Can anyone help me with hyperfocal settings ?

[quickben]

I'm trying to find the optimal setting for my Tamron 28-75 F2.8 Xr Di for hyperfocus shooting on my 10D.

Where do you set the focus point ?

What aperture is the best ?

I've seen the charts going around but I haven't got the faintest idea how to use them :oops:

Please help, I'm planning on doing some landscape shooting this weekend.

Gary.

PS: I'm in the process of changing my avatar :shock:

[cmM]

Hyperfocal distance differs with aperture and focal length.

You want to do landscape photography, right? You most likely will use the lens at the wide end (28mm) and stop down the aperture to f/11- f/16 to get enough DOF.

There are formulas to calculate hyperfocal distances and stuff like that, but if you go by the charts, you will see that there is a focal length and an aperture. where that row&column intersect, there is a distance. Focus manually on a point which is *that* distance away from your lens to maximize your depth of field.

Is all that really necessary? Depends. Some say yes, I say not always.

[Jon]

The best aperture is the smallest you can use, as this gives maximum depth of field. Hyperfocal distance varies with focal length, and degree of apparent enlargement (how much it's enlarged, and what the viewing distance is going to be). Most of the tables that are out there are for 35 mm. With the 10D's crop factor, your numbers will be different. At 28 mm, f/22 your hyperfocal distance is about 6'. At 75 mm, it's 43'. You can find an on-line calculator here (http://www.dofmaster.com/dofjs.htm).

Basically, once you've found the hyperfocal distance for your lens and aperture, focus the lens to that point and everything from half that distance to infinity will be in (commonly-held) acceptable limits of focus.

[Scottes]

It is generally accepted that a lens will perform it's best when stopped down 2 stops from it's widest. So for this lens, set it to f/8. This is not always the case of course, but seems to be the usual case. (And often enough, there's no discernible qulity betwen 2 and 3 stops, so f/11 might be OK for you, too.) So you set the Av mode and use that aperture of f/8.

Next is to check what zoom you need to take the picture. 28mm? 43mm? 64mm? Find the hyperfocal distance for the combo of zoom length & aperture. You'll find that as lens length increases you will have to go to a smaller aperture to get a decent hyperfocal. At 28mm f/8 hyperfocal is 17 feet. At 75mm f/8 hyperfocal is now 121 feet. But 75mm f/45 hyperfocal is 21 feet - about the same as 28mm f/8.

So fo 28mm f/8 you would aim at something 17 feet away, and expect everything to be in focus from about 9 feet away to infinity. Set the lens to manual focus, set the distance to 17 feet using the distance ring of the lens, aim and shoot.

[quickben]

Thanks for the advice !!

I think the only problem I'll have is estimating where *that* distance is.

I can only try and learn, I suppose.

Thanks again.

Gary

[cmM]

I think the only problem I'll have is estimating where *that* distance is.

Does your lens have a distance scale?
If not, take a long tape measure :wink:

[Scottes]

Thanks for the advice !!

I think the only problem I'll have is estimating where *that* distance is.


Don't estimate. Use a chart - here (http://www.johnhendry.com/gadget/dof.php) or here (http://www.dofmaster.com/dofjs.html)

Use the distance ring on the lens to get the distance correct, if at all possible.

[quickben]

Scottes wrote : "So fo 28mm f/8 you would aim at something 17 feet away, and expect everything to be in focus from about 9 feet away to infinity. Set the lens to manual focus, set the distance to 17 feet using the distance ring of the lens, aim and shoot."

The only problem I have with that, is that my lens' distance ring only goes from 1.08 feet to 7 feet then infinity.

I can't see how I can dial in 17 feet. What would be the focusing distance if I used f11 or f13 ?

Thanks everyone.

Gary.

[Scottes]

The only problem I have with that, is that my lens' distance ring only goes from 1.08 feet to 7 feet then infinity.

I can't see how I can dial in 17 feet. What would be the focusing distance if I used f11 or f13 ?

That kind of distance scale makes it pretty difficult to dial in 17 feet. Yep. So a tape measure and/or getting used to eyeballing the distance is necessary. You *could* take a tape measure at home and measure off some distances and put tiny scratches to make your own distance guage.

As to "what would be the focusing distance" well there are three links above that list DoF calculators with Hyperfocal Settings. Just plug in some numbers into the calculators. And then print them down in a little chart and carry it with you.

[Rayz]

The shorter the focal length of your lens, the less critical you have to be about hitting the exact hyperfocal spot. I think on some of my lenses there's nothing between 2 metres and infinity. With such a lens I don't bother doing any calculations. I just focus on the main subject of interest in the composition and use f11. I've never found in any my lenses I've tested, more than the most marginal fall-off in resolution at f11.

Another point you should bear in mind, wide angle lenses change the perspective of the composition. That is, objects in the foreground appear immensely large in comparison with objects in the background. With telephoto lenses you get the reverse effect.

Clearly, if objects in the background appear tiny on the print, they don't need to have great resolution to appear sharp. If the objects in the background are large, as is the case with a telephoto lens, they do need to have great resolution to appear sharp.

