min f stop for those that use 85mm 1.4 for portrait shots

Qarik

I have been on a depth of field study recently. I have the nikkor 85mm 1.4 and am trying to get a more natural feel for what I should be setting my aperture at to get a portrait in focus. My finding were a bit surprising to me so i thought I would share.

So I brought out my tape measure and laid it on the floor and had my wife stand against the door. I then got to proper distance for tight a head shot. Yeah yeah..she blinked!



I looked down at my tape measure and noted I was about 4 feet give or take few inches.

Then I moved to get a upper torso shot:

This was at about 8 feet.

Then I moved back further to get an almost complete head to ankle shot:

This was at about 16 feet

So what f stop should I use to get everything in focus for each of these types of portraits? I had been choosing say 2.8 for a head shot previously..was this justified? 1st I made these assumptions which may use some tweaking. I assumed for the tight head shot you need say 4 inches of DOF if focused on her eyes to get the the face, ears, hair, in focus. Using DOF calculator you need to be at f/13.5 to achieve this! That seemed rather high to me. I swear I was getting in focus shots with a lot lower f stops.

For the waist up shot at 8 feet I assumed a 6 in DOF is needed and that calculated to a min fstop of f/4.8

For the head to ankle shot at 16 feet I assumed 8 in of DOF is needed and that calculated to a min f stop of f/1.4.

Really? I can only use my lens wide open on a person from 16 feet away?

Then I began thinking, well what does the DOF really mean? Perhaps it is technically what is in perfect focus..but what about what is "acceptable" focus. So then I set up this experiment. I laid a measuring tape out on a table and took a photo of it with single site focus on the 85mm lens at 1.4 and 54 inches away centered closely to the 12 in mark on the tape. Note the min focusing distance of the lens is round 3 feet.



a 100% crop of the 12 in mark



Now you can measure what you judge to be "acceptable" focus. I know everyone will have different opinion on this and it depends on different factors. My eye tells me that the focus is reasonable from 12 inche mark to about 12 and 7/8 mark. This translates to 7/8 of an inch or 0.875 inches on the TAPE.

Now to compensate for the fact that I am shooting at an angle to the tape measure, *busts out 9th grade geometry book*. The triangle that I formed is right triangle with the hypotenuse of 54 in and legs 48 in and 25 in. The tape measure is on the 48 in leg but the DOF of the lens and camera is centered on the 54 in leg. I have to multiply the 0.875 with 48/54 to get the DOF on the plane of the camera/lens. The result is 0.78 inches of DOF.

Now what does the DOF calculator say? at 85mm and 1.4 and 54 inches away, the official DOF = 0.54 inches. So what does that mean? It appears (at least in my eye) that the "acceptable" DOF is actually larger then the DOF you get from the calculator by quite a bit. Note that an f stop of 2 will give you a DOF of 0.77 in which is quite close to the 0.78 in of "acceptable DOF". This translate roughly to 1 full fstop between the calculated and "acceptable" f stop.

So one might roughly assume on 85mm
1) head shot at calculated fstop is 13.5 but you can get away with say 9.5.
2) torso shot at calculated fstop is 4.8 but you can get away with say 3.5.
3) full length shot, you can use any fstop!

This kind of analysis allows you to to set an fstop and fire away at heads or torso knowing you are going to be in full focus and not miss a nose or ear, etc. Assuming you hit the right spot focal point.

Note my wife is short! The calculus will change a bit for wilt chamberlain!:wink

I realize this is not scientific but it makes sense to me! I would like to hear if folks agree with this basic assessment/methodology...though of course we may diasgree on the numbers a bit :ivar

pathfinder

A full frame camera, like the D3, will have shallower DOF than your D40 also.

Qarik

Path finder, do you see any issue with my methodology on this?

Also the one thing that has been bothering me..the data on the DOF calculators..how is that determined? How do they know what is in focus and what is not? It seems every lens would be different to some extent according to the design of the optics.

