What am I doing wrong?? (DOF question)

[Madison]

I have been travelling up to the Museum Of Natural History for two days now because I got obsessed with this incredibly weird Baboon. His mounth is terrifying, his eyes look so sad (this is a dead animal on display in one of the staircases, not really a main attraction).





Here's the problem. Yesterday I went and took a shot. The mouth was in focus, the rest was not (picture is here: http://community.the-digital-picture.com/forums/t/850.aspx?PageIndex=26 halfway on the page).


I went back today, to re-do it nd make sure everything was okay. I decided to shoot at F8, which should be more than enough, right? Wrong. Now, the eyes are in focus and the mouth is not. Is it because I am too close to the monkey? That's it, isn't it? I need some tips because no matter what I do, I cannot get the entire monkey in focus. And I can get as close or as far away as I want. I can lick it if I have to.





Tips?


This is the picture from today: (Oh the monkey is up against a white wall with glass next to it and loads of daylight so it's a horror to shoot it anyway because of the enormous contrast and the reflectve surface behind it that I want to keep out of the picture).





/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.25.11/natuurhistorischmuseum_2D00_14.jpg

[Sean Setters]

A few questions:


What camera/lens are you using?


What's the EXIF info? (shutter speed, ISO, focal length)


Just how close were you to the subject?


What focus point were you using and where was it on the animal?

[Fred Doane]

Madison,





All you need is this...


DOF Calculator ("http://www.dofmaster.com/dofjs.html)


DOF has to do with a lot of things. Distance to the subject, focal length,apperature, and even if your using a FF or crop sensor. The calculator will help u with deciding what you'll be able to get in focus. Don't forget after a certain point defraction will be a concern if you use too high of a f-stop. Usually it's said you want the eyes in focus but if the teeth are your main subject you may have to live with the eyes being a little out of focus if you can't use a smaller apperature. I believe all Canon cameras have a DOF preview button next to where the lens is mounted that if you press down the image will get darker inyour viewfinder but it will allow you to see what will be in and out of focus. You might choose to focus on a point between the teeth and eyes since when the lens is stopped down there will be a point in front and behind your focus point that is still in focus (see "in front of subject" & "behind subject" in the online DOF Calculator) Hope this was somewhat helpful.





Fred~

[Madison]

A few questions: What camera/lens are you using?, What's the EXIF info? (shutter speed, ISO, focal length), Just how close were you to the subject?, What focus point were you using and where was it on the animal?





Hi Sean!


I am using the 5D MkII with the 24-105. EXIF for this particular shot is: 1/13 at F8. ISO400. 58mm. I need to be closer to the subject that I would like (not close enough for me to lose my minimum focus distance though) to avoid problems. (There are white walls everywhere, there is a giant glass wall directly behint the baboon that I am trying to hide in the shot, there is MASSIVE back-light as well and high contrast so I overexposed to get the proper exposure for the baboon itself).


I tried focusing with a focus point on the teeth (sharp teeth, unsharp eyes (while the distance between them is minimal) at F8. And I tried using a focus point for the yes resulting in sharp eyes and not so sharp teeth.





Now tell me. With an object like this, it shouldnt be difficult at all to get it all in focus, right? Or am I saying/doing something incredibly stupid? It wouldn't be the frst time.

[Chuck Lee]

I'm going to guess 5DMII with 24-105 f4L.


Based on the info below I'm also going to guess that you were 3ft-5ft away and shooting at 70mm - 105mm to get this close up head shot.


Min focus distance is aprox 18". I'm thinking that if your in a hall you can get a little further away. Say 4-6 ft.


@ 50mm- DOF @f8 from 4ft= 11" 6ft = 25.5"


@ 70mm - DOF @f8 from 4ft= 5.4" 6ft = 12.5"


@ 105mm- DOF @f8 from 4ft = 2.25" 6ft = 5.4"


Like Sean said: What were the settings?


The catch 22: The further back you get the more DOF you will have. The more you zoom in the less DOF you have.