This phenomenon is a great source of confusion on the Luminous Lanscape where Michael Reichmann has published a controversial article (tutorial) on Focal Length & DoF, in which he claims that all lenses show the same DoF provided the subject of interest is the same size on the sensor or film. http://www.luminous-landscape.com/tutorials/dof2.shtml

To demonstrate this point, Reichmann takes a number of shots of a stuffed doll with different lenses ranging from 400mm to 17mm. He varies the distance with each shot so the doll appears the same size in the image in each case. Clearly, the 400mm shot appears to have a shallow DoF and the 17mm shot appears to have great DoF as one would expect. Not so, says Reichmann, and proceeds to enlarge a tiny tower on the horizon in the 17mm shot to demonstrate it really has no more detail than the extremely blurry and huge tower in the 400mm shot.

Really! Michael. Pull the other leg :D . Is this what DoF means?

[Jon]

This phenomenon is a great source of confusion on the Luminous Lanscape where Michael Reichmann has published a controversial article (tutorial) on Focal Length & DoF, in which he claims that all lenses show the same DoF provided the subject of interest is the same size on the sensor or film. http://www.luminous-landscape.com/tutorials/dof2.shtml

To demonstrate this point, Reichmann takes a number of shots of a stuffed doll with different lenses ranging from 400mm to 17mm. He varies the distance with each shot so the doll appears the same size in the image in each case. Clearly, the 400mm shot appears to have a shallow DoF and the 17mm shot appears to have great DoF as one would expect. Not so, says Reichmann, and proceeds to enlarge a tiny tower on the horizon in the 17mm shot to demonstrate it really has no more detail than the extremely blurry and huge tower in the 400mm shot.

Really! Michael. Pull the other leg :D . Is this what DoF means?

He's 100% correct. If you take a picture with any two rectilinear lenses from the same distance, crop the same area out of both of them, and enlarge them to the same final print size, the DoF will be the same in both. This is because the degree of enlargement changes the size of the circle of confusion (which determines the DoF).

[Rayz]

He's 100% correct. If you take a picture with any two rectilinear lenses from the same distance, crop the same area out of both of them, and enlarge them to the same final print size, the DoF will be the same in both. This is because the degree of enlargement changes the size of the circle of confusion (which determines the DoF).

Jon,
Looks as though you've also been infected with the confusion :D . Also you don't appear to have read the Reichmann article I referred to. In this tutorial, Reichmann is taking shots from different distances with different lenses but at the same f stop of f5.6. The point he's trying to make is that the well known fact that telephoto lenses exhibit less Depth of Field than wider angle lenses is false, provided the subject is the same size in the frame.

Where on earth he got this absurd idea from beats me. The Luminous Landscape is a great site. There's a wealth of useful information there and some examples of great photography from Michael, but sometimes Michael is just plain wrong and unable to see it. Okay! So none of us is perfect, but Depth of Field and how it's affected by focal length is such a basic issue in photography, it's difficult to understand how an experienced photographer like Michael could be so confused on this issue.

I hope the moderators do not consider these remarks inflammatory or a personal attack on Michael. I'm just calling a spade a spade. If it doesn't look like a duck, doesn't walk like a duck and doesn't squawk like a duck, I think one can be fairly certain it isn't a duck.

Ergo, if the shots don't look as though they've got the same DoF, if they ignore accepted definitions of the term DoF and don't comply with the basic mathematical formulas that describe DoF, then I think one can be fairly certain the shots have not got the same DoF.[/i]

[Jon]

He's 100% correct. If you take a picture with any two rectilinear lenses from the same distance, crop the same area out of both of them, and enlarge them to the same final print size, the DoF will be the same in both. This is because the degree of enlargement changes the size of the circle of confusion (which determines the DoF).

Jon,
Looks as though you've also been infected with the confusion :D . Also you don't appear to have read the Reichmann article I referred to. In this tutorial, Reichmann is taking shots from different distances with different lenses but at the same f stop of f5.6. The point he's trying to make is that the well known fact that telephoto lenses exhibit less Depth of Field than wider angle lenses is false, provided the subject is the same size in the frame.



Ah, but the tower's essentially at infinity WRT the camera position in all cases. As he stated, it's on the order of a mile from the camera position.

[Mark Kemp]

http://www.dofmaster.com/dofjs.html

http://www.outsight.com/hyperfocal.html

Are both quite helpful

[Rayz]

Ah, but the tower's essentially at infinity WRT the camera position in all cases. As he stated, it's on the order of a mile from the camera position.

I've just re-read Reichmann's first article on DoF where he confesses that he doesn't fully understand the issue himself and had to delve into his text books for help. He provides the following definition:

Definition: "The area in front of and behind a focused subject in which the photographed image appears sharp".

This definition is similar to most definitions of DoF which usually mention 'areas of acceptable' sharpness, or out-of-focus areas which appear as sharp as areas that are in focus. The significant thing about DoF is that it's an 'appearance', it's 'relative' and it depends on the size of the print being viewed.

If 2 equal size prints viewed from the same distance 'appear' to have the same DoF, then they have the same DoF. The mathematical formulas tell you why and allow you to predict what will 'appear' sharp and under what conditions.