Finally..with the FF sesnor, I have to disagree. The 85mm the DX sensor is like a 130mm while on FF sensor it is a true 85mm. It would appear the FF would have more DOF then the DX. The calculators proves this out when I choose a D3 vs D300 as well.

ziggy53

Qarik wrote:

...the data on the DOF calculators..how is that determined? How do they know what is in focus and what is not? It seems every lens would be different to some extent according to the design of the optics.

...

Start your education with:

http://en.wikipedia.org/wiki/Depth_of_field

Then:

http://www.luminous-landscape.com/tutorials/understanding-series/dof.shtml

Qarik wrote:

... It would appear the FF would have more DOF then the DX. The calculators proves this out when I choose a D3 vs D300 as well.

Nope, the calculators indicate just the opposite and agree with Pathfinder. I think the confusion is that you need to adjust the distance as well as the imager size as the smaller imager requires more distance to subject for the same view at the same focal length.

The classic and often quoted site is here:

http://dofmaster.com/dofjs.html

DOF calculators and tables are just guidelines and starting points. Always temper the knowledge with practical experience and adjust according to the particular application.

pathfinder

Qarik wrote:

Path finder, do you see any issue with my methodology on this?

I think your basic methodology is sound.


For myself, I have never really played with the DOF calculators, I just use the depth of preview button on my camera and my experience by the seat of my pants. That my sound like heresy to some folks, but I usually know what aperture I need for the DOF I want from past exerience. As I have said previously, I rarely shoot at apertures smaller than f4 without a specific decision to do so. I have been burned with f2.8 too many times. For groups, I rarely shoot smaller than f5.6 at a minimum, for the same reason. Bad experience. Do not assume that I never shoot with large apertures though, I have a series of candids that I shot at f1.2 ( love the bokeh), and f2. You just need to know and understand the requirements of our tools.

Also the one thing that has been bothering me..the data on the DOF calculators..how is that determined? How do they know what is in focus and what is not? It seems every lens would be different to some extent according to the design of the optics.

The data in the DOF calcculators is directly from an optical formula. The depth of field is a function of the distance to the subject plane and the aperture, focal length is a minor consideration.

Focal length really has a minimal effect on DOF - IF - the subject is held to a constant size on the sensor. Now IF you move the subject closer to the camera, the DOF decreases as you approach the camera, that is why the dof is the same for a wide angle lens and a telephoto IF the subject size is held constant. But it rarely is held constant, that is why we choose a wide angle, and feel that they have greater dof than long telephotos. But those relationships only hold up for a given format size. A telephoto focal length ( say 40mm ) on a P&S, is a normal lens on a DSLR, and a very wide angle lens on an 8x10 view camera.

The determination of the DOF calculators is done by calculating a least circle of confusion, a numerical value representing how small something must be to be seen in the film plane with an appropriate degree of magnification, if your negative going to be magnified in an enlarger 4x, 10x, 30x etc. In other words, the circle of confusion size is ultimately chosen by a human being. Circles of confusion relate to resolution of line pairs. Most of those decisions were made ~ 80 years ago in the optical industry in Germany at Zeiss, Leica and others.

For myself, if an image is truly sharp at 100% after capture sharpening in Adobe Raw converter, I am satisfied. I know that I can make 24 x 26 in prints that are acceptable with a full frame DSLR used with good technique if I do my part and the lens is good quality. I have the prints to prove that to my satisfaction.

Finally..with the FF sesnor, I have to disagree. The 85mm the DX sensor is like a 130mm while on FF sensor it is a true 85mm. It would appear the FF would have more DOF then the DX. The calculators proves this out when I choose a D3 vs D300 as well.

I will need to see your DOF calculator.

What distance did you choose for a film to subject distance for each format with the 85mm lens?

I assure you that full frame DSLRs have less DOF than a APS sensor DSLR from long experience. Now you have to be careful here and define what lens and what distance and what aperture. But as a general rule, the smaller the format, the greater the DOF for a given focal length. That is why point and shoots make such great macro cameras - they have smaller format sized sensors, and thus inherently greater DOF particularly at close distances.

You can read about DOF here - http://en.wikipedia.org/wiki/Depth_of_field

Maybe Ziggy will drop in here and straighten all this out.