Try shooting at a wider angle and cropping or shoot at a higher aperture. If you can stand another stop go to f16 same composure/crop.


Sorry for stating the obvious.


I'm still struggling with FF and this very same thing.


Oh and...........Don't lick the monkey! It'll leave a bad taste in your mouth I'm sure.


Chuck

[Daniel Browning]

Now, the eyes are in focus and the mouth is not. Is it because I am too
close to the monkey? That's it, isn't it? I need some tips because no
matter what I do, I cannot get the entire monkey in focus.





Once you've decided on a composition, there is no way to get deeper depth of field except by using a narrower iris diameter (i.e., stopping down). [There are some hard ways, such as using a tilt-shift lens or focus stacking.]






And I can
get as close or as far away as I want. I can lick it if I have to.





If you move forward and zoom out (to get the same framing), the DOF will remain the same.


If you move backward and zoom in (to get the same framing), the DOF will remain the same.


So moving doesn't really affect the DOF. However, it does affect the perspective and background blur. Further away + longer focal length = flatter perspective and more background blur.






Tips?





If you want the eyes *and* teeth in focus, then you should focus in between them: not on one or the other. With liveview this is straightforward:

Compose your shot
Start liveview
Zoom in to focus on the closest feature (front teeth)
Hold down the DOF preview button (or use a liveview mode that does it for you).

See how far you can move the focus point back before the teeth go out of focus.
Zoom in on the furthest feature (back eye) to see if it is in focus.

If not, stop down further and go back to step 6.



If you're only using the image on the web (i.e. very low resolution), then you don't need to use the full 10X zoom (that would more accurately depict the DOF of a large print or heavy crop.)






I am using the 5D MkII with the 24-105. EXIF for this particular shot is: 1/13 at F8. ISO400. 58mm.





For most people, 1/13 would not be near enough to prevent camera shake, even with image stabilization. That may be what caused the blur in your image. I would suggest a monopod if you're not using one already. (Of course a tripod is better, but not always an option.) If you only plan to use the image on the web (low resolution), you could use a higher ISO and faster shutter speed.






With an object like this, it shouldnt be difficult at all to get it all in focus, right?





It's not the object, per se, that makes DOF difficult: it's the composition and light. It's possible to get deeper DOF if you change your composition, e.g. doing a full body portrait instead of headshot, but that doesn't help you if you've already got the composition you want. More light it would be easier as well, since then you could use a narrower iris diameter (i.e. stop down).


By the way, the reason I'm using the term "iris diameter" instead of f-number is because it works for all sensor sizes. Some people think that small sensors (e.g. digicams) have a DOF "advantage" over large sensors (e.g. 5D2). But that's not true: once you stop down the 5D2 to the same iris diameter (focal length divided by f-number, e.g. f/45 on 5D2 vs f/2.8 on the digicam): the DOF, noise, and diffraction are all the same as the digicam.

[Madison]

Chuck: EXF data is in my earlier post.


Daniel: Monopods or tripods are not allowed inside the museum. Changing the composition is not an option. First of all there is no body attached to the head, and any other angle than this one will create a huge reflection of the rest of the room because the head hangs next to a giant glass wall (that I am trying to hide from the shot).

I will need to up the ISO and make sure I get the DOF right, I figured at F8 (which I thought should cover it), it would not be a problem but I guess I was wrong.

[Sean Setters]

I think Daniel hit the nail on the head with most of his points. Also, I certainly agree taking a look at the DOF calculator. I've used it several times to get an idea of what my settings ought to be. My first instinct was to suggest a monopod, but I figured they wouldn't be allowed in the museum (as well as flash, probably).


You might want to spend about 5 minutes researching the absolute best way to handhold a camera. I know that seems silly, but I read Joe McNally's Hot Shoe Diaries a while back and he devoted a few pages to the different techniques he uses to steady the camera (while sometimes holding a flash) when he doesn't have access to a tripod/monopod. One thing in particular is that he locks the elbow holding the lens to his body, stands straight up, with his chest weight distributed above his center of gravity. Holding the camera for maximum steadiness probably buys you an extra stop of exposure.

