagram is wrong.


OK in answer to your question - The grass nearby and the background are not actually in pin sharp focus. The small circles (Circles of Confusion) that they are made up of are very, very, very small though and to your eye they appear to be pin sharp.

The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens. is this really correct

Is that what you meant by parallel, that the light rays allowed through by the stopped down aperture will all be a similar angle to the centre axis of the lens?

 

[Forbiddenbiker]

I agree about the Wiki article - the top RH diagram is OK, as for the rest . In fact the 2nd diagram is wrong.


OK in answer to your question - The grass nearby and the background are not actually in pin sharp focus. The small circles (Circles of Confusion) that they are made up of are very, very, very small though and to your eye they appear to be pin sharp.

The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens.

Is that what you meant by parallel, that the light rays allowed through by the stopped down aperture will all be a similar angle to the centre axis of the lens?

Yes that is was partially what meant in my first sentence, and the reason behind it . ta for getting back to that.

Not simply because there to the centre axis of the lens though, the aperture doesn't block any part of the scene, as you close down you get less of the wider reflected rays .... compensate exposure etc...leaving you with more of the nearer to parallel CoC in comparison, and so on as you stop down further.

..as you know.

edit soz, slow reply very poor browser here..

 

[Suvv]

Yes that is was partially what meant in my first sentence, and the reason behind it . ta for getting back to that.

Not simply because there to the centre axis of the lens though, the aperture doesn't block any part of the scene, as you close down you get less of the wider reflected rays .... compensate exposure etc...leaving you with more of the nearer to parallel CoC in comparison, and so on as you stop down further.

..as you know.

edit soz, slow reply very poor browser here..

Thanks for that - I have found that in a lot of these discussions people are saying the same thing but in completely different ways. It was only when I was writing my reply that I thought about the rays being a smaller angle to the centre axis and therefore could be considered parallel.

 

[Forbiddenbiker]

Ive done it myself a few times Im sure. its very easy to missunderstand

 

[Suvv]

mrcrow
The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens. is this really correct


Yes it is right. The rays from a single point on the subject, say a pore on a face, spread out like a solid cone and hit every part of the front of the lens. The lens then bends the light by different amounts depending on where on the lens the light hits, all of the light then focuses at a single point. If the camera had been focused on the face then the sensor will be where all of the light is focused. As you close down the aperture then light rays from the edges of the lens will be blocked so not so much light from the pore reaches the sensor so it will be darker.
Any points that are nearer than the pore will be focused nearer than the sensor and the rays will continue and spread out and reach the sensor as a circle (Circle of Confusion).
Any points that are further than the pore will be focused further than the sensor and the rays will reach the sensor as a circle (Circle of Confusion).
That circle will be made smaller if the aperture is closed down as the aperture will not allow light from the edges of the sensor through.
I am not sure if my diagram helps or not - it is a bit crude. S is a point on the subject, the orange line is the aperture, you have to imagine where the sensor is.
If the sensor is where the focused rays meet then the point will be in focus and aperture will just block some rays and make the point darker.
If the sensor is behind where the focused rays meet then you can see that the rays have spread out again and the point will not be in focus - it will be a circle, if you now imagine some of the rays being stopped by the aperture the spread out circle will be smaller. The smaller the aperture the more rays that are blocked and the smaller the circle - the more the point will seem in focus to the eye.
A similar argument applies if the sensor is in front of where the focused rays meet.
I know that it is not the sensor that moves but too much detail on the diagram makes it even more confusing. Just imagine each case - in focus, nearer than the focus point and further than the focus point seperately.

 

[HoppyUK]

All this confusion started when the OP said this:

So as i've read. DOF is basically the depth of the picture which is in focus. The lower the f/ number the smaller the section of the image that is in focus, and the higher f/ number the larger the section of the image that is in focus.

Am i correct?

Yes, that is 100% correct. At least as far as it goes (the other consideration is magnification). But FB replied with:

Yes on the first bit ...near on the second.

