Depth of Field, Circles of Confusion and Sensor Size.

[pxl8]

There is a lot of confusion surrounding the effect of sensor size on DoF, some articles say crop bodies have more DoF, others say they have less.

The truth is that sensor size doesn't change the DoF at all.

DoF is a function of magnification and aperture. If the subject is resolved to the same size in the frame (ignoring cropping) and you use the same aperture then it doesn't matter what the focal length or sensor size is, the DoF will be the same. It's as simple as that.... almost.

There are a few scenarios we need to examine to understand what is happening.

Firstly, same shot, same lens, different sensor size.
Let's say our “subject” is a ruler, a few feet away. Regardless of crop factor 10cm on the ruler will be the same size on the sensor. For this example we'll assume 10cm on the rules is 1mm on the sensor.

If we printed both shots 30cm high then the FF image needs less enlargement than the crop sensor image. This is obvious when we look at the print, the ruler and our 10cm sample will be smaller on the FF print (remember the 10cm sample is 1mm on both sensors). In order to get our 10cm sample the same size on both prints the one from the FF sensor would need to be enlarged by a factor of around 1.6.

Next, same shot, different lenses.
This time the focal length for the FF is 1.6x longer than the crop body. The 10cm sample from the ruler would now be 1.6mm on the FF sensor but the prints would be identical in terms of scale. The real difference here is that we've increased the magnification on the FF shot.

Finally, same lens, different shooting distance.
This is actually the same as the last example but instead of changing lens we've moved the FF sensor 1.6x closer to the subject. The 10cm sample is still 1.6mm on the FF sensor and the prints are again identical.

Circles of Confusion
The CoC is the size of a “point” on a print that, when viewed under normal conditions appears to be in sharp focus. The CoC changes depending on the size of the sensor. The smaller the sensor the greater the enlargement of the image for a given print size. A 10” print is 254mm, a FF sensor is 36mm and a crop sensor might be 22.5mm.

254mm / 36mm = 7.05
254mm / 22.5mm = 11.28

Clearly a crop sensor needs more enlargement than a FF sensor for a given print size. Going back to the first example. The 10cm sample on the print was 1mm on the sensor so it would be 7.05mm on the FF print and 11.28mm on the crop sensor print. If we printed the FF image 1.6x larger then then 10cm sample would also be 11.28mm on the print (7.05 x 1.6 = 11.28).

How does this affect DoF?
How sharp or blurred detail on a print will be is defined by the CoC, the more you enlarge the more blurred the CoC will become. If we print so the 10cm sample is the same physical size then the DoF will also be the same, as we saw earlier to get the 10cm sample printed the same size as the crop we needed to enlarge it another 1.6x in effect making the CoC the same size. In reality the DoF (optically) hasn't changed but the size of the printed details changes our perception of what is or isn't in focus.

Back in the real world...
In reality we print to a particular size no matter what the sensor size might be. This means that a 10x8” print a from crop sensor will have its CoC enlarged more than a print from a FF sensor which will give less apparent DoF than a shot from a FF camera.

When you start trying to match the framing (subject size on the sensor or print) between crop and FF then the DoF will be same. If you crop the FF image to match you also reduce the CoC size. If you increase magnification on the FF (either by moving closer or using a longer lens) then you decrease the DoF but that will be offset by the CoC size when printing. The FF image no longer needs cropping so for a given print size the FF CoC will be enlarged 1.6x less than the print from the crop sensor.

Hopefully I haven't made this too confusing, if it helps you can just summarise it as:

• Subject size the same on the print, DoF is the same regardless of sensor size.
• Print to a particular size (subject bigger on crop) then crop sensors have less (apparent) DoF.

 

[StewartR]

There is a lot of confusion surrounding the effect of sensor size on DoF ....

Indeed there is.

I've spent the last hour debating this with pxl8 via PM. I think he's got something subtly wrong, but I can't persuade him and he can't persuade me.