[Chuck Lee]

Madison,


Is there another time of day that would cut down on the glaring light?


Do you have an assistant that could hold up a gobo or two? Experiment with some subtractive lighting.


Daniels suggestion to focus between the eyes and mouth was spot on.


58mm would put you out around 2.5 ft depending on the size of the head? Try ISO800, f11 to ISO1600, f16. Take bursts of 3 and keep the sharpest.Sean's Joe McNally suggestion is good too. The steadier the better.


Would the addition of flash help? Try on camera and off. In fact, you could kill a lot of that back light by introducing key light into that scene. Keep your aperture down around f11 and play your ISO up and down along with shutter speed. Aperture sets the power of the flash on the subject, shutter speed controls the ambient power.


What do you think Sean? Would flash help him in this situation? I think it would.


Unless the museum doesn't allow it. That'd be wack. But it is a stairwell.


Just think'n out loud,


Chuck

[Daniel Browning]

Good post Chuck Lee, just one minor correction:






Try shooting at a wider angle and cropping...





Cropping in post ("digital zoom") will make the DOF just as thin as if you had zoomed optically. The only time a wider angle has more DOF is if you don't compensate for the change in framing (composition). When you compensate for it by moving closer or cropping, it makes the DOF thinner again.

[Madison]

Flash is not allowed anywhere. They're very strict.


It's an old antique building. The stairwell is a huge square room. The head is in one corner. But the light is tricky. I will reshoot in a few days with all these wonderful tips (thanks everyone: i am most grateful for it) and will come back to discuss further!

[Sean Setters]

What do you think Sean? Would flash help him in this situation? I think it would.
<div style="CLEAR: both"]</div>



Flash is kind of a moot point when it comes to museums. Typically speaking, unless the museum is paying you to take pictures of the work, flash is prohibited. It makes sense, of course, because the bursts of light could fade paintings quite severely I imagine. Plus, it's very distracting...

[Rodger]

You might want to spend about 5 minutes researching the absolute best way to handhold a camera.





Madison, I did the research for you :)


It's for left eye focusers (which I am not), but I find that if a situation calls for extreme steadiness, Im alright with focusing through left for a few shots.


Joe McNalley: Da Grip ("http://www.youtube.com/watch?v=EDsx3-FWfwk)


The HTML editor wouldn't let me post the code for the embedded video, but that's the link.


Best of luck to you on the re-reshoot!


-Rodger

[Chuck Lee]

OK, hang on Daniel,


Your confusing me........ again. LOL


According to the DOF calculator my distance of acceptable sharpness increases with:


1) increased aperture (i.e. larger number = smaller iris)


2) increased camera to subject distance


Let's say the subject is yardstick laying on the ground. My camera is on a rail 45 degrees to the ground plane. I am focusing on the 18" mark with center focus point.


At10ft awayI take a photo ofthe subject using 100mm at f4, I should get the same distance of acceptable sharpness as if I had taken the same photo from 5ft away with a 50mm at f4. Themaindifferencesare the field of view and the amount ofbackground blur. This distance on a 5D FF is .71 ft for both shots. So I see the middle 8-10" of the yardstick in acceptable focus.


So what happens when I takea photo from 10ft away with a 50mm at f4?The distance(or depth) of acceptable sharpness is now 2.97ft. I see almost all of the yardstick in acceptablefocus, do I not? The difference is that the yardstick appears half the size in the frame as the other two images.


That's easy to remedy. I just crop the image in half. It goes from 3000 pixels to 1500 pixels. The yardstick now appears to be correctly porportioned in the frame.


The yardstick also appears to be almost completely in focus unlike the other two shots.


What an I missing Daniel? and I mean that respectfully and sincerely.......