Which is not correct

Thanks for that - I have found that in a lot of these discussions people are saying the same thing but in completely different ways....

Are they?

Light rays are only ever parallel at infinity, and infinity changes distance according to focal length. Either way, I don't think the concept of parallel rays is helpful to a discussion about depth of field.

In practise, the relevant considerations are f/number and magnification which hasn't been touched on.

There are two things which change magnification and affect depth of field. Move closer, the subject gets bigger, magnification increases, and depth of field is reduced. Likewise, if you fit a longer focal length lens, the subject gets bigger etc and depth of field is reduced. These two factors cancel out in that when image size is maintained, by fitting a longer lens but then moving back so that the subject remains the same size in the frame, depth of field also remains the same.

If you change camera format, that also changes magnification (bigger format, less depth of field).

 

[Forbiddenbiker]

What confusion, more like a bit of misunderstanding.

You say its 100% correct, except for the other considerations ...so its nearly correct then?

Hoppy your such a hair splitter.... :nono:

A


Light rays are only ever parallel at infinity, and infinity changes distance according to focal length. Either way, I don't think the concept of parallel rays is helpful to a discussion about depth of field.

Don't you, well I think it will be of interest myself. My tangent of an opinion on this subject might be more usefull than re-reading the same old replies for some.

 

[HoppyUK]

What confusion, more like a bit of misunderstanding.

Well I was confused by your explanation. Still am TBH.

You say its 100% correct, except for the other considerations ...so its nearly correct then?

Hoppy your such a hair splitter.... :nono:

What the OP said was 100% correct, as far as was stated. Not nearly correct. Who is splitting hairs?

The OP wanted to know what controls depth of field, so I tried to explain it.

Don't you, well I think it will be of interest myself. My tangent of an opinion on this subject might be more usefull than re-reading the same old replies for some.

Maybe the same old replies are right! Nothing to do with tangent of opinion, depth of field is a fact of physics.

 

[Suvv]

Then remember that when a lens is focused on an object it is in fact adjusting the light rays from that subject to hit the sensor as near parallel light rays as possible to render the subject in sharp focus.

I have to agree with Hoppy on this.

For example the statement above is completely wrong. If you have the lens aperture open wide then the light rays hitting the sensor from an in focus point will be coming from all angles and will be focused at a point on the sensor. They are not parallel.

 

[mrtoad]

mrcrow
The size of these small circles are dependent on the aperture size as the closed down aperture blocks the light rays which would otherwise be coming from the edges of the lens and only allows through the rays which are near the centre of the lens. is this really correct


Yes it is right. The rays from a single point on the subject, say a pore on a face, spread out like a solid cone and hit every part of the front of the lens. The lens then bends the light by different amounts depending on where on the lens the light hits, all of the light then focuses at a single point. If the camera had been focused on the face then the sensor will be where all of the light is focused. As you close down the aperture then light rays from the edges of the lens will be blocked so not so much light from the pore reaches the sensor so it will be darker.
Any points that are nearer than the pore will be focused nearer than the sensor and the rays will continue and spread out and reach the sensor as a circle (Circle of Confusion).
Any points that are further than the pore will be focused further than the sensor and the rays will reach the sensor as a circle (Circle of Confusion).
That circle will be made smaller if the aperture is closed down as the aperture will not allow light from the edges of the sensor through.
I am not sure if my diagram helps or not - it is a bit crude. S is a point on the subject, the orange line is the aperture, you have to imagine where the sensor is.
If the sensor is where the focused rays meet then the point will be in focus and aperture will just block some rays and make the point darker.
If the sensor is behind where the focused rays meet then you can see that the rays have spread out again and the point will not be in focus - it will be a circle, if you now imagine some of the rays being stopped by the aperture the spread out circle will be smaller. The smaller the aperture the more rays that are blocked and the smaller the circle - the more the point will seem in focus to the eye.
A similar argument applies if the sensor is in front of where the focused rays meet.
I know that it is not the sensor that moves but too much detail on the diagram makes it even more confusing. Just imagine each case - in focus, nearer than the focus point and further than the focus point seperately.