If you move to Michael's second article on DoF where he shows samples of the same subject taken from different distances with different lenses, it must be quite clear that the 17mm shot appears to have much greater DoF than the 400mm shot, and I suggest that the basic maths supports this view. Try the formula H=f squared over nC where C=0.03mm. I think you'll find that a 17mm lens at f5.6 has a hyperfocal distance of 1.7 metres, ie. with the lens focussed at 1.7m, everything will have 'acceptable' sharpness, or 'appear' sharp, from 1.7/2m (850mm) to infinity on an 8x12" print.

[Jon]

As I recall, you were challenging his assertion that, when he enlarged the image of the tower from the 17 mm shot until it was the same size as the one from the 400 mm shot, the two images were equally fuzzy, thus depth of field for that case was independent of the lens. Which is what his exercise demonstrates.
Circle of Confusion for an image is a consistent value determined by the total degree of magnification (degree of magnification in the "negative", in the "print", and in the viewing distance) in the finished print and the aperture used. It's not subjective; it's measurable and computable by formula. The size of the circle of confusion at which point blur becomes unacceptable is subjective, however. DoF tables rely on the most commonly accepted values of CoC.

[Rayz]

As I recall, you were challenging his assertion that, when he enlarged the image of the tower from the 17 mm shot until it was the same size as the one from the 400 mm shot, the two images were equally fuzzy, thus depth of field for that case was independent of the lens. Which is what his exercise demonstrates.


This is where the total confusion lies in Reichmann's concept. If you selectively enlarge a part of an image in order to compare that part to the same part in a different image that has been enlarged to a different degree, or if you compare vastly different sized prints of the same subject, then you can prove any silly theory you like.

DoF is about the relative sharpness of near and far objects in the same image. If you wish to compare the DoF in the same scene taken with two different cameras and/or lenses, then it's only sensible to compare same size prints.


Circle of Confusion for an image is a consistent value determined by the total degree of magnification (degree of magnification in the "negative", in the "print", and in the viewing distance) in the finished print and the aperture used. It's not subjective; it's measurable and computable by formula. The size of the circle of confusion at which point blur becomes unacceptable is subjective, however. DoF tables rely on the most commonly accepted values of CoC.

The Circle of Confusion is a consistently variable value which, on the film or sensor, is dependent only upon the focal length of the lens, the f stop used, and the distance between camera and subject (for any given format). The Circle of Confusion on the print is determined by the degree of enlargement.

The degree of enlargement can also affect the appearance of sharpness in different parts of the image relative to each other. For example, a standard 320mm lens on an 8x10 field camera used at f8 will probably produce reasonable DoF on a contact print, as viewed from the same distance as one normally views an 8x10 print. However, enlarge the 8x10 film about 4x to make a 32x40" print, and it will become apparent that the DoF is very shallow, which is why LF photographers tend to use large f stop numbers like F45 or F64.

If we consider in more detail the 17mm shot of the gremlin doll in Michael's article, Michael states that he had to get closer than 3ft to take that shot. If MR had focussed on the hyperfocal distance of 1.7m instead of the doll, this shot would have been as sharp as required for an 8x12' print, ie. the CoC would have been no greater than 0.03mm from the doll to infinity (on the film) and 0.03x8= 0.24 on the print, which is all you need.

However, it appears that Michael focussed on the subject in each shot. If we rework the formula for hyperfocal distance, then for an H value of say 850mm (less than 3ft), we can find a new value for C which will determine what degree of enlargement is allowed for acceptable sharpness.

According to my calculations, the CoC becomes 0.06mm on the film. Divide into 0.24 (print CoC) and you get roughly a postcard size enlargement (4x) in which everything is acceptably sharp from the doll to infinity. There's no way the tower in the 400mm shot could have acceptable sharpness in any print larger than the smallest postage stamp :D .

What these example images in Michael's article demonstrate is that different focal length lenses will produce different perspectives and that longer focal length lenses tend to produce less DoF than shorter focal length lenses.

If you think they produce equal DoF, then you've been conned :D .

[maderito]

I found this thread very disorienting when I first followed it - like listening to an argument between wife and husband - both alwys right - sort of.

Today I shot a mini-marathan road race with a Canon 200mm f/4.0L lens and felt somewhat frustrated by DOF problems - a point brought up in this and other forum threads recently.

So what happens to DOF when the image size is kept constant. IMO, it is explained best here: http://www.dofmaster.com/dof_imagesize.html

The bottom line is that that the nonlinear behavior of optics, and therefore, of DOF equations, is most dramatic with short focal length lenses and/or at close distances (from camera to subject). For other cases, the point that Reichmann is making is almost correct and good enough for day-to-day work when you've left your DOF chart and calculator at home: DOF is roughly similar among standard and telephoto lenses when image size is kept constant.

Reichmann did overstate his case to make a simple point - but doesn't he usually. :)

[Rayz]

I found this thread very disorienting when I first followed it - like listening to an argument between wife and husband - both alwys right - sort of.


Give me a wife like that! I want a wife that has all the usual physical attributes that attract men (you know what I mean) plus the ability to argue sensibly on matters such as DoF. Please! Pulease God! Give me such a woman!

Okay! To more serious matters :D . You're quite at liberty to assert that Michael Reichmann is right on this issue. All I want from you is the mathematical verification of this point of view. And the reason I want it is because those sample images in the second article 'Focal length and DoF' do not look, to my eyes, as though they have equal DoF.

