Something I learned...

If you're trying to see the difference aperture makes in terms of sharpness (as opposed to depth-of-field), put your camera on a tripod, shoot at an object with fine detail (classified ads in a newspaper will work), place it far enough (say 6-8 ft), position it so that it is perpendicular to the lens axis, turn your AF/VR off, go into manual focus and focus (manually) at the center of the target using live-view magnified 10x. Take your first shot wide open, then reduce the aperture by a stop and do another shot. Keep going until you're reached the minimum aperture. Now upload your shots and check two spots: the center, and the upper right corner.<BR><BR>Many lenses will be somewhat soft wide open, then sharpen to their maximum around the mid-range apertures, then start softening again at higher apertures. It is quite instructive to see the difference between the edges and the center. If you go through this exercise, you will have an excellent idea of just how sharp your lens is, and where in the aperture range you will get this sharpness.<BR><BR>Another exercise, while you got the camera on a tripod and the target in place, is to change the camera to auto-focus, and get the camera to focus. Often, you'll find that the AF is less sharp than your manual efforts. This could be a case of lens/camera calibration or misadjustment. If the manual and auto focus are identical, be happy. If they are not, you may want to use a different test that shows you whether you are back- or front-focussing.

Okay so after following the advice of pgriz, here are my results. I shot using every aperture my 50mm f1.8 lens had (1.8-22). #1 is at f1.8, #2 is at f8, and #3 is at f22. I can't believe I have been shooting almost all my photos at f1.8 in order to get the shallow DoF. Meanwhile they are all out of focus. I can't believe I never noticed though. These were all shot in Av mode, with a tripod, and I changed the WB to tungsten. Don't know why I didn't just shoot JPEG? Also I manually focused using the live view at 10x magnification and focused on the line "activities. Basic math and" in the middle and towards the bottom of the page. Anywho here are the photos:

1. f1.8
f18.jpg



2. f8.0
f80.jpg


3. f22.0
f220.jpg


Thank you all for your advice and comments. I know this might seem insifnifcant to some but I am thrilled with what I have just learned.
 
I hold my breath when shooting if I'm extremely worried about shooting. I'll also shorten my camera straps and spread them apart with my elbows to make the strap taught to further steady the camera.

For testing purposes I can see why a tripod is necessary.
 
ConradM said:
Tripod is a must for pin pointing the narrow focus. I use a tripod + zoom focus assist.

EDIT: if you were to try and hand hold @ f1.8 with something that close, the focus would be off just by you breathing.

Tripod is not a must for shooting wide open. I have pictures shot close up, handheld at 1.4 that are in perfect focus. So, no a tripod isn't necessary at shallow DOF. Neither is live view. For what the OP is doing - then yeah use a tripod.

I do use a tripod and live view when I am shooting macro but other than that - no. I have even taken macro shots handheld and they were in focus. Macro at f/11 has less DOF then f/1.8.

It takes practice to nail focus with shallow DOF but it can be done with and without a tripod.
 
I can't believe I have been shooting almost all my photos at f1.8 in order to get the shallow DoF. Meanwhile they are all out of focus.

I think DOF is the least of your focusing problems... If you did as pgriz suggested, then you either need glasses or you misunderstood his suggestion. That first shot at f/1.8 is STILL OOF. Its not even that your lens is soft at that aperture either because the bottom left corner is much much sharper than the center. Sure the others are sharper since you've increased the DOF, but you are still missing focus regardless.
 
With the right technique you can achieve pretty good sharpness wide open with a relatively slow shutter. Notice how the edges are quite soft, which is the result of shooting 1.8. When used in the right context, it can work though. I don't think it takes anything away from this photo.

100ASA film
1/30 shutter
f1.8 50mm

 
@ birdfish: You&#8217;re part of the way there. Your target was not square ( ie, perpendicular) to the lens axis, and you got very obvious DOF issues as a result. So part of the exercise that you should have achieved , you&#8217;re not getting. S&#8217;OK. Let&#8217;s review what you have learned. Getting your focus nailed when wide-open (at f/1.8) is effing hard when you&#8217;re manual and relying on the camera&#8217;s AF mechanism. Deep DOF (through a small aperture) covers a lot of focusing errors. Too little aperture (as in the f/22 example), starts killing the image quality due to diffraction.
You still don&#8217;t know if your AF gives you the same results as carefully manually focusing. That is important information.
You still don&#8217;t know what your lens &#8220;sweet spot&#8221; is in terms of aperture.
I&#8217;d suggest re-doing the test, but this time make sure that you are shooting at the CENTER (maybe mark it with a dot), that the target is fully square vs. the lens axis, and post 100% crops of the center and the corners for comparision. If done right, the difference in focus between the center and the edges should not be great (unless you have a really bad copy of a lens).
@ ballistics: A tripod is not a must when shooting &#8220;normally&#8221;. However, when shooting a test sequence, you want to minimize the number of uncontrolled variables, and the tripod helps there by keeping the same camera-target distance, eliminating the shift in focus, and eliminating blur due to hand-shake, camera-shake, and the like.
 
@pgriz- I redid like you said making sure to shoot at the center and making sure my camera was perpendicular to my subject. I can't figure out how to get 100% crops of the center and right corner but just looking at them I don't see very much difference. (except at f1.8, 2.0, 2.2) But for the rest of them, they are about the same so it is still hard for me to figure out my sweet spot. But I would say maybe f5.6?? Here are a few of them:

1. f1.8
f18-1.jpg


2. f3.2
f32.jpg


3. f5.6
f56.jpg


4. f8.0
f80-1.jpg


5. f22.0
f220-1.jpg
 
To get 100% crops, you can use photoediting software (Elements, CS5, etc) to crop a 200 pixel x 200 pixel area around the center then copy and paste it into a new image and save that image without any resizing.
Same for the corner crop.

As for your images, f/5.6 and f/8 look pretty sharp, but at f/22 you are starting to get softness due to diffraction. I'd say that you're good from f/5.6 to f/8, and maybe even f/11.

Now, the next test you need to do is to see if your autofocus will be as accurate as manual focusing. Set your aperture to f/1.8 (to get the thinnest Depth-of-field), select your center AF point, place it over the center dot and let it come to a focus. Snap the image. Now, without moving anything, set your lens to manual focus and refocus on the same point using live-view and 10x magnification. Snap the second image. Now compare the two. If they are identically sharp, then the AF focus is pretty dead-on. If the AF image is softer, then you know the AF is either front or back-focusing. But that's another test.
 
Gah my eyes need checked or something, the 215 in all of those look in focus to me :/
 
Try putting your target about twice as far away as you had it. That will "stress" the fine detail, and we should be able to see more clearly at which f/stops your lens is good.
 
I like number 1 in this setting. I use narrow apertures (larger f numbers) for my product photography when I have solid backgrounds. That way, more of the item is in focus and the viewer is not distracted by the background.
 
I agree that f/9 looks sharper everywhere.

But I would challenge you a bit and suggest that f/1.8 is the better photo regardless, since the background detracts from the picture.

Something in between would probably be ideal, probably f/4.

It used to be common wisdom that the sharpest aperture for most lenses was two stops down from wide open, which in your case, would be f/4 or a bit less. Give it a try, and see if that still yields enough background blur to get rid of the stock pot in the background. ;)

Side note: This is one advantage of REALLY fast lenses. An f/1.4 lens is SHARP at f/2.8, while an f/2.8 lens is not as sharp.
 

Most reactions

Back
Top