Chuck

[wickerprints]

Check your AF accuracy. You may need a microfocus adjustment.
You can slightly reduce camera shake further with a makeshift "monopod." Here's how it works: (a) Take a bolt that threads into the tripod mount of your camera. (b) Measure a length of string that is longer than the distance from the floor to eye level. (c) Tie one end of the string to the bolt. (d) Get a flat washer and tie it to the other end of the string. (e) To use this device, screw the bolt into the tripod mount, unwind the string, and let the washer fall to the floor. Step on the washer and pull the string taut as you compose your photo. The induced tension on the string eliminates vertical camera shake. (f) It also helps to use the timer feature.

Increase ISO by one more stop.
Understand the shape of your subject and select a focus point that has the appropriate contrast and places the DOF in such a way as to maximize the amount of subject in acceptable focus. That is to say, if AF selects a point on the subject that is closest to you, then you are only utilizing the rear DOF. If the AF selects a point on the subject furthest away to you, then you are only utilizing the front DOF.

[Chuck Lee]

wickerprints,


Have you actually tried #2 with success?


Chuck

[Chuck Lee]

If not flash, what about a small handheld reflector? Fold it up an sneak it in?


Anything to get some more light on the camera side of that monkey.


I'll keep throw'n s%&amp;t at the wall. Sooner or later it might stick.

[Sean Setters]

If not flash, what about a small handheld reflector? Fold it up an sneak it in?


Anything to get some more light on the camera side of that monkey.


I'll keep throw'n s%&amp;t at the wall. Sooner or later it might stick.
<div style="CLEAR: both"]</div>



LOL! You know, that's not a half bad idea. However, I imagine he'd need tobe holding it, and that would make his camera grip less steady. I like resourcefulness of the idea, though. If the extra light were needed more forcreativity rather than exposure necessity, I think it'd be a good idea.

[Daniel Browning]

At 10ft away I take a photo of the subject using 100mm at f4, I should get the same distance of acceptable sharpness as if I had taken the same photo from 5ft away with a 50mm at f4.


Agreed.



The main differences are the field of view and the amount of background blur.


I concur. For the reader I would add that perspective is different and point out that it's only the angular field of view (angle of view) that is different: the absolute field of view (i.e. subject framing in feet) remains the same.



This distance on a 5D FF is .71 ft for both shots. So I see the middle 8-10" of the yardstick in acceptable focus.


Yep. Your DOF calculator uses 0.03mm for an acceptably sharp CoC on the 5D2. I get the exact same numbers as you. Here they are in metric:

50mm 1.524m distance f/4 DOF = 0.21m
100mm 3.048m distance f/4 DOF = 0.21m



So what happens when I take a photo from 10ft away with a 50mm at f4? The distance (or depth) of acceptable sharpness is now 2.97ft. I see almost all of the yardstick in acceptable focus, do I not?


Yes. That's what I get too:

50mm 3.048m distance f/4 DOF = 0.90m



The difference is that the yardstick appears half the size in the frame as the other two images. That's easy to remedy. I just crop the image in half. It goes from 3000 pixels to 1500 pixels. The yardstick now appears to be correctly porportioned in the frame.


So far so good.



The yardstick also appears to be almost completely in focus unlike the other two shots.


This part is incorrect. It will have thin DOF like the other two shots. (In some cases it will be blurry *all over* the image, but that doesn't count as a deeper DOF advantage because you could add blur to the thin DOF shots too if you wanted to go for that blurry look.)





I think if you actually try the experiment you described you'll see what I'm talking about. I think I recall seeing similar experiments already on the web, so we should be able to find one with GIYF.

The reason why cropping affects DOF is the same reason why print size affects DOF: reproduction magnification. If you crop the image in half *and* display it at half size (print or screen), then the reproduction magnification stays the same and therefore the DOF is the same. But if you crop the image in half and display it at the *same* size, (as people do in real life) then the reproduction magnification increases, and the DOF goes back to beeing thin.