At the beginning of the first article, "Understanding DoF", Reichmann makes the following statement:

"The term "Depth of Field" describes the range in a photograph, from near to far, that appears to be in focus."

Later in the article he makes this statement:

"At the beginning of this tutorial I wrote, "Most people also believe that wide angle lenses have more depth of field than telephoto lenses (false)." Why is this such a common misconception?
It seems logical. Wide angle lenses appear to have more depth of field than long lenses, don't they? Yes, they appear to but only when you don't take subject size into account."

Well, you can't have it both ways. If DoF is defined as a range that appears sharp, you can't then say, hey! this range appears sharp but it's not really.

DoF does is not defined by 'absolute' sharpness but 'relative' sharpness.

If it was defined by absolute sharpness, then pinhole cameras would have no DoF at all, because as we all know such photos are very blurry.

[maderito]

Well ... all if fair in love & war except misquotation. I didn't say MR was right - only that he was "almost correct" and that his point was "good enough for day-to-day work".

My dear mom use to work for the government, and so I grew fond of the phrase - "good enough for government work." MR's argument fits into this category - good enough, but certainly not correct or accurate.

The photographer who regularly shoots concerts asks the question: should I sit in the front row, 20 ft from the string section, and use my favorite 80 mm lens or 50 ft back and use my trusted 200 mm lens?

Assuming a Canon 10D, and desiring a constant image size of the 1st violinist's Stradavarius, and using DOF=near to far distance of acceptable sharpness:

200mm f/5.6 focused @ 50ft
DOF = 3.994 ft

80mm f/5.6 focused @ 20ft
DOF = 4.028 ft

Pretty close...

Examining the results, the photographer then carries his/her experiment to the extreme and confronts these facts while now perched on the stage to gain a different perspective with a new 10-20mm zoom lens:

20mm f/5.6 focused @ 5ft
DOF = 4.742 ft

10mm f/5.6 focused @ 2.5 ft
DOF= 10.96 ft
(and focused at 3.1 ft, DOF = 1.6 ft to infinity)

My assertion was simply that as long as you are not working with WA lenses or at short camera-to-subject distances, DOF is approximately the same for constant image sizes across lenses of different focal lengths.

As far as I can discern, your claim is that MR is wrong - and for that simple claim, you are right (join the crowd). :)

[Rayz]

My assertion was simply that as long as you are not working with WA lenses or at short camera-to-subject distances, DOF is approximately the same for constant image sizes across lenses of different focal lengths.



Actually there does appear to be a principle here, but my figures are different from yours (I've been using dofmaster). I get the impression that DoF can be the same for constant image size provided the camera to subject distances are short. Ie. Use a 400mm lens from 20ft, a 200mm lens from 10ft and so on, and you can go all the way to a 10mm lens from 0.5ft with little change in DoF.

The principle seems to work best if you stay well clear of all hyperfocal distances. Consider the following situations. For equal subject size a 400mm lens with subject distance of 400ft = 200mm at 200ft = 100mm at 100ft and so on. Following are the results for DoF behind the subject. I've used 35mm format, a CoC of 0.03mm and constant aperture of f5.6.

400mm...... 59.2ft

200mm...... 69.5ft

100mm......106.4ft

80mm........145ft

50mm....... infinite DoF

It's clear the DoF is getting progressively greater at these distances. By the time we get to 50mm at a subject distance of 50ft, we're at the hyperfocal distance.

Working with your examples of a 10D with 200mm lens trained on the Stradavarius, for DoF behind the violinist, I get:-

200mm .... 2.11ft
80mm ..... 2.25ft
50mm ..... 2.41
20mm ...... 3.39

Perhaps I should apologise to Michael. He seems to be at least half right :lol:

Ray

[maderito]

Perhaps I should apologise to Michael. He seems to be at least half right :lol:
Ray

I'm sure Reichmann would be happy to know he's half right and not half wrong.

I get the same numbers as you using your assumptions. My assumptions were a DSLR (1.6 crop) and DOF=near + far DOF.

The problem with generalizations about DOF is that you invariably get back to hyperfocal distance (where this whole thread began). NOBODY knows the hyperfocal distance of each lens they own at all working f-stops. And only landscaper photographers take time to pay attention to it - both the concept and the actual numbers - when shooting.

For DOF generalizations, you typically have to be concerned with optical behavior when (1) subject distance is near the lens focal length, (2) when subject distance is near the hyperfocal distance, and (3) when subject distance is far beyond the hyperfocal distance. These are approximate break points when the optical formulas change from one type of behavior (near linear or nonlinear) to another. Who can keep all this clear in their heads while shooting?

Given the type of shooting I do (mostly available light and in close with medium focal length fixed and zoom lenses), it's good enough for me to know that the DOF doesn't change much between lenses or degrees of zoom. Then I can try to remember how thin or thick the DOF is at various apertures. For my shots, the DOF is usually quite thin and probably accounts for more OOF shots than any other variable I have under my control.

A rock concert photographer once bemoaned the fact that his subjects kept swaying in and out of his DOF - ruining his shots!

We had a recent thread about a website which allowed you to visualize DOF. It's a must see for all: http://www.liquidsculpture.com/DOF/DOF.htm (courtesy of Scottes).