This is an extremely useful essay ... if we're 100% sure it's right. Let's not all barge in with opinions until we're quite sure of our ground.

I'm going to go away and wrap a cold towel round my head for a couple of hours. I'll be very happy to apologise to Andy and retract my words if, when I've finally got my head round it, I decide he's right after all.

 

[pxl8]

To help demonstrate I created this simple image:

On the left is the crop sensor, FF on the right. They are in proportion and then both images have been scaled to the same size "print" in the large version. The clearly shows the effect CoC has on prints.

So to clarify. When you you shoot FF so the subject size in the viewfinder matches that of a crop camera, say using an 80mm lens instead of 50mm to start with you lose DoF due to the increase magnification from the lens. But when you print the CoC comes into play and you regain the loss because the FF shot when printed isn't enlarged as much.

There'll now be a brief interlude whilst everyone nips off for some asprin

Pixel density and count do come into play but you'd really have to push the limits to notice because even on a low pixel count sensor the pixels are still much smaller than the CoC so it's that is the deciding factor.

Quiet, darkened room anyone?

 

[StewartR]

OK, the cold towel is off. (It only took one hour, not two.)

Let's take this slowly.

First, the setup. We have two cameras:
* one with a full-frame sensor (say a 1Ds) and an 80mm lens
* one with a 1.6x crop sensor (say a 40D) and a 50mm lens
Both cameras are the same distance from the subject and both use the same aperture. So the two images will be identically composed in the viewfinders, and then we will print them at the same size with no cropping and see what happens. OK so far?

First, let's look at what happens in the camera. What happens is that point sources of light in the field of view will become tiny circular blurs on the sensor; the closer to the plane of focus, the smaller the blur. But how big will those circles be? Critically, they will not be the same size in the two cameras. The Wikipedia article on Circle Of Confusion shows (about half way down) how to calculate the sizes of the blur circles. It shows that the size of the blur circle is (broadly) proportional to the square of the focal length. In order to achieve the same composition, our 1Ds has a lens which has a focal length 1.6x as long as the lens on the 40D. So the diameters of the blur circles on the 1Ds sensor will be 1.6 x 1.6 = 2.56 times as large as those on the 40D sensor.

Then, we need to magnify the image to form the print. The 1Ds sensor has a linear dimension which is 1.6x as large as the 40D sensor, so we need to magnify the image from the 40D sensor 1.6x as much in order to achieve the same print size. (For example, a 12"x8" print requires the 1Ds image to be magnified 8.47x, and the 40D image to be magnified 13.56x.)

So the blur circles are 2.56x as big on the 1Ds sensor as they are on the 40D sensor, but the 40D image is being magnified 1.6x more than the 1Ds image. So the result is that the 1Ds blur circles are 1.6x as big as the 40D blur circles on the print.

Let's say that again. The 1Ds blur circles are 1.6x as big as the 40D blur circles on the print.

Bits of the image which are slightly-out-of-focus will produce blur circles which are discernible as circles on the 1Ds print but which are too small to be discernible as circles on the 40D print. In other words, they will appear sharp on the 40D print but not on the 1Ds print. In other words, we have less DOF in the print made from the 1Ds image.

 

[pxl8]

Hmmm, how to explain this.

The CoC you've described there is for the sensor but there's another CoC size for the print.

For the same of simplicity let's say the CoC for the print (pCOC) is 10, for the crop body sensor (cCoC) it's 1 and for the FF body (fCoC) it's 1.6.

So you're going from the sensor size CoC to the print size.

pCoC / cCoC = 10
pCoC / fCoC = 6.25

The CoC from the FF sensor doesn't need to be enlarged as much to reach the CoC for the target print size. This means blur circles won't be enlarged as much. This is the balancing side of the equation I am talking about. FF need 80mm for the shot which means less DoF (due to magnification) but less CoC enlargement for the print so you get the sharpness back again.

Perhaps a much simpler way to look at this is with film. Under the enlarger you print the full frame to 10" wide. If you then crop to match the 1.6x body you've "enlarged" the CoC from the neg that much more - the blur circles are that much bigger.