Cropping in post is very similar to using a "crop" sensor, and has the same result in DOF (if the pixel size is the same on the crop camera and the FF camera). Let's take the 30D and 5D2 for example, since they have the exact same pixel size.

Using 3m focus distance and h/CoC=1200:

30D 50mm f/4 DOF = 0.54m
5D2 50mm f/4 DOF = 0.88m

With the same lens, the 30D has thinner DOF, because it is only seeing the center crop of what the 5D2 sees, but it is still magnifying it to the same print size. What if you crop the 5D2 digitally so it matches the framing of the 30D?

30D 50mm f/4 12x18 print DOF = 0.54m
5D2 50mm f/4 digitally cropped by 1.6X, 12x18 print DOF = 0.54m

Now the 5D2 has the same thin DOF as the 30D. Another variable is print size. If you print it at 1.6X smaller size (7.5x11.25) the DOF goes back to being the same. In fact, it looks like you took a knife to the 5D2 print and just remove the center portion to make the 30D print:

30D 50mm f/4 7.5x11.25 print DOF = 0.88m
5D2 50mm f/4 12x18 print DOF = 0.88m

In other words:

1. 5D2 + lens A + cropping = thinner DOF than 5D2 + lens A + no cropping
2. 30D + lens A = thinner DOF than 5D2 + lens A + no cropping

The best way to look at it is the *total* magnification from the size of the original subject to its size to the viewer.

Any thing you do to make the subject bigger for the viewer will make DOF thinner. Get closer, use a longer lens, use a smaller sensor (with same lens), crop digitally, print larger, move the viewer closer to the print, etc. All of that increases the magnification, and any increase in magnification makes the DOF thinner.

[wickerprints]

Put a bit more simply, the CoC is partly a function of reproduction magnification and partly a function of viewing distance. Changing this magnification necessarily changes the CoC, and in turn, DOF, since now you have affected what you consider to be "acceptably sharp."


DOF is a convenient concept but it is not really how sharpness as a function of focusing distance really behaves. The loss of sharpness is not sudden but a continuous gradation.

[Chuck Lee]

Daniel,


Truly, you have a dizzying intellect!!


All righty then. It's yardsticks at dawn!!


LOL!!


I'll post results tomorrow.


ITMT...............


Madison,


Do you have anyone who can help you? If not, I'd be willing to bet that you could finda museum employee to hold a reflector for you for five minutes.

[Daniel Browning]

Truly, you have a dizzying intellect!!


Wait till I get going!


...


Where was I?


Ah, yes: DOF. And you must have suspected I would have known the<span style="font-family: Century Gothic;"] DOF, so I can clearly not choose the wine in front of me.



All righty then. It's yardsticks at dawn!!


[:D] I'll be there with bells on (and pocket protectors), but be warned: my character wields a +2 sword of nerdliness and +1 buckler of textbooks. [;)]

[Sean Setters]

my character wields a +2 sword of nerdliness and +1 buckler of textbooks. /emoticons/emotion-5.gif
<div style="clear: both;"]</div>





....and a +12 year-old prophylactic, never used, apparently...


he he he

[Rodger]

Daniel,


Your movie reference made me chuckle haha.

[Madison]

...wow I expected something much simpler like (focus on the nose and the eyes will be in focus if they are behint it and your aperture is narrow enough). I have DOF troubles with FF a bit, getting to grips with it. I honestly never shoot something this close and I never have focus issues with the camera or the lens but this monkey is trouble.





I thought. Small object. Stay close to minimum focus distance. Choose F8 to make SURE that everything is in focus (come on, its F8 and a small object, is what I thought) and shoot!





Shoot itw as. But it sounded slightly different than shoot when I said it.

I appreciate the (incredibly technical) DOf explanations but I have to be honest and say that I am having trouble understanding it for a number of reasons (all my fault). For one I have severe discalc (which means my brain somehow refuses anything to do with math, for real), I also don't understand inches (we are metric over here) and because English is not my native language when thinsg get this technical I am in way over my head.