My goal in life is to always be at least half-right and never half-wrong. :)

[Scottes]

it's good enough for me to know that the DOF doesn't change much between lenses or degrees of zoom.

I just had an experience that counters this. So my "field observations" first, then some math...

I usually shoot butterflies with my 100-400 and an extension tube. (I want a 100-200 macro with IS, dammit!). Yesterday I shot with my 105mm Macro, and blew most of my shots due to DoF. I didn't even think about it, and was mostly using apertures in the f/8 - f/11 range. Well at 105mm and 18" working distance I had *way* too much DoF and many images became cluttered because of too-sharp or semi-sharp backgrounds. Such an image would have been well-blurred on my 100-400.

Why?

Well, I just checked John Hendry's DoF calculator, and it says that 105mm at 18" at f/8 on a 10D = 0.02" of DoF. Less than 1/2 a millimeter. 400mm at the same setting (not accounting for the extension tube) produces 0.00" of DoF.

The pictures don't look the same. While DoF is defnitely deeper with the 105mm, nothing is sharp - it's just not so fully blurred like the 100-400 would do. The pictures with the 100-400 look perfect because of nicely-blurred backgrounds, while the 105mm looks blurred but not blurred enough.

One point to realize is that Depth of Field per the charts is not a point in space like the charts make it seem. That is, if a chart says that Near DoF is 1' and Far DoF is 2', then something that is 2.01' away will not suddenly become blurred. It is simply blurred a little more than 2', but past the mathematical point of acceptability.

With the 400mm the fall-off (or "rate of blur") is much more pronounced at the same distances - the blur difference between 2.0' and 2.1' is noticable. With the 105mm the fall-off is less noticable, and there's little difference between 2.0' and 2.1'.


With the 105mm at f/8:
Distance - DoF
10' - 0.81'
12' - 1.18'
14' - 1.61'
16' - 2.11

With the 400mm at f/8:
10' - 0.05'
12' - 0.07'
14' - 0.10'
16' - 0.14

So DoF increases equally mathematically for both - DoF at 16' is almost 3 times the DoF at 10', regardless of lens length. But the amount of blur is dramatically different. A secondary subject at 18' will be blurred with the 105mm focused at 10' and barely acceptably sharp focused at 16'. But with the 400mm that secondary subject will be completely blurred whether the focus distance is 10' or 16' or even 17.5'.


Note that I am not discussing "the same image" now, but just distance. Also, the above calc do not take the tubes into effect, though I'd expect the DoF to be less that what's stated.

[maderito]

Note that I am not discussing "the same image" now, but just distance.

Right. A shifting, wandering thread here.

We had been talking about DOF given constant image size.

The effects you note show that a longer lens with focus set at the same distance as a shorter focal length lens will have a very different and shallower DOF. Moreover, this difference will be magnified as you move closer to the object. In fact, in your first example at 18", the actual calculated difference in DOF (using exact equations) is almost 100-fold between the 105mm and 400mm lens.

Macro photography is a specific example of where lenses of different focal length make a big differences in DOF because camera-to-subject differences are close to the lens focal length. The beahvior of the optical equations in this region (subject distance approx. equals lens focal length) are extremely nonlinear.

I don't do macro photography. However, for those who do, I'd guess that there are actual differences with aesthetic consequences on how image sharpness falls off from the plane of focus to points near and far depending on the lens focal length. That said, most posts I read talk about convenience of shooting (how close to the subject) and not aesthetics when discussing macro photography with lenses of different focal lengths.

[Scottes]

Right. A shifting, wandering thread here.

We had been talking about DOF given constant image size.

And so was I - at least when I started.

I have many BF pics showing the same approximate image size - same or similar BFs at same or similar sizing, taken with two completely different rigs. The math gets completely screwed up because of the tube on the 100-400. So I can't *mathematically* prove it, but I can see a huge difference in DoF between these two rigs, even with (approximately) the same image size and definitely the same aperture.

I'll have to dig through some images....

[Mark Kemp]

Can anyone help 'me' with hyperfocal settings?

Seems like everybody can !

Pity they don't all have the same advice !

My head hurts now - I will rely on guesswork and luck --- its not as accurate, but it doesn't involve complex theory and hard sums.

:D

[Scottes]

OK, cancel my last posts. I checked through images and found that many taken with the 100-400 were usually at 170-200mm, and I really couldn't tell *enough* difference in DoF.

Maybe I'll measurebate it some day...

[Rayz]

Can anyone help 'me' with hyperfocal settings?



Mark,
If you don't want to do the calculations, then you'll just have to rely on experience :D . Just bear in mind, the shorter the lens, the higher the f stop # and the greater the distance between camera and subject, then the greater the DoF. And of course, the reverse is true.

The basic formula for calculating hyperfocal distance is H=focal length squared over ( f stop# times CoC). The CoC varies depending on desired print enlargement and format size, but generally a CoC of 0.019mm is used for formats such as the 10D and 0.03mm for 35mm. If you are going to use this formula, which is really very simple, make sure all values are in the same units, ie. everything in millimetres for example. (Except of course the f stop number :D ).