This is why the CoC on a crop body is smaller than that of FF, 1.6x smaller (for a 1.6x crop). The more you enlarge the CoC the more blurry the image becomes. It's also important to remember that pixel resolution doesn't define sharpness here, the circle is an "optical" construct if you like which is why the film example helps visualise the effect. More pixels on the sensor simply means more discreet steps for the circle of blur and CoC for the sensor doesn't change with megapixel size.

 

[AliB]

Pass the wet towel will you, I'm going for a lie down!

 

[StewartR]

The CoC from the FF sensor doesn't need to be enlarged as much to reach the CoC for the target print size. This means blur circles won't be enlarged as much. This is the balancing side of the equation I am talking about.

But the whole point (and the reason you're wrong) is that it doesn't balance.

The magnification required is inversely proportional to the sensor size. We all agree there. The print from the FF sensor needs less magnification (1.6x less) than the print from the crop sensor.

But the sizes of the blur circles which are formed on the sensor are proportional to the square of the focal length. The blur circles on the FF sensor are 1.6 x 1.6 = 2.56 times as big as the blur circles on the crop sensor. This is the critical point which you've missed. (And no shame there, I should add. I didn't appreciate it until I got the wet towel out and did some research.)

So the blur circles on the FF sensor start off bigger. We don't have to magnify them as much to make the print, but they still end up bigger. (1.6x bigger, in fact.) And bigger blur circles is the phenomenon which we perceive as smaller DOF.

 

[jgs001]

I'm almost following this I think... And that also explains why, with a P&S or Bridge camera, like my Z2 which has a very small sensor, and a lens which goes from 6.3 to 63mm (35mm eq 38-380) has oodles of DOF even if you get the working distance down really pretty short.

Although I think I need to borrow the towel too please.

 

[pxl8]

A bigger CoC doesn't mean more blur, quite the opposite. From the wiki article:

Circles with a diameter less than the circle of confusion will appear to be in focus.

The smaller CoC on the crop body means you can't enlarge as much as you can from the FF body before the image looks OOF.

FF CoC is 0.018mm
APS-C CoC is 0.029mm

0.029 / 0.018 = 1.6

80mm / 50mm = 1.6

Balance is restored

I'm not sure which bit of the wiki article you got for the CoC calculation but the figures are widely published (wiki includes them) for various formats and those are the ones I've been using all along.

 

[pxl8]

I should add a disclaimer to all this. It's all give or take, roughly approximated. Lenses are only ever their named focal length when focused at infinity and even then numbers are rounded up/down for niceness. So a 50mm lens might be 48mm at infinity and 42mm at min. focus.

As a result, any calculations you do based on lens values are really best guesses based on rounded/truncated values in the first place but results will be slightly better than a random number generator

 

[StewartR]

Oh dear. This is just getting to be a replay of the discussion we had via PM ...

Please Andy, stop and reflect for a minute.

Everybody knows that P&S camera give you huge depth of field. Indeed, the ability to produce an image with shallow DOF is often cited as one of the big reasons to switch from P&S to SLR.

But you've just denied this. Are you really right, and everbody else wrong?

No. Here's why.

The Circle Of Confusion is more properly referred to as the Maximum Acceptable Circle Of Confusion. It is the largest circle on the sensor which cannot be perceived as a circle on the final print. Anything smaller than the CoC will be perceived as sharp and therefore in focus. Anything larger than the CoC will be perceived as a circle and therefore not in focus. I think we agree about this.

But it's important to understand why the CoC isn't the same size for each sensor. It's precisely because different sensors need different amounts of magnification to get to the same final print size. The way it's calculated is by starting with the size of the smallest detail which is resolvable by the human eye (typically taken to be 0.2mm), then making some assumptions about print size and viewing distance. The results are typically about 0.029mm for a FF sensor and about 0.018mm for a 1.6x crop, but that is precisely because the image from the crop sensor needs 1.6x more magnification. No other reason. It is so ingrained that it is almost part of the definition, not an emergent property.