Qarik

thanks for teh detailed reply. Bothe you and ziggy agree that the DOF will be shallower on a FF. But then how do I reconcile with this DOF calculator?

http://dofmaster.com/dofjs.html

I choose the d300 as the camera (DX) then 85mm, 1.4 and 10 feet as inputs. I get a DOF of 0.23ft

Then I simply change the camera to D3 (FX) and the DOF changes to 0.35ft

LiquidAir

Qarik wrote:

thanks for teh detailed reply. Bothe you and ziggy agree that the DOF will be shallower on a FF. But then how do I reconcile with this DOF calculator?

http://dofmaster.com/dofjs.html

I choose the d300 as the camera (DX) then 85mm, 1.4 and 10 feet as inputs. I get a DOF of 0.23ft

Then I simply change the camera to D3 (FX) and the DOF changes to 0.35ft

To get the same field of view with a FF frame camera and the same lens, you will need to move closer than 10 feet. Alternately you can stay at the same distance and use a longer lens. More or less the FF equivalent of 85mm is 135mm.

Qarik

heh..that doesn't make sense to me semantically. If I have to meove closer with the FF camera to achieve the same DOF as the DX..I would intuitively conclude that the FX sensor has more DOF..because in fact it does? ie at the same distance, fstop and zoom..you get more DOF of FX then the DX?

I must be missing something here.

pathfinder

Qarik wrote:

thanks for teh detailed reply. Bothe you and ziggy agree that the DOF will be shallower on a FF. But then how do I reconcile with this DOF calculator?

http://dofmaster.com/dofjs.html

I choose the d300 as the camera (DX) then 85mm, 1.4 and 10 feet as inputs. I get a DOF of 0.23ft

Then I simply change the camera to D3 (FX) and the DOF changes to 0.35ft

First of all, from a practical standpoint in the field, 0.23 and 0.35 are the same to me.

But the field of view of the D3 with an 85mm lens will be significantly larger ( wider view angle ) at 10 feet than it is on the D300. The 85mm on a D300 acts like a stronger telephoto - narrower angle of view.

In other words, your DOF calculator is calculating that you stand in the same position with the same lens and aperture but different bodies. But you won't really do that because the view is so different when using an 85mm lens on a full frame body, and then using the same lens on a DX body, which is what Liquid Air was referring to.

I remember the first time I shot a 40D with a 135mm lens, and a 200mm lens on a 1DsMkll and realized I saw exactly the same angle of view in the viewfinder. It is one thing to know this in your head, and another to see it through the viewfinder.

As I said earlier, a 40mm lens on a point and shoot is a telephoto angle of view, but a 40mm normal lens on a full frame 35mm camera, and on an 8 x 10 viewcamera becomes an extreme wide angle lens. The focal length is exactly the same, what changes is the area of the image circle at the film plane. It is easy to keep things in focus in the small image circle of a point and shoot sensor, but to do that for an 8x10 in negative is much harder. That is why we never seen aperture of f64 for a point and shoot ( and diffraction also of course ) but 8x10 shooters frequently do use an f64 aperture.

That is why I don't use a DOF calculator, but look through my viewfinder ( that viewfinder is what is great about an SLR ), press the depth of field preview button, let my eye adapt to the darkness, and look for dof myself.


I know that for the same aperture, focal length, and film to subject distance, the DOF is less for larger format than for smaller formats. One of my sources for this is here - http://en.wikipedia.org/wiki/Depth_of_field#DOF_vs._format_size

pathfinder

Hey Ziggy,

It has been a while since I last read Michael Reichman's treatise on DOF http://www.luminous-landscape.com/tutorials/understanding-series/dof.shtml

Why does he call the Circle of Confusion COF instead of the more usual CC

ziggy53

Qarik wrote:

heh..that doesn't make sense to me semantically. If I have to meove closer with the FF camera to achieve the same DOF as the DX..I would intuitively conclude that the FX sensor has more DOF..because in fact it does? ie at the same distance, fstop and zoom..you get more DOF of FX then the DX?