I have a year subscription to the museum anyway so I will go back and figure it out with all the help here.


Not today though. Pooring cats, dogs and monkeys out there.

[alex]

Truly, you have a dizzying intellect!!


Wait till I get going!


...


Where was I?


Ah, yes: DOF. And you must have suspected I would have known the<span style="font-family:Century Gothic;"] DOF, so I can clearly not choose the wine in front of me.






Any thread on The-Digital-Picture.com that contains a Princess Bride reference I hereby propose automaticallybe placed in the Thread Hall of Fame. Because it's just THAT awesome.


DOF....you keep using that word. I do not think it means what you think it means.


I cannot contribute any technical information to this thread, so that's what you get.

[wickerprints]

I thought. Small object. Stay close to minimum focus distance. Choose F8 to make SURE that everything is in focus (come on, its F8 and a small object, is what I thought) and shoot!








You should be aware that the 24-105/4L IS has a "macro" focusing range that enables a maximum magnification of 0.23x. While that is not anywhere near 1:1 true macro (hence the use of quotes), it's quite respectable for a zoom lens of this design. So if you are really getting that close so as to fill the entire frame with this detail, you are losing a lot of DOF even at f/8.


If camera shake turns out not to be the overriding issue with respect to sharpness, you may be able to stop down to f/11 or f/16 to enable more forgiving DOF.

[Chuck Lee]

Daniel,


You only think I guessed wrong! That's what's so funny! I switched glasses when your back was turned! Ha ha! You fool! You fell victim to one of the classic blunders! The most famous is neverbuy a Hasselbladin Asia, but only slightly less well-known is this: never go in against Chuck when a opinion about DOF is on the line!! Ha ha ha ha ha ha ha!! Ha ha ha ha ha ha ha!! Ha ha ha-- I die laughing!!


Here's the test:


10ft was too far to get what I wanted so I back down to 6ft [1.83m] I did not have a "yardstick". Very strange. I can't ever remember in my life not having one in my house. Oh well,I used a tape measure instead. As stated in the earlier post I am taking one photo from 1) 3ft [.91m] 50mm @ f4 and two photos from 6ft [1.83m] 2) 100mm @ f4 &amp; 3) 50mm @ f4. The 3rd photo is cropped to 1/2 of the size of the original image. All images were processed with SilkyPix, No sharpening, +.3ev adjustment on the 100mm f4 shot to match exposure. All images were developed @ 2184 x 1456 pixels which is 1/2 the native resolution of a 5D sensor and USM of 200, .6 was applied. Text was added in CS3 and 640 pixel reductions were created using bicubic sharper for display here. Distance to focus line was established by height to the tripod studplate vs. the linear distance from the stud to the focus line. 45 degree inclination resulted in 51" height x 72" hypotenuse and 25.5" x 36" for half distance. The reading of the tape measure at 3' and 6' simply confirm those setups and does not indicate some horizontal distance to the tripod.


http://www.partsense.com/Photos/DOFTest/DOFTest_01_640.jpg


Original Click Here ("http://www.partsense.com/Photos/DOFTest/DOFTest_01.jpg)


http://www.partsense.com/Photos/DOFTest/DOFTest_02_640.jpg


Original Click Here ("http://www.partsense.com/Photos/DOFTest/DOFTest_02.jpg)


http://www.partsense.com/Photos/DOFTest/DOFTest_03_640.jpg


Original Click Here ("http://www.partsense.com/Photos/DOFTest/DOFTest_03.jpg)


I have to admit that the last image wasn't exactly what I was looking for. The camera/lenssomewhat front focused. I have a second image that is just the opposite and looks backfocused. They were taken back to back using center point focusing on the6' mark. I choose to use the front focused image here.


Results:


We have a great illustration of the perspective difference betweena 50mm @ 3'[.91m]and 100mm @ 6'[1.83] lens.


We also have proof positive that the depth of field does not thin in digital cropping as was the object of this experiment.