But it's not all about the size of the CoC. That's just a limiting value. What we also need to consider - and this is the key point I've been trying to get across - is the size of the blur circle on the sensor. If the blur circle is smaller than the CoC, the point will appear to be sharp and in focus. If the blur circle is larger than the CoC, the point will appear as a circle and not in focus.

And, I repeat, the size of the blur circle is much bigger for the full-frame sensor. It's not 1.6x bigger; it's 2.56x bigger, because the size of the blur circle is proportional to the square of the focal length, and therefore proportional to the square of the sensor size.

Suppose we look at a point in the image which produces a blur circle 0.015mm across on the crop sensor. That's less than the CoC of 0.019mm, so it will appear as an in-focus point on the print. But when we shoot the same image with the FF sensor, we've had to increase the focal length from 50mm to 80mm in order to preserve the composition, and the blur circle will measure 0.15mm x 1.6 x 1.6 = 0.038mm. This is larger than the FF CoC of 0.29mm, so the point will appear as an out-of-focus circle on the print.

I'm exhausted now. I really don't think I can explain it any more clearly. If you still disagree, that is your right. But it doesn't mean you are right.

 

[pxl8]

My one objection is that I did not say P&S cameras have less DoF than crop/ff.:nono:

The good news is I'm found the point where my maths went astray, thanks to Stewart for sticking with it and keeping me thinking on this one

My mistake was to only think of the enlargement of the CoC, thinking that the smaller CoC of the crop sensor meant it had less scope for enlargement. What I failed to think about how the CoC values came out at print size to start with.

We have our two sensors:

FF has CoC of 0.03mm.
Crop with CoC of 0.019.

In both cases they are around 1/1200 of the width of the sensor. Now enlarge the width of the sensor a print size of 254mm (10")

254/36 means an enlargement of 7.05, a new CoC size of .21mm
254/22.5 means an enlargement of 11.28, a new CoC size of .21mm

What this shows, and what I missed until now, is that the CoC on the print is the same to start with. My apologies for leading everyone up the garden path on this one - I did say it was confusing

Now, could someone just, oh there it is, thanks

 

[malla1962]

lets keep it simple.
http://www.luminous-landscape.com/tutorials/understanding-series/dslr-mag.shtml

 

[purpleclouds]

You confused me alright!

 

[StewartR]

lets keep it simple.
http://www.luminous-landscape.com/tutorials/understanding-series/dslr-mag.shtml

Not really.

The only thing that article has to say about this topic is this:

Depth of Field is an illusion based on minimum sized Circles of Confusion and is related to the enlargement factor of the final print and the distance of the subject in multiples of the focal length of the lens being used. In other words, short focal length lenses have greatest depth of field at a given subject distance.

The DSLR has greater depth of field when the subject is framed the same as on a full frame SLR as the angle of view of the lens is less and the subject distance must therefore be greater.

But I have two problems with that:

(1) It's only referring to the case where you have the same lens on different cameras, and it's the subject distance which changes.

(2) It states that the DSLR has greater depth of field - which is true - but it doesn't say why.

 

[Le_Petit_Lapin]

Does that help? I started drawing it to make a point, but then forgot what the point was.

Left Cropped, Right Full Frame, Red Ruler.

 

[CT]

Quote:
The DSLR has greater depth of field when the subject is framed the same as on a full frame SLR as the angle of view of the lens is less and the subject distance must therefore be greater.

Not really.

The only thing that article has to say about this topic is this:But I have two problems with that:


(2) It states that the DSLR has greater depth of field - which is true - but it doesn't say why.

It could have been put better, but presumably he means a crop sensor SLR will be further from the subject than a full frame sensor to achieve the same FOV, and that as DOF increases with greater camera to subject distance, the full frame sensor will have less DOF?

 

[malla1962]

Does that help? I started drawing it to make a point, but then forgot what the point was.