Ray

[Rayz]

Given the type of shooting I do (mostly available light and in close with medium focal length fixed and zoom lenses), it's good enough for me to know that the DOF doesn't change much between lenses or degrees of zoom. Then I can try to remember how thin or thick the DOF is at various apertures. For my shots, the DOF is usually quite thin and probably accounts for more OOF shots than any other variable I have under my control.

My goal in life is to always be at least half-right and never half-wrong. :)

Woody,
There's another issue here that the maths doesn't address and that's the subjective nature of DoF. When viewing a print, one doesn't bring out the pocket calculator to calculate DoF. It doesn't really matter what the maths says. If it looks as though it's got shallow DoF then it has got shallow DoF and no amount of mathematical formulas can prove otherwise.

Let's look at what really happens when you use say a 100mm lens from 25ft instead of a 200mm lens from 50ft, at the same f stop.

The stuff in the background in both shots will have the same resolution, according to the DoF calculator. But the background detail in the 200mm shot will appear larger in relation to the main subject. The eye uses such cues in order to get a sense of depth of field. If an object behind something, that we recognise and have some idea of its size, appears quite large in relation to an object in front of it, then we assume the object behind is fairly close to the object in front. If it looks fairly close (even though it isn't) and is also rather blurred (same resolution as in the 100mm shot), then we get a sense of shallow DoF.

There's another factor which also influences the subjective sense of DoF. The longer lens has a narrower field of view and therefore excludes detail from the background which again the eye uses as a cue for depth of field. If the calculated DoF is say 10ft behind the subject, then the 100mm shot might include something that's sharp and 10ft behind but a bit to the right. That same background object might be excluded from the 200mm shot, which will contribute to the sense that the 200mm shot has a shallower DoF.

In conclusion, I have to say I'm being a little too generous in saying Reichmann is half right. He's half right on the technical side and completely wrong on the subjective side of DoF. On balance, I would say one quarter right :lol: .

[maderito]

There's another issue here that the maths doesn't address and that's the subjective nature of DoF. When viewing a print, one doesn't bring out the pocket calculator to calculate DoF. It doesn't really matter what the maths says. If it looks as though it's got shallow DoF then it has got shallow DoF and no amount of mathematical formulas can prove otherwise.

Let's look at what really happens when you use say a 100mm lens from 25ft instead of a 200mm lens from 50ft, at the same f stop...

Ray,

An excellent post. Long telephoto lenses deliver a "pop" that seemingly cannot be duplicated with shorter focal length lenses – even though DOF is not very different. I've puzzled over this for a long time. Many of these lenses are "L" grade Canon lenses that have excellent resolution and contrast (which adds to apparent resolution). They are typically used to get in "tight" on the subject against a distant background. They are often shot at or near wide open apertures with lenses that have multiple apertures blades which creates superb bokeh. The sum of these and other effects you mention is to create the impression of outstanding sharpness with unusually narrow DOF.

I think it is mostly an illusion - as it should be. Sharpness is perception, not reality. Just today I uploaded this gallery (http://www.pbase.com/maderito/nh_roadrace) to PBase and noted on a post in the Critique forum here that none of the images were truly sharp. All shots were taken with a Canon 70-200 L lens @ f/4.0 and 200 mm. Judge for yourself -- it's not the DOF but other image features that make many of them seem sharp with very small DOF.

I don't have to revisit the formulas. They lie. Just use the eyes and analyze what you're seeing.

[Rayz]

Woody,
Interesting shots! You've got an out-of-focus background to emphasis the subject, but not too-out-of-focus so that the background is unrecognisable. However, it does appear that the images are not 'tack sharp'. Since the lens is high quality, I can only assume that the shutter speed you used could ideally have been faster. The combination of runner movement and camera movement requires a really fast shutter speed to freeze the moment.

This is one area where the 20D is going to be really useful for me. Upgrading from a D60, I'm going to really enjoy getting tack sharp and relatively noise-free images with my 100-400 zoom which, incidentally appears to be sharpest at f11. I might even be persuaded to buy the 70-200 F4L. The lack of IS, which is of concern with the D60, should be of less concern with the 20D if I can shoot at ISO 800.

You see how the cost mounts up :D . A new camera body; a new 10-22mm lens plus some other good value non-IS lenses which suddenly become more useful. Geez! This is an expensive hobby!

[maderito]

Woody,
Interesting shots! You've got an out-of-focus background to emphasis the subject, but not too-out-of-focus so that the background is unrecognisable. However, it does appear that the images are not 'tack sharp'. Since the lens is high quality, I can only assume that the shutter speed you used could ideally have been faster. The combination of runner movement and camera movement requires a really fast shutter speed to freeze the moment.

This is one area where the 20D is going to be really useful for me. Upgrading from a D60, I'm going to really enjoy getting tack sharp and relatively noise-free images with my 100-400 zoom which, incidentally appears to be sharpest at f11. I might even be persuaded to buy the 70-200 F4L. The lack of IS, which is of concern with the D60, should be of less concern with the 20D if I can shoot at ISO 800...

Well now you've done it - taken us so far off topic that I'm almost embarassed to reply. :)

Among the claims made for the 20D, I don't believe it has been said that it improves hyperfocal distance! However, the relationship between hyperfocal distance and DOF has driven me to the same conclusion you have reached - I'm gonna try out that 20D.