If you want to take the time download the 3 images and look at each one in succession noting thetext "New Arctic Ocean" in red next to the Nat Geo Magazine Title. That text appears OOF about the same in all three images. Now look at the number "4" in the 50mm @ 3'[.91m] shot, then the number 3" on the last two. Obviously, as theorized, the number 3" in the cropped 50mm shot appears to be "acceptably sharp". Much more than the 100mm version.


Conclusion: I used a 5D to do this. A 5D MII would yield a 10 megapixel image. This would be comparable (less crop sensor symantics) to the output of a 40D. Focus accuracy is a concern, based on the swing in front-back focus I experienced, So Madison, if you haven't given up on that monkey yet, I suggest when trying this technique, take multiple images, refocusing each time.


I also conclude from this test that I could do the same thing if limited to 3ft by using 25mm @f4, crop 1/2 to increase DOF X4 over 50mm. Or at least twice from what I'm seeing here.


That's my take, so for now I'll shut up.


Anybody want a peanut?

LOL [:D]


Chuck

[Daniel Browning]

We also have proof positive that the depth of field does not thin in digital cropping as was the object of this experiment.


You'd like to think that, wouldn't you? You've bested my sense of humor, which means you must have studied, and in studying you must have learned that DOF is subjective, so you would have put the focus point as far from yourself as possible, so I can clearly not choose the test image in front of me. But, you've also bested my forum post, which means you're exceptionally strong, so you could've put the front-focused shot in your own test image, trusting on your strength to save you, so I can clearly not choose the test image in front of you.

Seriously: I was wrong, and your tests prove it. Backing up and cropping does indeed result in deeper DOF. Furthermore, it shows that digital zoom has deeper DOF than optical zoom. Thank you for your effort. I will try to verify your experiment when I have time.

[David Selby]

your shutter speed is far too low for a hand held image, and your minimum focus distance my be pushing it on that camera.





Finally, you may not have enough light for accurate focusing, but certainly your shutter speed is way to low for 58mm it needs to be at least 1/focal length for hand holding with a full frame and 1/focal length x 1.6 for crops. I recommend always getting even higher if possible.

[peety3]

Once you've decided on a composition, there is no way to get deeper depth of field except by using a narrower iris diameter (i.e., stopping down). [There are some hard ways, such as using a tilt-shift lens or focus stacking.]


If you move forward and zoom out (to get the same framing), the DOF will remain the same.


If you move backward and zoom in (to get the same framing), the DOF will remain the same.


So moving doesn't really affect the DOF. However, it does affect the perspective and background blur. Further away + longer focal length = flatter perspective and more background blur.





So I chose to pull out my handy-dandy iPhone and bring up DoF Plus. It's set to the Canon 40D at the moment.


Field of view, 35mm at 10m to subject, is 6.32m wide by 4.21m high.


Field of view, 70mm at 20m to subject, is 6.32m wide by 4.21m high.


Depth of field, 35mm at f/4 (8.75mm iris) at 10m to subject, is 6.17m to 26.19m (-3.83m and +16.19m).


Depth of field, 70mm at f/4 (17.5mm iris) at 20m to subject, is 15.27m to 28.94m (-4.73m and +8.94m).


Depth of field, 70mm at f/8 (8.75mm iris) at 20m to subject, is 12.35m to 52.38m. (-7.65m and +32.38m)


So, if I keep the same aperture, I get different DoF. If I compensate for iris diameter, I get double the DoF. I get the same numbers using the link found elsewhere in this thread. Am I mis-interpreting something? I'm not seeing the DoF staying the same with unchanged iris diameter.

[Daniel Browning]

I was wrong; thank you for the correction, peety3. I made at least two mistakes:

1. I should have said f-number instead of iris diameter. ("Once you've decided on a field of view, it doesn't matter how close or far away you are: there is no way to get deeper depth of field except by using a narrower f-number.")