Left Cropped, Right Full Frame, Red Ruler.

Not sure what you are trying to say here? I think we know how different sensors record a image size its DOF
we are trying to get our heads round, the smaller the sensor the greater DOF but why?

 

[malla1962]

Quote:
The DSLR has greater depth of field when the subject is framed the same as on a full frame SLR as the angle of view of the lens is less and the subject distance must therefore be greater.





It could have been put better, but presumably he means a crop sensor SLR will be further from the subject than a full frame sensor to achieve the same FOV, and that as DOF increases with camera to subject distance, the full frame sensor will have less DOF?

Im sure your right CT not sure how old the article is and if FF digital was about at the time, his dslr a d60 is 1.6 crop.

 

[CT]

Im sure your right CT not sure how old the article is and if FF digital was about at the time, his dslr a d60 is 1.6 crop.

Good point- he was obviously comparing a crop DSLR with 35mm film.

 

[fletch5]

wheres the over the head smilie???? and you left out the faithful point and shoot

 

[Canon Bob]

Quote:
The DSLR has greater depth of field when the subject is framed the same as on a full frame SLR as the angle of view of the lens is less and the subject distance must therefore be greater.





It could have been put better, but presumably he means a crop sensor SLR will be further from the subject than a full frame sensor to achieve the same FOV, and that as DOF increases with greater camera to subject distance, the full frame sensor will have less DOF?

I've been sitting back here in the auditorium as a casual observer and see that there's one subtle little thing that's been missed...or at least I don't recognise it from any of the preceding discussion...and I think it may be a different interpretation of the quote above.

Aperture affects DOF....larger aperture, smaller DOF. The danger is that we always think of aperture as f/x and if x is the same then the aperture is the same. It isn't.
If you frame the shot the same with a 1.6 crop and a FF body then, for example using Stewart's focal lengths above, you'd have 50mm on the crop and 85mm on the FF body. Having both lenses at the same aperture does not mean both at f/4. Set the 85mm to f/4 and you'd have to set the 50mm to f/2.5 to achieve the same aperture. It is the physical size of the aperture that determines its part of the DOF equation.
Hence, for the same framing, the crop sensor will have a larger DOF at the same f/x setting but it will have equal DOF when the aperture is set to the same physical size.

Bob

If you want me to leave now....just say the word

 

[Le_Petit_Lapin]

Not sure what you are trying to say here? I think we know how different sensors record a image size its DOF
we are trying to get our heads round, the smaller the sensor the greater DOF but why?

I don't know, honestly got 1/2 way through my little doodle and completely forgot.

 

[CT]

I've been sitting back here in the auditorium as a casual observer and see that there's one subtle little thing that's been missed...or at least I don't recognise it from any of the preceding discussion...and I think it may be a different interpretation of the quote above.

Aperture affects DOF....larger aperture, smaller DOF. The danger is that we always think of aperture as f/x and if x is the same then the aperture is the same. It isn't.
If you frame the shot the same with a 1.6 crop and a FF body then, for example using Stewart's focal lengths above, you'd have 50mm on the crop and 85mm on the FF body. Having both lenses at the same aperture does not mean both at f/4. Set the 85mm to f/4 and you'd have to set the 50mm to f/2.5 to achieve the same aperture. It is the physical size of the aperture that determines its part of the DOF equation.
Hence, for the same framing, the crop sensor will have a larger DOF at the same f/x setting but it will have equal DOF when the aperture is set to the same physical size.

Bob

If you want me to leave now....just say the word

That's actually a very good point Bob and one which has been well missed I'd say. I think it's a bit of stretch as an interpretation of the quote though when he's clearly talking about subject distance and angle of view.

Why would I want you leave - and what's the word?

 

[Canon Bob]

That's actually a very good point Bob and one which has been well missed I'd say. I think it's a bit of stretch as an interpretation of the quote though when he's clearly talking about subject distance and angle of view.

Why would I want you leave - and what's the word?