I must be missing something here.

Please do read the articles at the links I posted:

ziggy53 wrote:

Start your education with:

http://en.wikipedia.org/wiki/Depth_of_field

Then:

http://www.luminous-landscape.com/tutorials/understanding-series/dof.shtml

...

pathfinder

There was a great article I saw years ago, I may have to duplicate it myself, where they photographed a doll with a series of lenses from 24mm to 400mm on a 35mm film body.

In each shot, the camera was moved so that the doll was always the same vertical dimension at the film plane. The fascinating thing was that the depth of field was the same in each image, proving that the DOF is not really a function of focal length like we typically believe it to be. This was true because the subject to film plane was not constant but changed for each lens. They held subject size constant.

Google is our friend as is M Reichman

Here is a link demonstrating that DOF is not a function of focal length if subject size is constant

ziggy53

pathfinder wrote:

Hey Ziggy,

It has been a while since I last read Michael Reichman's treatise on DOF http://www.luminous-landscape.com/tutorials/understanding-series/dof.shtml

Why does he call the Circle of Confusion COF instead of the more usual CC

It's always been either:

CoC
COC
CC

... as far as I know.

Then again:

http://www.photoxels.com/tutorial_dof.html

Qarik

Thanks for the links Ziggy. I read both articles and had an "AHA" moment. I think I was thinking about it incorrectly...with all else being equal if I merely swap out the camera body from DX to FX, the DOF does increase but so does the shot.

For example if I had done a tight headshot on the DX then it might more of head and shoulders shot on the FX with the same lens and distance. From this POV it does not make sense to compare the DOF of the 2 images because they are different relative size on the sensor. NOW if i step closer on the FX camera to achieve the relative same size (headshot) on the sensor then indeed the FX is shallower.

I guess you have compare DOF with the same relative image size on the sensor and not the same distance.

I get it now. Thanks for the patience.

NateW

What about this?
Scheimpflug
(I'm sure I've not yet fully grasped the science behind various concepts you guys seem to have grasped, so I'm somewhat hopeful that y'all can explain this in a manner than makes sense to us relative ignoramuses. :my-brain-hurts:)

Thanks!

ziggy53

NateW wrote:

What about this?
Scheimpflug
(I'm sure I've not yet fully grasped the science behind various concepts you guys seem to have grasped, so I'm somewhat hopeful that y'all can explain this in a manner than makes sense to us relative ignoramuses. :my-brain-hurts:)

Thanks!

The "Scheimpflug Principle" is easiest to understand when you see it demonstrated on a view camera with tilts and swings. The only time a normal dSLR user would deal wih it is if you had a lens capable of similar actions like the Canon TS-E lenses or the Nikkor PC series lenses.

If you don't have access to one of those lenses take a standard/normal lens off the camera and make sure the aperture is wide open. Point the lens at a very well lit target like a newspaper and hold a piece of paper behind the lens to project the image upon. Start with the newspaper parallel to the projected image on the paper.

Make sure you get a sharp and clear image projected and then tilt the lens fore and aft and observe how it affects the upper and lower parts of the image.

Now tilt the newspaper with respect to the projected image. Again, tilt the lens until you get the sharpest image top and bottom. The angle of tilt of the lens at sharpest focus should be halfway between the tilt of the newspaper and the projected image.

The "Scheimpflug Principle" would allow you to predict and set the proper angle for the lens by knowing the angles of the subject and the respective angle of the focal plane of the camera.

You would also need additional attention to the f-setting and a smaller f-stop would generally be selected depending on the effect required.

Here are a couple of Wiki sites to explore:

http://en.wikipedia.org/wiki/Scheimpflug
http://en.wikipedia.org/wiki/View_camera

NateW

Now that I've got it a bit better, I understand why this didn't come up before in the thread, though the tape measure image is definitely what triggered the memory.

ziggy53 wrote:

... <good stuff> ...
http://en.wikipedia.org/wiki/Scheimpflug
http://en.wikipedia.org/wiki/View_camera

Thanks!
That does help to explain in much more layman's terms.
To further simplify (and confirm I've got it):
Scheimpflug Principle explains what lens angle will allow the maximum amount of a subject will be acceptably sharp when subject plane and film plane are not parallel.
I.e.: how to get a receding picket fence (or road) in all in focus without shrinking aperture too much.