IMO, Rob Galbraith posted the most comprehensive review of the 20D, and because of that review and because of my frustration with OOF sports shots on the 10D in AI servo mode, I'm standing in line for a 20D.

Here's Rob Galbraith's update on his opinion about the 20D AF speed posted here (http://www.robgalbraith.com/ubbthreads/showflat.php?Cat=&Board=UBB8&Number=263780&Forum=A ll_Forums&Words=servo&Searchpage=0&Limit=25&Main=2 63778&Search=true&where=bodysub&Name=&daterange=0& newerval=&newertype=&olderval=&oldertype=&bodyprev =#Post263780)
(scroll down to RG's response):
9/7/2004
Since posting the preview article, Mike Sturk and I have been out shooting a bunch with the 20D and EOS-1D Mark II, including more soccer and football, both in nice daylight and under icky stadium light. These experiences have only cemented our initial opinion of the 20D's AF: it's a quantum leap forward from the 10D, and crosses into the usable-for-sports AF territory in a way that few digital SLR's in this price range have done to date.

But the EOS-1D Mark II is better still, enough that if I were shooting mostly sports and could afford it, I'd want the EOS-1D Mark II. The biggest difference seems to be in tracking performance: the EOS-1D Mark II simply yields more sharp frames in an extended sequence than the 20D, though the 20D is capable of getting the all-important first frame in focus a lot of the time. But the EOS-1D Mark II almost certainly has an edge in this area too.

If money is no object and you're looking for a body to strap your Canon lenses to, the EOS-1D Mark II will suit your needs the best I suspect. If money is a factor (as it is for practically everybody), then I'd second the recommendation to at least consider a used EOS-1D. It offers better AF than the 20D and 8 fps, but not the resolution or the superb high ISO performance of the 20D (which is slightly but noticeably better at ISO 1600 and 3200 than even the very good EOS-1D Mark II). If you shoot sports under perpetually bad lighting conditions, the 20D might actually be the better choice of the two, but the EOS-1D is still a really sweet sports camera in many ways.

Ultimately, it's your money, so you'll have to decide.

FYI, we'll be publishing some sports photos taken with the 20D and EOS-1D Mark II soonish, once we've finished work on a couple of offline projects.

[maderito]

Woody,
Interesting shots! You've got an out-of-focus background to emphasis the subject, but not too-out-of-focus so that the background is unrecognisable. However, it does appear that the images are not 'tack sharp'. Since the lens is high quality, I can only assume that the shutter speed you used could ideally have been faster. The combination of runner movement and camera movement requires a really fast shutter speed to freeze the moment.

This is one area where the 20D is going to be really useful for me. Upgrading from a D60, I'm going to really enjoy getting tack sharp and relatively noise-free images with my 100-400 zoom which, incidentally appears to be sharpest at f11. I might even be persuaded to buy the 70-200 F4L. The lack of IS, which is of concern with the D60, should be of less concern with the 20D if I can shoot at ISO 800...
Well now you've done it - taken us so far off topic that I'm almost embarassed to reply. :)

Among the claims made for the 20D, I don't believe it has been said that it improves hyperfocal distance! However, the relationship between hyperfocal distance and DOF has driven me to the same conclusion you have reached - I'm gonna try out that 20D.

IMO, Rob Galbraith posted the most comprehensive review of the 20D, and because of that review and because of my frustration with OOF sports shots on the 10D in AI servo mode, I'm standing in line for a 20D.

Here's Rob Galbraith's update on his opinion about the 20D AF speed posted here (http://www.robgalbraith.com/ubbthreads/showflat.php?Cat=&Board=UBB8&Number=263780&Forum=A ll_Forums&Words=servo&Searchpage=0&Limit=25&Main=2 63778&Search=true&where=bodysub&Name=&daterange=0& newerval=&newertype=&olderval=&oldertype=&bodyprev =#Post263780)
(scroll down to RG's response):
9/7/2004
Since posting the preview article, Mike Sturk and I have been out shooting a bunch with the 20D and EOS-1D Mark II, including more soccer and football, both in nice daylight and under icky stadium light. These experiences have only cemented our initial opinion of the 20D's AF: it's a quantum leap forward from the 10D, and crosses into the usable-for-sports AF territory in a way that few digital SLR's in this price range have done to date.

But the EOS-1D Mark II is better still, enough that if I were shooting mostly sports and could afford it, I'd want the EOS-1D Mark II. The biggest difference seems to be in tracking performance: the EOS-1D Mark II simply yields more sharp frames in an extended sequence than the 20D, though the 20D is capable of getting the all-important first frame in focus a lot of the time. But the EOS-1D Mark II almost certainly has an edge in this area too.

If money is no object and you're looking for a body to strap your Canon lenses to, the EOS-1D Mark II will suit your needs the best I suspect. If money is a factor (as it is for practically everybody), then I'd second the recommendation to at least consider a used EOS-1D. It offers better AF than the 20D and 8 fps, but not the resolution or the superb high ISO performance of the 20D (which is slightly but noticeably better at ISO 1600 and 3200 than even the very good EOS-1D Mark II). If you shoot sports under perpetually bad lighting conditions, the 20D might actually be the better choice of the two, but the EOS-1D is still a really sweet sports camera in many ways.