2. Even that corrected version is inaccurate the closer you get to the hyperfocal distance. If you change your example from 10 meters to 1 meter, you will find that it is very accurate:


Depth of field, 35mm at f/4 at 1m to subject, is 122mm


Depth of field, 70mm at f/4 at 2m to subject, is 122mm


My mistake was that I didn't point out this important weakness in the approximation. Even so, I think the approximation is still very useful because a lot of photography, including the OP, is not close to hyperfocal, but I will remember to point that out in the future. While I'm at it I might point out that extreme macro can throw off the approximation as well, thanks to bellows factor.

[dmckinny]

Madison,


After wading through the fire swamp of posts on DOF I remembered a suggestions made by a gentleman name Scott Kelby in on of his books. Basically, go to someone in charge of the museum and see if you can arrange special access. I'm guessing if you talk to them for a few minutes and explain the situation you might be able to some help. If you can get in the museum either 30 minutes before public opening, or stay for 30 minutes after closing, they should have no problem with you setting up a tripod, and maybe even flash. You could offer to give them a copy of the finished picture as part of the negotiation.


Now that my hero Daniel has been bested, I have to assume that this thread is mostly dead.


Have fun storming the castle!


David

[Daniel Browning]

Now that my hero Daniel has been bested, I have to assume that this thread is mostly dead.





Whoo-hoo-hoo! Look who knows so much! It just so happens that your thread here is only MOSTLY dead. There's a big difference between
mostly dead and all dead. Mostly dead is slightly alive. With all dead,
well, with all dead there's usually only one thing you can do. Go through the posts and look for loose pixels.







Have fun storming the castle!





Since the invention of the forum, there have only been five threads that were rated the most passionate, the most pure. This one left them all behind.

[peety3]

I was wrong; thank you for the correction, peety3. I made at least two mistakes:

1. I should have said f-number instead of iris diameter. ("Once you've decided on a field of view, it doesn't matter how close or far away you are: there is no way to get deeper depth of field except by using a narrower f-number.")

2. Even that corrected version is inaccurate the closer you get to the hyperfocal distance. If you change your example from 10 meters to 1 meter, you will find that it is very accurate:


Depth of field, 35mm at f/4 at 1m to subject, is 122mm


Depth of field, 70mm at f/4 at 2m to subject, is 122mm





I still don't understand comment #1. I'll repeat the stats I found (with arbitrary numbers):


Depth of field, 35mm at f/4 at 10m to subject, is 6.17m to 26.19m (-3.83m and +16.19m).


Depth of field, 70mm at f/4 at 20m to subject, is 15.27m to 28.94m (-4.73m and +8.94m).


Both of these have identical field of view according to DoF Plus. These do not have the same DoF.


I think your guideline may only apply at compositions where the field of view is larger than the focal length, or perhaps where the DoF is within 2 x &lt;focal length&gt;, or something like that.






Another variable is print size. If you print it at 1.6X smaller size
(7.5x11.25) the DOF goes back to being the same. In fact, it looks like
you took a knife to the 5D2 print and just remove the center portion to
make the 30D print:

30D 50mm f/4 7.5x11.25 print DOF = 0.88m
5D2 50mm f/4 12x18 print DOF = 0.88m





So why isn't there a "print size knob" in ANY of the DoF calculators that I've found?

[Daniel Browning]

I still don't understand comment #1. I'll repeat the stats I found (with arbitrary numbers):


Depth of field, 35mm at f/4 at 10m to subject, is 6.17m to 26.19m (-3.83m and +16.19m).


Depth of field, 70mm at f/4 at 20m to subject, is 15.27m to 28.94m (-4.73m and +8.94m).





It only applies to situations that are not hyperfocal, as mentioned in #2. Try with closer focus distances such as 1m.



So why isn't there a "print size knob" in ANY of the DoF calculators that I've found?


Because most of them assume "same, but unspecified, print size". The best ones do have print size knobs:


http://eosdoc.com/jlcalc/ ("http://eosdoc.com/jlcalc/)