All this is inspiring me to pick up the Ansel books again....
My dad had written (highly modified?) something trying to explain the Scheimpflug Principle a couple years ago (back when I was originally reading the Ansel Adams' books) and I didn't really get it then. If what I've oversimplified above is right, then I think I've finally got it. <img src="https://us.v-cdn.net/6029383/emoji/clap.gif&quot; border="0" alt="" >

Thanks!
Nate

PS: What my dad wrote, just for fun:

Jay Warner wrote:

"Scheimpflug"
-- sch as in "sh" of "shush up"
-- ei as in i - long i - I went to the store.
-- eimp as in I'm, with a bit of a cut off: I'm|p
-- flu as in "flue"
-- g as in a hard g

Accent on first syllable

Shimp-flug only with a long i. Matches "shine" spoken by a Scot. -
word clipped short.

Take a photograph of a brick wall at an angle to the wall so that
only one vertical line is in focus, and as you blow it up, less and
less of it will pass your 'in-focus' criterion.
When the Scheimpflug principle is used to adjust the film and lens
planes and get more of the flat road into focus on the film, the
depth of field, measured at each point on the image, remains the
same - it is controlled by the lens aperture (in f/ stops), the
focal length, and the distance of the subject. If you use a wide
open lens at say f/4 (that's going to be a large view camera lens!)
and adjust to get the horizontal roadway into focus, you will see
that a person standing in the road has their feet in focus, but
their head may be out of focus. If they are reasonably close to the
camera, a smaller f/ stop will be needed to get both the road and
the person's head into focus at once, and no amount of Scheimpflug
adjustment will get away from that.

Qarik

I was unbale to sleep the other night and I found my mind wandering back to this subject of DOF and full frame cameras. I was trying to reconcile my observations, the DOF calculators and the comments from the folks on this board.

My initial assertion that the DOF was greater for the FF camera when the zoom, aperture, and distace are the same..I think still stands. The revelation I had was that the apparent image size on the sensor changes and so it is difficult to make an apples to apples comparison with dx sensor. If you match the image size on the sensor with FX camera by moving closer to the subject then the DOF is indeed shallower. This the apples to apples comparison!

Pathfinder finder pointed out that DOF is the same for different zooms as well with a sample shots using different zooms. I thought about this as well. The sample shots show the same image size on the sensor..but obviously you need to be at different distances to the subject to get those shots. The key is that the distances are indeed different. I wouldn't say that different zooms have the same DOF...I would say that you can achieve the same DOF for different zooms by matching the subjects relative image size on the the sensor (by changing the distnce to the subject).

In the end DOF is still a function of zoom, distance, and aperture.

So thanks for the links and input. I really did learn much about DOF in this process/threads, etc.

LiquidAir

Qarik wrote:

My initial assertion that the DOF was greater for the FF camera when the zoom, aperture, and distace are the same..I think still stands.\

Sort of...

If you crop the FF image to match the APS-C camera then the two images will be exactly the same (ignoring resolution differences).

The reason most DoF calculators will tell you a larger sensor has more DoF at the same distance and focal length is because the CoC size is determined assuming a standard print size. This means that the CoC gets larger as the sensor gets larger with no regard to resolution.

However, if you upgrade from a 8MP 20D to a 15MP 50D you probably increase your standard for sharpness since you now have more pixels. That increased standard of sharpness means your CoC should get smaller; so arguably a 50D has less DoF that a 20D despite the fact that the sensor size stays the same. Arguably in the world of digital where we pixel peep sharpness, the CoC should be determined by the sensor resolving power rather than either the sensor size or the sensor resolution.

In the end, DoF is somewhat subjective; you have to decide what standard of sharpness to use based on what you plan to do with the image.

ziggy53

LiquidAir wrote:

Sort of...