Ultimately, it's your money, so you'll have to decide.

FYI, we'll be publishing some sports photos taken with the 20D and EOS-1D Mark II soonish, once we've finished work on a couple of offline projects.

[maderito]

Woody,
Interesting shots! You've got an out-of-focus background to emphasis the subject, but not too-out-of-focus so that the background is unrecognisable. However, it does appear that the images are not 'tack sharp'. Since the lens is high quality, I can only assume that the shutter speed you used could ideally have been faster. The combination of runner movement and camera movement requires a really fast shutter speed to freeze the moment.

This is one area where the 20D is going to be really useful for me. Upgrading from a D60, I'm going to really enjoy getting tack sharp and relatively noise-free images with my 100-400 zoom which, incidentally appears to be sharpest at f11. I might even be persuaded to buy the 70-200 F4L. The lack of IS, which is of concern with the D60, should be of less concern with the 20D if I can shoot at ISO 800...
Well now you've done it - taken us so far off topic that I'm almost embarassed to reply. :)

Among the claims made for the 20D, I don't believe it has been said that it improves hyperfocal distance! However, the relationship between hyperfocal distance and DOF has driven me to the same conclusion you have reached - I'm gonna try out that 20D.

IMO, Rob Galbraith posted the most comprehensive review of the 20D, and because of that review and because of my frustration with OOF sports shots on the 10D in AI servo mode, I'm standing in line for a 20D.

Here's Rob Galbraith's update on his opinion about the 20D AF speed posted here (http://www.robgalbraith.com/ubbthreads/showflat.php?Cat=&Board=UBB8&Number=263780&Forum=A ll_Forums&Words=servo&Searchpage=0&Limit=25&Main=2 63778&Search=true&where=bodysub&Name=&daterange=0& newerval=&newertype=&olderval=&oldertype=&bodyprev =#Post263780)
(scroll down to RG's response):
9/7/2004
Since posting the preview article, Mike Sturk and I have been out shooting a bunch with the 20D and EOS-1D Mark II, including more soccer and football, both in nice daylight and under icky stadium light. These experiences have only cemented our initial opinion of the 20D's AF: it's a quantum leap forward from the 10D, and crosses into the usable-for-sports AF territory in a way that few digital SLR's in this price range have done to date.

But the EOS-1D Mark II is better still, enough that if I were shooting mostly sports and could afford it, I'd want the EOS-1D Mark II. The biggest difference seems to be in tracking performance: the EOS-1D Mark II simply yields more sharp frames in an extended sequence than the 20D, though the 20D is capable of getting the all-important first frame in focus a lot of the time. But the EOS-1D Mark II almost certainly has an edge in this area too.

If money is no object and you're looking for a body to strap your Canon lenses to, the EOS-1D Mark II will suit your needs the best I suspect. If money is a factor (as it is for practically everybody), then I'd second the recommendation to at least consider a used EOS-1D. It offers better AF than the 20D and 8 fps, but not the resolution or the superb high ISO performance of the 20D (which is slightly but noticeably better at ISO 1600 and 3200 than even the very good EOS-1D Mark II). If you shoot sports under perpetually bad lighting conditions, the 20D might actually be the better choice of the two, but the EOS-1D is still a really sweet sports camera in many ways.

Ultimately, it's your money, so you'll have to decide.

FYI, we'll be publishing some sports photos taken with the 20D and EOS-1D Mark II soonish, once we've finished work on a couple of offline projects.

[maderito]

Oops!! a little problem with the forum server - perhaps the mods can delete the extra posts.

[Mark Kemp]

Can anyone help 'me' with hyperfocal settings?



Mark,
If you don't want to do the calculations, then you'll just have to rely on experience :D . Just bear in mind, the shorter the lens, the higher the f stop # and the greater the distance between camera and subject, then the greater the DoF. And of course, the reverse is true.

The basic formula for calculating hyperfocal distance is H=focal length squared over ( f stop# times CoC). The CoC varies depending on desired print enlargement and format size, but generally a CoC of 0.019mm is used for formats such as the 10D and 0.03mm for 35mm. If you are going to use this formula, which is really very simple, make sure all values are in the same units, ie. everything in millimetres for example. (Except of course the f stop number :D ).

Ray

Rayz, I was trying to be ironic ---

Can anyone help.... was the original title of this thread, thats why I put 'me' in quotes.

Oh well - in any case thanks for the info, but I can actually do the sums if I really have to. I just try to avoid it as it makes my head hurt!

[Rayz]

but I can actually do the sums if I really have to. I just try to avoid it as it makes my head hurt!

Dear me! Have you tried Asprin? (Just being ironic :D ).

[Rayz]

Among the claims made for the 20D, I don't believe it has been said that it improves hyperfocal distance!

It's true we're off topic here; one of my failings I'm afraid :D . However, I think I could make a case for the 20D improving (or shortening)hyperfocal distance. Since hyperfocal distance is related to f stop, and f stop is related to shutter speed, and shutter speed is related to ISO, and ISO is related to noise, and noise is related to whether or not you're going to bother taking the shot at all ....... enough said :D .