The rays from a single point on the subject, say a pore on a face, spread out like a solid cone and hit every part of the front of the lens. The lens then bends the light by different amounts depending on where on the lens the light hits, all of the light then focuses at a single point. If the camera had been focused on the face then the sensor will be where all of the light is focused. As you close down the aperture then light rays from the edges of the lens will be blocked so not so much light from the pore reaches the sensor so it will be darker.

thanks i can understand this and your sketch

does this lead to the conclusion that with really small apertures light from the edges (pore) gets so dark as to be outwith the exposure set..so tending towards vignetting ...

its quite a phenomenon and i havent noticed it so far...my present cam only goes down to f8 but i have used macro lenses at f32..

at this stage i cant really add anything just merely consider and accept
as i hope the OP can as well

i never thought of dof as affecting exposure so long as the shutter was suitably adjusted...and the exposure values were the same right across the circle projected into the body but curtailed by the sensors position which i assumed slightly inside the image circle

wish i had f16

 

[Suvv]

I may have confused you a bit - sorry

The rays from a single point on the subject, say a pore on a face, spread out like a solid cone and hit every part of the front of the lens. The lens then bends the light by different amounts depending on where on the lens the light hits, all of the light then focuses at a single point. If the camera had been focused on the face then the sensor will be where all of the light is focused. As you close down the aperture then light rays from the edges of the lens will be blocked so not so much light from the pore reaches the sensor so it will be darker.

thanks i can understand this and your sketch

does this lead to the conclusion that with really small apertures light from the edges (pore) gets so dark as to be outwith the exposure set..so tending towards vignetting ...

All light reflected towards the camera from all parts of the subject hits all parts of the lens, the lens then focuses all that light onto the sensor at the correct position to re-establish the scene. The aperture blocks light from the edges of the lens - this includes light from all parts of the subject. This does make the sensor image darker but you change the shutter speed to compensate.


its quite a phenomenon and i havent noticed it so far...my present cam only goes down to f8 but i have used macro lenses at f32..

at this stage i cant really add anything just merely consider and accept
as i hope the OP can as well

i never thought of dof as affecting exposure so long as the shutter was suitably adjusted...and the exposure values were the same right across the circle projected into the body but curtailed by the sensors position which i assumed slightly inside the image circle

As my comment above - decreased aperture reduces light across all of the image

wish i had f16

 

[mrtoad]

I may have confused you a bit - sorry

As my comment above - decreased aperture reduces light across all of the image

right!!

so one exposes longer...:|..etc.

depth of field is something i dont think about too much..its either f8 for maximum hopefully bringing in my foreground item which i focus a bit beyond..and let the background take care of itself...or f2.8'ish to blow out the background and intensify the main image..

but i am inclined to use P and let the whole shot fall into place.

 

[mrtoad]

I may have confused you a bit - sorry

As my comment above - decreased aperture reduces light across all of the image

right!!

so one exposes longer...:|..etc.

depth of field is something i dont think about too much..its either f8 for maximum hopefully bringing in my foreground item which i focus a bit beyond..and let the background take care of itself...or f2.8'ish to blow out the background and intensify the main image..

but i am inclined to use P and let the whole shot fall into place.

 

[Forbiddenbiker]

Well I was confused by your explanation. Still am TBH.

I have to agree with Hoppy on this. ...

Okay guys fair enough.

I'l work on my explanation angle for another day. using the word, narrowing, instead of near parallel I think.

----

... For example the statement above is completely wrong. If you have the lens aperture open wide then the light rays hitting the sensor from an in focus point will be coming from all angles and will be focused at a point on the sensor. They are not parallel.

Yes, its badly written I see that now.

....Its the angles of the light within the focus point (at the focal plane) on the sensor that determine the sharp DOF. The narrower the cluster's (CoC's) the larger the DOF.

I think that makes better sense.

err maybe.