If you crop the FF image to match the APS-C camera then the two images will be exactly the same (ignoring resolution differences).

The reason most DoF calculators will tell you a larger sensor has more DoF at the same distance and focal length is because the CoC size is determined assuming a standard print size. This means that the CoC gets larger as the sensor gets larger with no regard to resolution.

However, if you upgrade from a 8MP 20D to a 15MP 50D you probably increase your standard for sharpness since you now have more pixels. That increased standard of sharpness means your CoC should get smaller; so arguably a 50D has less DoF that a 20D despite the fact that the sensor size stays the same. Arguably in the world of digital where we pixel peep sharpness, the CoC should be determined by the sensor resolving power rather than either the sensor size or the sensor resolution.

In the end, DoF is somewhat subjective; you have to decide what standard of sharpness to use based on what you plan to do with the image.

Excellent explanation LA.

Scott_Quier

LiquidAir wrote:

Sort of...

If you crop the FF image to match the APS-C camera then the two images will be exactly the same (ignoring resolution differences).

The reason most DoF calculators will tell you a larger sensor has more DoF at the same distance and focal length is because the CoC size is determined assuming a standard print size. This means that the CoC gets larger as the sensor gets larger with no regard to resolution.

However, if you upgrade from a 8MP 20D to a 15MP 50D you probably increase your standard for sharpness since you now have more pixels. That increased standard of sharpness means your CoC should get smaller; so arguably a 50D has less DoF that a 20D despite the fact that the sensor size stays the same. Arguably in the world of digital where we pixel peep sharpness, the CoC should be determined by the sensor resolving power rather than either the sensor size or the sensor resolution.

In the end, DoF is somewhat subjective; you have to decide what standard of sharpness to use based on what you plan to do with the image.

The highlighted statement above would seem to encourage one to determine the appropriate aperture by the seat of their pants, ala "the PF technique", rather than by using such tools as DOFMaster - at least until such time as such tools are updated to account for pixel pitch. That sure does make me feel better. That's the way I've been doing it for the last couple of years as I could never get DOFMaster results to reconcile with my experential data/results.

pathfinder

Come on Scott, using the DOF preview button is not my idea, but has been used by shooters for many decades.:D I know you know this also!

I just don't have the patience to sit and work out DOF with the online computers - as I need the information in the field, at the time of shooting.

I admire you folks who do take the time to think their shots through this accurately, but I just tend to use my viewfinder, and my previous experience. I tend to shoot at higher apertures also - as I said, when I shoot at f2.8 or f1.8, it is for a specific, defined reason, as I know that unless I am very careful, my DOF will not be large enough.

Qarik

hmm..do I even have DOF preview button on my D90? I don;t recall seeing one heh.:deadhorse

Scott_Quier

pathfinder wrote:

Come on Scott, using the DOF preview button is not my idea, but has been used by shooters for many decades.:D I know you know this also!

Of course - I was just messing with you

pathfinder

Qarik wrote:

hmm..do I even have DOF preview button on my D90? I don;t recall seeing one heh.:deadhorse

I am sure you do - it is the button near your lens mount that makes the image go all dark when you press it. It stops the iris diaphragm down to the aperture the photo will be shot at - you view through the lens in a modern SLR with the iris diaphragm wide open, but when you press the shutter, before the shutter actually opens, the iris diaphragm stops down to the selected aperture, the shutter opens and closes after the mirror clangs up out of the way, and then the shutter closes, the mirror slams back down, and the iris diaphragm opens back up.

The DOF preview button stops the iris diaphragm down so you can see what the image actually looks like at f5.6. pr f11. or f16. I am sure Nikons do this as well as Pentax, or Mamiya or Canon. This is not true with Point and shoots though, just SLRs.

Now if you stop a lens down from f2.8 to f8 - that is 3 full stops less light, or 1/8th the amount of light through the aperture, and it immediately looks DARK!! Slow down, wait just fifteen seconds, and your eye begins to adjust and you can begin to see with you lens stopped down. This is not a 1 second affair - it does not work that way. But done right, can be quite useful, that is why they include a DOF preview button on your camera.