Trying to explain the concept and workings of DOF frequently turns a knowledgeable and literate genius into someone akin to a bumbling rambling idiot.....I've read so many interpretations now that all the possibilities have been used up.

Bob

Edit...In my example, I used 50 and 85mm (standard lenses) but 80mm would have made the math's more precise...you get the drift anyway.

 

[StewartR]

I've been sitting back here in the auditorium as a casual observer and see that there's one subtle little thing that's been missed...or at least I don't recognise it from any of the preceding discussion...and I think it may be a different interpretation of the quote above.

Aperture affects DOF....larger aperture, smaller DOF. The danger is that we always think of aperture as f/x and if x is the same then the aperture is the same. It isn't.
If you frame the shot the same with a 1.6 crop and a FF body then, for example using Stewart's focal lengths above, you'd have 50mm on the crop and 85mm on the FF body. Having both lenses at the same aperture does not mean both at f/4. Set the 85mm to f/4 and you'd have to set the 50mm to f/2.5 to achieve the same aperture. It is the physical size of the aperture that determines its part of the DOF equation.
Hence, for the same framing, the crop sensor will have a larger DOF at the same f/x setting but it will have equal DOF when the aperture is set to the same physical size.

Not missed or overlooked, Bob. Just not mentioned because (a) it wasn't immediately relevant to the discussion, which was all about comparing the effects at constant f-numbers; and (b) it's not the most helpful approach because photographers tend to think of apertures as f-numbers rather than absolute sizes.

In fact the most straightforward approach to determining the size of a blur circle or a CoC, which is set out very nicely and clearly in the Wikipedia "Circle of confusion" article, naturally comes up with a formula whereby the size of the circle is proportional to the focal length times the aperture. It's only by introducing the f-number (ie aperture = focal length / f-number) that we get to the result that the size of the circle at any given f-number is proportional to the square of the focal length.

And it's this formulation which is the key to appreciating the proof I outlined above:

The magnification required is inversely proportional to the sensor size. We all agree there. The print from the FF sensor needs less magnification (1.6x less) than the print from the crop sensor.

But the sizes of the blur circles which are formed on the sensor are proportional to the square of the focal length. The blur circles on the FF sensor are 1.6 x 1.6 = 2.56 times as big as the blur circles on the crop sensor. This is the critical point which you've missed. (And no shame there, I should add. I didn't appreciate it until I got the wet towel out and did some research.)

So the blur circles on the FF sensor start off bigger. We don't have to magnify them as much to make the print, but they still end up bigger. (1.6x bigger, in fact.) And bigger blur circles is the phenomenon which we perceive as smaller DOF.

 

[Canon Bob]

No problem Stewart. I've read many explanations of DOF and inevitably assumptions are made about the factual accuracy....frequently dangerous. It's also common that the final enlargement (at print level) is said to negate the benefit....clearly disproved by your maths.

Thanks for laying it out this way.

Bob

 

[StewartR]

So where do we go with this now, Andy/pxl8? As I said earlier, I think having an accessible reference available would be of benefit to TP members. Do you want to edit your essay to correct it? (Actually, it would probably be better to post a revised version at the end of the thread, to avoid later confusion.) I'd be happy to proof-read if you like.

 

[pxl8]

A good question, I did think about an edit but thought the thread also served another purpose in highlighting the problems/issues/my stupidity

I'll do an update and email you a copy for checking, you're very kind to do so. Thanks again for showing me the light (groan)

 

[HoppyUK]

Can I add my 2'penneth? (I'd not seen this when I wrote my reply to another thread.)

Depth of field is affected by f/number and image size (focal length, subject distance). That's it, and this has been explained pretty well.

In my view, most of the confusion arises around image size and format size. It leads into all sorts of hugely complex maths which is irrelevant to photography in practise, and we haven't dwelt long on pixel desities yet!. If you forget comparing actual image sizes on various formats, and start comparing frame-fillers vs frame-fillers regardless of format (which is what matters) then smaller formats always give more depth of field. This is depsite the obvious fact that they need to be enlarged more to produce the same size print. That's just the way physics made it, hence the big depth of field for compacts, and increased depth of field for crop-format SLRs vs full-frame (factor 1.25 stops for Nikon/Canon, ie f/stop x crop factor).

The other thing that gets slighly difficult for the picky peeps is the assertion that wide-angles and telephotos give the same depth of field provided image size remians the same. This is true to an almost absolute extent, but if you look at the numbers depth of field does vary very slighly and the depth in front and behind the subject also varies very fractionally too, but to all intents and purposes it is true within a few percentage points.

A much more significant difference between wide angles and telephotos is the field of view (a function of viewpoint) which, while it doesn't alter depth of field it alters the amount of background included and this certainly gives a very strong impression of reducing depth of field. It's technically an illusion of course, but it surely works

Regards,

Richard.

 

[Garry Edwards]

I've been sitting back here in the auditorium as a casual observer and see that there's one subtle little thing that's been missed...or at least I don't recognise it from any of the preceding discussion...and I think it may be a different interpretation of the quote above.

Aperture affects DOF....larger aperture, smaller DOF. The danger is that we always think of aperture as f/x and if x is the same then the aperture is the same. It isn't.
If you frame the shot the same with a 1.6 crop and a FF body then, for example using Stewart's focal lengths above, you'd have 50mm on the crop and 85mm on the FF body. Having both lenses at the same aperture does not mean both at f/4. Set the 85mm to f/4 and you'd have to set the 50mm to f/2.5 to achieve the same aperture. It is the physical size of the aperture that determines its part of the DOF equation.
Hence, for the same framing, the crop sensor will have a larger DOF at the same f/x setting but it will have equal DOF when the aperture is set to the same physical size.

Bob

If you want me to leave now....just say the word

Yes, that's the right answer.
All things being equal, the same aperture will produce the same dof on any lens.
But (say) a 5mm aperture on a 10mm lens will be f/2, on a 50mm lens it will be f/5 and on a 85mm lens it will be f8.5
Forget the f/number, think about the actual physical size of the aperture.

 

[pxl8]

Of course it should also be remembered that it's only true if you use the entire frame, start cropping then the rules change.

It's also easier to work out the aperture difference by multiplying or dividing by the crop factor (depending which way you're going). 50mm @ f/4 on a 1.6 crop means 80mm @ f/6.4 on full frame.

 

[pxl8]

Dragging this thread up from the depths...

The recent thread about DoF and macro made me think about this in another context and I would be interested to hear thoughts and opinions on how DoF is affected in the following:

f/8 1:1 mag, full frame
f/8 1:1 mag, 1.6x crop

f/8 1:1 mag, full frame cropped to 1.6x
f/8 1:1 mag, 1.6x crop

In both example which shot has the greater DoF? Why?

 

[oldfella]

Just getting my big guns loaded up before I enter this affray.

 

[jerry12953]

Just getting my big guns loaded up before I enter this affray.

Oooer......

 

[oldfella]

Bang whoooosh kapow. all dead

 

[oldfella]

Dragging this thread up from the depths...

The recent thread about DoF and macro made me think about this in another context and I would be interested to hear thoughts and opinions on how DoF is affected in the following:

f/8 1:1 mag, full frame
f/8 1:1 mag, 1.6x crop

f/8 1:1 mag, full frame cropped to 1.6x
f/8 1:1 mag, 1.6x crop

In both example which shot has the greater DoF? Why?

Full macro at F8 oh my

 

[seriousrikk]

This topic always gives me a headache but I have never gone away for the wet towel approach and worked it out. Generally I solve the problem with a nice glass of red wine or a bottle of quality Ale.

However I would like to give Andy & Stewart a big thumbs up for discussing this openly and honestly - and I look forward to reading the revised version in full (And believe me - I will).

 

[pxl8]

Full macro at F8 oh my

Got anything sensible to add to the thread then? :razz:

 

[CT]

Full macro at F8 oh my

Erm..